Journal articles on the topic 'Retina-net'
To see the other types of publications on this topic, follow the link: Retina-net.
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 50 journal articles for your research on the topic 'Retina-net.'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Browse journal articles on a wide variety of disciplines and organise your bibliography correctly.
1
Provencio, Ignacio, Mark D. Rollag, and Ana Maria Castrucci. "Photoreceptive net in the mammalian retina." Nature 415, no. 6871 (January 2002): 493. http://dx.doi.org/10.1038/415493a.
Full text
2
Liesegang, Thomas J. "Photoreceptive net in the mammalian retina." American Journal of Ophthalmology 133, no. 5 (May 2002): 739. http://dx.doi.org/10.1016/s0002-9394(02)01447-2.
Full text
3
Priya, S. Sathiya, and J. G. R. Sathiaseelan. "Enhanced Retina Blood Vessel Segmentation by Super Resolution Generative Adversarial Networks based U-Net." Indian Journal of Science and Technology 14, no. 43 (November 12, 2021): 3246–53. http://dx.doi.org/10.17485/ijst/v14i43.1502.
Full text
4
Zhao, Shun, Tao Liu, Bowen Liu, and Kun Ruan. "Attention residual convolution neural network based on U-net (AttentionResU-Net) for retina vessel segmentation." IOP Conference Series: Earth and Environmental Science 440 (March 19, 2020): 032138. http://dx.doi.org/10.1088/1755-1315/440/3/032138.
Full text
5
Tsuboi, S., R. Manabe, and S. Iizuka. "Aspects of electrolyte transport across isolated dog retinal pigment epithelium." American Journal of Physiology-Renal Physiology 250, no. 5 (May 1, 1986): F781—F784. http://dx.doi.org/10.1152/ajprenal.1986.250.5.f781.
Full textAbstract:
Transport of Na and Cl across the isolated dog retinal pigment epithelium (RPE) choroid was investigated. Under the short-circuit condition, a net Na flux was observed from choroid to retina and a net Cl flux was determined in the opposite direction. The current created by the net flux of these two ions was larger than the short-circuit current (SCC). Addition of 10(-5) M ouabain to the apical side inhibited net fluxes of both Na and Cl, whereas it reduced the SCC 84%. Addition of 10(-4) M furosemide to the apical side inhibited net Cl flux but had no effect on the net Na transport. The 10(-4) M furosemide reduced the SCC 38%. These drugs had no effect when applied to the basal side. Thus the transport of both Na and Cl depends on the Na-K-ATPase in the apical membrane of the dog RPE. A furosemide-sensitive neutral carrier at the apical membrane is suggested for the transport of Cl. Replacement of HCO3 with SO4 in the bathing solution caused an increase in the SCC, indicating the choroid-to-retina movement of HCO3 across the short-circuited dog RPE choroid.
6
Momin, Shadab, Yang Lei, Zhen Tian, Justin Roper, Jolinta Lin, Shannon Kahn, Hui-Kuo Shu, Jeffrey Bradley, Tian Liu, and Xiaofeng Yang. "Cascaded mutual enhancing networks for brain tumor subregion segmentation in multiparametric MRI." Physics in Medicine & Biology 67, no. 8 (April 11, 2022): 085015. http://dx.doi.org/10.1088/1361-6560/ac5ed8.
Full textAbstract:
Abstract Accurate segmentation of glioma and its subregions plays an important role in radiotherapy treatment planning. Due to a very populated multiparameter magnetic resonance imaging image, manual segmentation tasks can be very time-consuming, meticulous, and prone to subjective errors. Here, we propose a novel deep learning framework based on mutual enhancing networks to automatically segment brain tumor subregions. The proposed framework is suitable for the segmentation of brain tumor subregions owing to the contribution of Retina U-Net followed by the implementation of a mutual enhancing strategy between the classification localization map (CLM) module and segmentation module. Retina U-Net is trained to accurately identify view-of-interest and feature maps of the whole tumor (WT), which are then transferred to the CLM module and segmentation module. Subsequently, CLM generated by the CLM module is integrated with the segmentation module to bring forth a mutual enhancing strategy. In this way, our proposed framework first focuses on WT through Retina U-Net, and since WT consists of subregions, a mutual enhancing strategy then further aims to classify and segment subregions embedded within WT. We implemented and evaluated our proposed framework on the BraTS 2020 dataset consisting of 369 cases. We performed a 5-fold cross-validation on 200 datasets and a hold-out test on the remaining 169 cases. To demonstrate the effectiveness of our network design, we compared our method against the networks without Retina U-Net, mutual enhancing strategy, and a recently published Cascaded U-Net architecture. Results of all four methods were compared to the ground truth for segmentation and localization accuracies. Our method yielded significantly (P < 0.01) better values of dice-similarity-coefficient, center-of-mass-distance, and volume difference compared to all three competing methods across all tumor labels (necrosis and non-enhancing, edema, enhancing tumor, WT, tumor core) on both validation and hold-out dataset. Overall quantitative and statistical results of this work demonstrate the ability of our method to both accurately and automatically segment brain tumor subregions.
7
Cameron, David A. "Asymmetric retinal growth in the adult teleost green sunfish (Lepomis cyanellus)." Visual Neuroscience 12, no. 1 (January 1995): 95–102. http://dx.doi.org/10.1017/s0952523800007343.
Full textAbstract:
AbstractPrevious studies on fish retina have suggested that a curved, non-fused embryonic fissure is associated with, and perhaps caused by, asymmetric growth along the retina's marginal germinal zone (where neurons and Miiller glia are added appositionally throughout life). In this report retinal growth was measured directly in adult green sunfish (Lepomis cyanellus), which has a curved, non-fused embryonic fissure. Growth was asymmetric in both small and large fish: ventral and nasal retina grew more than temporal and dorsal retina. This asymmetry was due to different net rates of cellular addition, rather than differential passive expansion. The absolute rates of retinal growth in the centroperipheral direction were roughly exponential functions of fish size—smaller fish grow faster than large fish—but the area of new retina added per unit time did not vary with fish size. Visual implications of asymmetric retinal growth are evaluated.
8
Wang, Hua, Jingfei Hu, and Jicong Zhang. "SCRD-Net: A Deep Convolutional Neural Network Model for Glaucoma Detection in Retina Tomography." Complexity 2021 (April 9, 2021): 1–11. http://dx.doi.org/10.1155/2021/9858343.
Full textAbstract:
Early and accurate diagnosis of glaucoma is critical for avoiding human vision deterioration and preventing blindness. A deep-neural-network model has been developed for the diagnosis of glaucoma based on Heidelberg retina tomography (HRT), called “Seeking Common Features and Reserving Differences Net” (SCRD-Net) to make full use of the HRT data. In this work, the proposed SCRD-Net model achieved an area under the curve (AUC) of 94.0%. For the two HRT image modalities, the model sensitivities were 91.2% and 78.3% at specificities of 0.85 and 0.95, respectively. These results demonstrate a significant improvement over earlier results. In addition, we visualized the network outputs to develop an interpretation of the learned mechanism for discriminating glaucoma and normal images. Thus, the SCRD-Net can be an effective diagnostic indicator of glaucoma during clinical screening. To facilitate SCRD-Net utilization by the scientific community, the code implementation will be made publicly available.
9
Osio, A. A., H. Â. Lê, S. Ayugi, F. Onyango, P. Odwe, and S. Lefèvre. "DETECTION OF DEGRADED ACACIA TREE SPECIES USING DEEP NEURAL NETWORKS ON UAV DRONE IMAGERY." ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences V-3-2022 (May 17, 2022): 455–62. http://dx.doi.org/10.5194/isprs-annals-v-3-2022-455-2022.
Full textAbstract:
Abstract. Deep-learning-based image classification and object detection has been applied successfully to tree monitoring. However, studies of tree crowns and fallen trees, especially on flood inundated areas, remain largely unexplored. Detection of degraded tree trunks on natural environments such as water, mudflats, and natural vegetated areas is challenging due to the mixed colour image backgrounds. In this paper, Unmanned Aerial Vehicles (UAVs), or drones, with embedded RGB cameras were used to capture the fallen Acacia Xanthophloea trees from six designated plots around Lake Nakuru, Kenya. Motivated by the need to detect fallen trees around the lake, two well-established deep neural networks, i.e. Faster Region-based Convolution Neural Network (Faster R-CNN) and Retina-Net were used for fallen tree detection. A total of 7,590 annotations of three classes on 256×256 image patches were used for this study. Experimental results show the relevance of deep learning in this context, with Retina-Net model achieving 38.9% precision and 57.9% recall.
10
Roecklein, K. A., D. L. Wescott, S. F. Smagula, A. M. Soehner, P. L. Franzen, and B. P. Hasler. "0037 Melanopsin Driven Pupil Responses and Physical Activity: Stability of Activity from Day-to-Day in Winter in Seasonal Affective Disorder." Sleep 43, Supplement_1 (April 2020): A15. http://dx.doi.org/10.1093/sleep/zsaa056.036.
Full textAbstract:
Abstract Introduction The post-illumination pupil response (PIPR) is a measure of the responsivity of intrinsically photosensitive retinal ganglion cells (ipRGCs), and reflects the cell biology of the photoentrainment pathway projecting from the retina to the circadian clock. Adequate signaling from the ipRGCs in the retina to the circadian clock is necessary to result in robust circadian output which we hypothesize would increase inter-daily stability (IS), a non-parametric modeling technique that examines stability of rest activity rhythms across successive days. Methods Participants were aged 18–66 years and recruited from the greater Pittsburgh area during the Winter with Seasonal Affective Disorder who completed both actigraphy and pupillometry (n = 16). PIPR measures were collected after a 1 second red or blue light pulse, and are calculated as the Net difference between red and blue at multiple time frames: at 6 seconds post stimulus (PIPR 6), from 10–30 seconds post-stimulus (PIPR 20), or from 10–40 seconds post-stimulus (PIPR 30). Using actigraphy, inter-daily stability (IS) was calculated as the amount of overall variability in the recording that is accounted for by the typical 24-hour profile, and reflects stability of the mean 24-h profile day-to-day. Results Inter-daily stability (IS) was associated with Net PIPR 20 (Β = 0.561; p = .031) and Net PIPR 30 (Β = 0.551; p = .034; all Β’s are standardized), but not Net PIPR 6 (Β = 0.298; p = .304). Retinal irradiance was calculated for each participant based on age and pupil diameter, to account for age-related differences in transmission of the stimulus to the retina. All raw Net PIPR values were adjusted for calculated retinal irradiance, and gender and time since wake were included as covariates. Conclusion Inter-daily stability (IS) values indicate greater stability of 24-hour activity profiles across days. If reduced responsivity to entraining pulses of light is associated with day-to-day instability in activity rhythms, as shown here, we might expect that amplifying entraining light through environmental changes or bright light therapy would normalize inter-daily stability in SAD, or the reverse, stabilizing activity profiles across days could improve depression and/or normalize retinal ipRGC responsivity. Support NIMH K.A.R. MH103303
11
Nawwar*, Nadia M., Kasban ., and Salama May. "Deep Learning Model Based on Mobile-Net with Haar-like Algorithm for Masked Face Recognition at Nuclear Facilities." Regular issue 10, no. 7 (May 30, 2021): 18–23. http://dx.doi.org/10.35940/ijitee.g8893.0510721.
Full textAbstract:
During the spread of the COVID-I9 pandemic in early 2020, the WHO organization advised all people in the world to wear face-mask to limit the spread of COVID-19. Many facilities required that their employees wear face-mask. For the safety of the facility, it was mandatory to recognize the identity of the individual wearing the mask. Hence, face recognition of the masked individuals was required. In this research, a novel technique is proposed based on a mobile-net and Haar-like algorithm for detecting and recognizing the masked face. Firstly, recognize the authorized person that enters the nuclear facility in case of wearing the masked-face using mobile-net. Secondly, applying Haar-like features to detect the retina of the person to extract the boundary box around the retina compares this with the dataset of the person without the mask for recognition. The results of the proposed modal, which was tested on a dataset from Kaggle, yielded 0.99 accuracies, a loss of 0.08, F1.score 0.98.
12
Arsalan, Muhammad, Adnan Haider, Ja Hyung Koo, and Kang Ryoung Park. "Segmenting Retinal Vessels Using a Shallow Segmentation Network to Aid Ophthalmic Analysis." Mathematics 10, no. 9 (May 3, 2022): 1536. http://dx.doi.org/10.3390/math10091536.
Full textAbstract:
Retinal blood vessels possess a complex structure in the retina and are considered an important biomarker for several retinal diseases. Ophthalmic diseases result in specific changes in the retinal vasculature; for example, diabetic retinopathy causes the retinal vessels to swell, and depending upon disease severity, fluid or blood can leak. Similarly, hypertensive retinopathy causes a change in the retinal vasculature due to the thinning of these vessels. Central retinal vein occlusion (CRVO) is a phenomenon in which the main vein causes drainage of the blood from the retina and this main vein can close completely or partially with symptoms of blurred vision and similar eye problems. Considering the importance of the retinal vasculature as an ophthalmic disease biomarker, ophthalmologists manually analyze retinal vascular changes. Manual analysis is a tedious task that requires constant observation to detect changes. The deep learning-based methods can ease the problem by learning from the annotations provided by an expert ophthalmologist. However, current deep learning-based methods are relatively inaccurate, computationally expensive, complex, and require image preprocessing for final detection. Moreover, existing methods are unable to provide a better true positive rate (sensitivity), which shows that the model can predict most of the vessel pixels. Therefore, this study presents the so-called vessel segmentation ultra-lite network (VSUL-Net) to accurately extract the retinal vasculature from the background. The proposed VSUL-Net comprises only 0.37 million trainable parameters and uses an original image as input without preprocessing. The VSUL-Net uses a retention block that specifically maintains the larger feature map size and low-level spatial information transfer. This retention block results in better sensitivity of the proposed VSUL-Net without using expensive preprocessing schemes. The proposed method was tested on three publicly available datasets: digital retinal images for vessel extraction (DRIVE), structured analysis of retina (STARE), and children’s heart health study in England database (CHASE-DB1) for retinal vasculature segmentation. The experimental results demonstrated that VSUL-Net provides robust segmentation of retinal vasculature with sensitivity (Sen), specificity (Spe), accuracy (Acc), and area under the curve (AUC) values of 83.80%, 98.21%, 96.95%, and 98.54%, respectively, for DRIVE, 81.73%, 98.35%, 97.17%, and 98.69%, respectively, for CHASE-DB1, and 86.64%, 98.13%, 97.27%, and 99.01%, respectively, for STARE datasets. The proposed method provides an accurate segmentation mask for deep ophthalmic analysis.
13
Guduru, Abhilash, Arushi Gupta, Mudit Tyagi, Subhadra Jalali, and Jay Chhablani. "Optical coherence tomography angiography characterisation of Best disease and associated choroidal neovascularisation." British Journal of Ophthalmology 102, no. 4 (August 1, 2017): 444–47. http://dx.doi.org/10.1136/bjophthalmol-2017-310586.
Full textAbstract:
AimsTo characterise the vasculature of the retina in patients with Best vitelliform dystrophy, including those with choroidal neovascularisation (CNV), using optical coherence tomography angiography (OCTA) and correlate with fluorescein angiography (FA).MethodsThis prospective observational study included 19 eyes of 10 patients with Best disease. Using OCTA, all layers of retina were qualitatively characterised for each eye. Patients with CNV also underwent FA, and areas of CNV were measured by OCTA and FA and correlated.ResultsRetinal characteristics revealed 14 (74%) eyes with abnormal foveal avascular zone (FAZ) in the superficial layer, 19 eyes (100%) had an abnormal FAZ in deep layers, 11 (58%) eyes had a hyper-reflective centre in the superficial layer, 18 (95%) had patchy vascularity loss in the deep layer, 17 (89%) eyes had hyporeflective centre in the choriocapillary (CC) layer and 12 (63%) of those eyes had hyper-reflective material within the hyporeflective centre. Also, notably 6 (86%) CNV eyes had a "halo" or a hypolucent area surrounded in the CNV complex in the outer retinal layer. CNV patterns resembled dense net, loose net, mixed and a new found pattern of a ring shape. CNV measurements revealed an average area of 1.66±1.18mm2 using OCTA and an average area of 0.88±0.76mm2 using FA (p=0.15).ConclusionOCTA reveals that eyes with Best disease have abnormal FAZ, patchy vascularity loss in the superficial and deep layers of the retina and capillary dropout with a hyporeflective centre in CC layer. Further, OCTA is superior to FA in measuring CNV.
14
Li, Wei, Cong Wu, YuQing Cheng, and Zhi Yang. "DM-Net:a Depth-separable convolution and Multi-Scale Network for retinal blood vessel segmentation." Journal of Physics: Conference Series 2213, no. 1 (March 1, 2022): 012040. http://dx.doi.org/10.1088/1742-6596/2213/1/012040.
Full textAbstract:
Abstract Retina segmentation plays an important role in the medical field,In recent years, some proposed networks have some problems, such as single receptive field, huge parameters, and difficulty in training, which affect the segmentation results. In this paper, a U-Net-based DM-Net with deep separable convolution and multi-scale is proposed, a residual multiscale module is designed to reduce the parameters and improve the feature extraction ability. In order to cope with the feature information fusion at different levels and the sudden decrease in the number of feature channels in the decoder, the channel attention mechanism is applied. Experiments on the public data set CHASE_DB 1 show that DM-Net has achieved good results compared with other networks, especially in ACC (0.9748) and SP (0.9882). At the same time, it has few parameters and fast convergence speed.
15
PADNICK-SILVER, LISSA, and ROBERT A. LINSENMEIER. "Quantification of in vivo anaerobic metabolism in the normal cat retina through intraretinal pH measurements." Visual Neuroscience 19, no. 6 (November 2002): 793–806. http://dx.doi.org/10.1017/s095252380219609x.
Full textAbstract:
We examined intraretinal [H+] in the intact retina of anesthetized cats using H+-sensitive microelectrodes to obtain spatial profiles of extracellular [H+]. One H+ is produced when an anaerobically generated ATP is utilized. We theorized that H+ production directly reflects anaerobic glucose consumption. From the choroid (pH ∼7.40), [H+]o steadily increased to a maximum concentration in the proximal portion of the outer nuclear layer (pH∼7.20). The shape of the profile was always concave down, indicating that a net production of H+ occurred across the avascular outer retina. A three-layer diffusion model of the outer retina was developed and fitted to the data to quantify photoreceptor H+ extrusion into the extracellular space (QOR-H+). It was determined that the outer segment (OS) layer had negligible H+ extrusion. The data were then refitted to a special three-layer model in which the OS layer QH+ was set equal to zero, but in which the inner segments and outer nuclear layer produced H+. The resulting QOR-H+ was several orders of magnitude lower than previous measurements of QOR-lactate, which were based on choroidal mass balances of lactate.
16
André, Pascal, Bruno Saubaméa, Véronique Cochois-Guégan, Cynthia Marie-Claire, Julie Cattelotte, Maria Smirnova, Alfred H. Schinkel, Jean-Michel Scherrmann, and Salvatore Cisternino. "Transport of Biogenic Amine Neurotransmitters at the Mouse Blood–Retina and Blood–Brain Barriers by Uptake1 and Uptake2." Journal of Cerebral Blood Flow & Metabolism 32, no. 11 (August 1, 2012): 1989–2001. http://dx.doi.org/10.1038/jcbfm.2012.109.
Full textAbstract:
Uptake1 and uptake2 transporters are involved in the extracellular clearance of biogenic amine neurotransmitters at synaptic clefts. We looked for them at the blood–brain barrier (BBB) and blood–retina barriers (BRB), where they could be involved in regulating the neurotransmitter concentration and modulate/terminate receptor-mediated effects within the neurovascular unit (NVU). Uptake2 (Oct1-3/Slc22a1-3, Pmat/Slc29a4) and Mate1/Slc47a1 transporters are also involved in the transport of xenobiotics. We used in situ carotid perfusion of prototypic substrates like [3H]-1-methyl-4-phenylpyridinium ([3H]-MPP+), [3H]-histamine, [3H]-serotonin, and [3H]-dopamine, changes in ionic composition and genetic deletion of Oct1-3 carriers to detect uptake1 and uptake2 at the BBB and BRB. We showed that uptake1 and uptake2 are involved in the transport of [3H]-dopamine and [3H]-MPP+ at the blood luminal BRB, but not at the BBB. These functional studies, together with quantitative RT-PCR and confocal imaging, suggest that the mouse BBB lacks uptake1 (Net/Slc6a2, Dat/Slc6a3, Sert/Slc6a4), uptake2, and Mate1 on both the luminal and abluminal sides. However, we found evidence for functional Net and Oct1 transporters at the luminal BRB. These heterogeneous transport properties of the brain and retina NVUs suggest that the BBB helps protect the brain against biogenic amine neurotransmitters in the plasma while the BRB has more of a metabolic/endocrine role.
17
Binh, Nguyen Thanh, Nguyen Mong Hien, and Dang Thanh Tin. "Improving U-Net architecture and graph cuts optimization to classify arterioles and venules in retina fundus images." Journal of Intelligent & Fuzzy Systems 42, no. 4 (March 4, 2022): 4015–26. http://dx.doi.org/10.3233/jifs-212259.
Full textAbstract:
The central retinal artery and its branches supply blood to the inner retina. Vascular manifestations in the retina indirectly reflect the vascular changes and damage in organs such as the heart, kidneys, and brain because of the similar vascular structure of these organs. The diabetic retinopathy and risk of stroke are caused by increased venular caliber. The degrees of these diseases depend on the changes of arterioles and venules. The ratio between the calibers of arterioles and venules (AVR) is various. AVR is considered as the useful diagnostic indicator of different associated health problems. However, the task is not easy because of the lack of information of the features being used to classify the retinal vessels as arterioles and venules. This paper proposed a method to classify the retinal vessels into the arterioles and venules based on improving U-Net architecture and graph cuts. The accuracy of the proposed method is about 97.6%. The results of the proposed method are better than the other methods in RITE dataset and AVRDB dataset.
18
Gargari, Manizheh Safarkhani, Mir Hojjat Seyedi, and Mehdi Alilou. "Segmentation of Retinal Blood Vessels Using U-Net++ Architecture and Disease Prediction." Electronics 11, no. 21 (October 29, 2022): 3516. http://dx.doi.org/10.3390/electronics11213516.
Full textAbstract:
This study presents a segmentation method for the blood vessels and provides a method for disease diagnosis in individuals based on retinal images. Blood vessel segmentation in images of the retina is very challenging in medical analysis and diagnosis. It is an essential tool for a wide range of medical diagnoses. After segmentation and binary image improvement operations, the resulting binary images are processed and the features in the blood vessels are used as feature vectors to categorize retinal images and diagnose the type of disease available. To carry out the segmentation task and disease diagnosis, we used a deep learning approach involving a convolutional neural network (CNN) and U-Net++ architecture. A multi-stage method is used in this study to better diagnose the disease using retinal images. Our proposed method includes improving the color image of the retina, applying the Gabor filter to produce images derived from the green channel, segmenting the green channel by receiving images produced from the Gabor filter using U-Net++, extracting HOG and LBP features from binary images, and finally disease diagnosis using a one-dimensional convolutional neural network. The DRIVE and MESSIDOR image banks have been used to segment the image, determine the areas related to blood vessels in the retinal image, and evaluate the proposed method for retinal disease diagnosis. The achieved results for accuracy, sensitivity, specificity, and F1-score are 98.9, 94.1, 98.8, 85.26, and, 98.14, respectively, in the DRIVE dataset and the obtained results for accuracy, sensitivity, and specificity are 98.6, 99, 98, respectively, in MESSIDOR dataset. Hence, the presented system outperforms the manual approach applied by skilled ophthalmologists.
19
DMITRIEV, A. V., V. I. GOVARDOVSKII, H. N. SCHWAHN, and R. H. STEINBERG. "Light-induced changes of extracellular ions and volume in the isolated chick retina–pigment epithelium preparation." Visual Neuroscience 16, no. 6 (November 1999): 1157–67. http://dx.doi.org/10.1017/s095252389916615x.
Full textAbstract:
To better understand the mechanisms of extracellular space volume regulation and their possible effects on retinal function, light-induced changes in the concentrations of the principal extracellular ions (Na+, K+, Ca2+, and Cl−) were measured with ion-sensitive microelectrodes in the chick retina–pigment epithelium–choroid preparation. Changes of extracellular space volume were assessed by measuring the concentration of an impermeant marker, tetramethylammonium. In the inner retina, transient ON/OFF Na+ decrease was about twice as large as K+ increase, and the charge difference was compensated by a decrease in Cl− concentration. The ion changes were accompanied by extracellular space-volume decreases here. In the subretinal space, [Na+]o increase was about twice as large as K+ decrease, yet [Cl−]o also decreased; this was accompanied by a sustained extracellular space-volume increase. The ionic changes in the inner retina are consistent with a model of extracellular space-volume regulation which assumes that neuronal depolarization causes net uptake of NaCl, cell swelling, and extracellular space shrinkage. However, to prevent the apparent violation of electroneutrality in the subretinal space, our simple model should be expanded to include the involvement of unidentified anion(s). Substantial changes in the subretinal space volume may influence interaction between the neural retina and pigment epithelium. Among ionic changes, only the light-induced [K+]o decrease around the photoreceptors and the [Ca2+]o increase near the photoreceptor bodies and synaptic terminals are large enough (−25% and 7.5%, respectively) to be likely candidates for integrated intercellular signaling.
20
Shi, Fei, Xuena Cheng, Shuanglang Feng, Changqing Yang, Shengyong Diao, Weifang Zhu, Dehui Xiang, et al. "Group-wise context selection network for choroid segmentation in optical coherence tomography." Physics in Medicine & Biology 66, no. 24 (December 9, 2021): 245010. http://dx.doi.org/10.1088/1361-6560/ac3a23.
Full textAbstract:
Abstract Choroid thickness measured from optical coherence tomography (OCT) images has emerged as a vital metric in the management of retinal diseases such as high myopia. In this paper, we propose a novel group-wise context selection network (referred to as GCS-Net) to segment the choroid of either normal or high myopia eyes. To deal with the diverse choroid thickness and the variable shape of the pathological retina, GCS-Net adopts the group-wise channel dilation (GCD) module and the group-wise spatial dilation module, which can automatically select group-wise multi-scale information under the guidance of channel attention or spatial attention, and enhance the consistency between the receptive field and the target area. Furthermore, a boundary optimization network with a new edge loss is incorporated to improve the resulting choroid boundary by deep supervision. Experimental results evaluated on a dataset composed of 1650 clinically obtained OCT B-scans show that the proposed GCS-Net can achieve a Dice similarity coefficient of 95.97 ± 0.54%, which outperforms some state-of-the-art segmentation networks.
21
Guimarães, Marília Zaluar P., and Jan Nora Hokoç. "Tyrosine hydroxylase expression in the Cebus monkey retina." Visual Neuroscience 14, no. 4 (July 1997): 705–15. http://dx.doi.org/10.1017/s0952523800012669.
Full textAbstract:
AbstractTyrosine hydroxylase (TH) expression was used as a marker to study the dopaminergic cells in the Cebus monkey retina. Two types of dopaminergic cells were identified by cell body size and location, level of arborization in the inner plexiform layer, and amount of immunolabeling. Type 1 cells displayed intense immunoreactivity and larger somata (12–24 μm) located in the inner nuclear layer or ganglion cell layer, whereas type 2 had smaller cell bodies (8–14 μm) found either in the inner plexiform layer or ganglion cell layer and were more faintly labeled. Interplexiform cells were characterized as type 1 dopaminergic cells. Immunoreactive axon-like processes were seen in the nerve fiber layer, and a net of fibers was visible in the foveal pit and in the extreme periphery of the retina. The population of TH+ cells was most numerous in the temporal superior quadrant and its density peaked at 1–2 mm from the fovea. Type 1 TH+ cells were more numerous than type 2 cells at any eccentricity. Along the horizontal meridian, type 1 cell density was slightly higher in temporal (29 cells/mm2) than in nasal (25 cells/mm2) retina, while type 2 cells had a homogeneous distribution (4.5 cells/mm2). Along the vertical meridian, type 1 cells reached lower peak density (average 17.7 cells/mm2) in the inferior retina (central 4 mm), compared to the superior portion (23.7 cells/mm2). Type 2 cell density varied from 4.5 cells/mm2 in the superior region to 9.4 cells/mm2 in the inferior region. The spatial density of the two cell types varied approximately inversely while the total density of TH+ cells was virtually constant across the retina. No correlation between dopaminergic cells and rod distribution was found. However, we suggest that dopaminergic cells could have a role in mesopic and/or photopic vision in this species, since TH+ fibers are present in cone-dominated regions like the foveola and extreme nasal Periphery.
22
Sollini, Martina, Margarita Kirienko, Noemi Gozzi, Alessandro Bruno, Chiara Torrisi, Luca Balzarini, Emanuele Voulaz, Marco Alloisio, and Arturo Chiti. "The Development of an Intelligent Agent to Detect and Non-Invasively Characterize Lung Lesions on CT Scans: Ready for the “Real World”?" Cancers 15, no. 2 (January 5, 2023): 357. http://dx.doi.org/10.3390/cancers15020357.
Full textAbstract:
(1) Background: Once lung lesions are identified on CT scans, they must be characterized by assessing the risk of malignancy. Despite the promising performance of computer-aided systems, some limitations related to the study design and technical issues undermine these tools’ efficiency; an “intelligent agent” to detect and non-invasively characterize lung lesions on CT scans is proposed. (2) Methods: Two main modules tackled the detection of lung nodules on CT scans and the diagnosis of each nodule into benign and malignant categories. Computer-aided detection (CADe) and computer aided-diagnosis (CADx) modules relied on deep learning techniques such as Retina U-Net and the convolutional neural network; (3) Results: Tests were conducted on one publicly available dataset and two local datasets featuring CT scans acquired with different devices to reveal deep learning performances in “real-world” clinical scenarios. The CADe module reached an accuracy rate of 78%, while the CADx’s accuracy, specificity, and sensitivity stand at 80%, 73%, and 85.7%, respectively; (4) Conclusions: Two different deep learning techniques have been adapted for CADe and CADx purposes in both publicly available and private CT scan datasets. Experiments have shown adequate performance in both detection and diagnosis tasks. Nevertheless, some drawbacks still characterize the supervised learning paradigm employed in networks such as CNN and Retina U-Net in real-world clinical scenarios, with CT scans from different devices with different sensors’ fingerprints and spatial resolution. Continuous reassessment of CADe and CADx’s performance is needed during their implementation in clinical practice.
23
Slaughter, M. M., and S. H. Bai. "Differential effects of baclofen on sustained and transient cells in the mudpuppy retina." Journal of Neurophysiology 61, no. 2 (February 1, 1989): 374–81. http://dx.doi.org/10.1152/jn.1989.61.2.374.
Full textAbstract:
1. Baclofen, a gamma-aminobutyric acid (GABA)/B receptor agonist, was bath applied while recording the responses of second- and third-order neurons in the mudpuppy retina. Baclofen receptors were largely restricted to amacrine and ganglion cells. 2. Baclofen hyperpolarized the membrane potential of many, but not all, third-order neurons. This involved an increase in input conductance, probably associated with an opening of potassium channels. 3. The maximal increase in input conductance associated with the activation of GABA/B receptors was approximately one-third of that produced by activation of GABA/A receptors. 4. Baclofen suppressed sustained responses but enhanced transient responses. The net effect was that responses throughout the inner retina became more transient in the presence of baclofen. 5. In sustained cells baclofen not only suppressed the sustained responses but also revealed large transient responses. Thus baclofen converted the light responses of these cells from sustained to transient. This suggests that sustained cells receive significant transient excitation which is normally masked by the sustained inputs. 6. The role of the GABA/B receptor in controlling response characteristics and information content of amacrine and ganglion cells is discussed.
24
Chen, Quan, John W. Olney, Peter D. Lukasiewicz, Todd Almli, and Carmelo Romano. "Ca2+-Independent Excitotoxic Neurodegeneration in Isolated Retina, an Intact Neural Net: A Role for Cl− and Inhibitory Transmitters." Molecular Pharmacology 53, no. 3 (March 1, 1998): 564–72. http://dx.doi.org/10.1124/mol.53.3.564.
Full text
25
Galli-Resta, Lucia, Elena Novelli, and Alessandro Viegi. "Dynamic microtubule-dependent interactions position homotypic neurones in regular monolayered arrays during retinal development." Development 129, no. 16 (August 15, 2002): 3803–14. http://dx.doi.org/10.1242/dev.129.16.3803.
Full textAbstract:
In the vertebrate retina cell layers support serial processing, while monolayered arrays of homotypic neurones tile each layer to allow parallel processing. How neurones form layers and arrays is still largely unknown. We show that monolayered retinal arrays are dynamic structures based on dendritic interactions between the array cells. The analysis of three developing retinal arrays shows that these become regular as a net of dendritic processes links neighbouring array cells. Molecular or pharmacological perturbations of microtubules within dendrites lead to a stereotyped and reversible disruption of array organization: array cells lose their regular spacing and the arrangement in a monolayer. This leads to a micro-mechanical explanation of how monolayers of regularly spaced ‘like-cells’ are formed.
26
Edelman, J. L., H. Lin, and S. S. Miller. "Potassium-induced chloride secretion across the frog retinal pigment epithelium." American Journal of Physiology-Cell Physiology 266, no. 4 (April 1, 1994): C957—C966. http://dx.doi.org/10.1152/ajpcell.1994.266.4.c957.
Full textAbstract:
In the intact eye, a transition from light to dark increases K concentration ([K]o) from approximately 2 to 5 mM in the extracellular (subretinal) space between the photoreceptors and the retinal pigment epithelium (RPE) apical membrane. In control (HCO3/CO2) Ringer solution, 36Cl was actively absorbed across isolated bullfrog RPE (retina to choroid) at a rate of 0.31 +/- 0.02 (SE) mu eq.cm-2.h-1 (n = 15). Elevating apical [K]o from 2 to 5 mM reversed active 36Cl transport to secretion (choroid to retina), with a rate of 0.76 +/- 0.17 mu eq.cm-2.h-1. This reversal was completely inhibited by 1 mM 4,4'-diisothiocyanostilbene-2,2'-disulfonic acid (DIDS) in either the apical or basal bath. In open circuit, elevating [K]o induced a similar reversal of net 36Cl flux and inhibited fluid absorption by approximately 25%. Apical Ba2+ (1 mM), decreased CO2 (5 to 1%), or increased apical bath HCO3 concentration ([HCO3]o) also caused a DIDS-inhibitable reversal of active 36Cl flux. A 10-fold reduction of apical bath Na or [HCO3]o significantly inhibited [K]o, Ba2+, and low CO2-induced Cl secretion. All of these results can be understood in terms of an intracellular pH-dependent stimulation of the basolateral membrane Cl-HCO3 exchanger.
27
Meng, Yongan, Hailei Lan, Yuqian Hu, Zailiang Chen, Pingbo Ouyang, and Jing Luo. "Application of Improved U-Net Convolutional Neural Network for Automatic Quantification of the Foveal Avascular Zone in Diabetic Macular Ischemia." Journal of Diabetes Research 2022 (February 26, 2022): 1–8. http://dx.doi.org/10.1155/2022/4612554.
Full textAbstract:
Objectives.The foveal avascular zone (FAZ) is a biomarker for quantifying diabetic macular ischemia (DMI), to automate the identification and quantification of the FAZ in DMI, using an improved U-Net convolutional neural network (CNN) and to establish a CNN model based on optical coherence tomography angiography (OCTA) images for the same purpose. Methods. The FAZ boundaries on the full-thickness retina of 6 × 6 mm en face OCTA images of DMI and normal eyes were manually marked. Seventy percent of OCTA images were used as the training set, and ten percent of these images were used as the validation set to train the improved U-Net CNN with two attention modules. Finally, twenty percent of the OCTA images were used as the test set to evaluate the accuracy of this model relative to that of the baseline U-Net model. This model was then applied to the public data set sFAZ to compare its effectiveness with existing models at identifying and quantifying the FAZ area. Results. This study included 110 OCTA images. The Dice score of the FAZ area predicted by the proposed method was 0.949, the Jaccard index was 0.912, and the area correlation coefficient was 0.996. The corresponding values for the baseline U-Net were 0.940, 0.898, and 0.995, respectively, and those based on the description data set sFAZ were 0.983, 0.968, and 0.950, respectively, which were better than those previously reported based on this data set. Conclusions. The improved U-Net CNN was more accurate at automatically measuring the FAZ area on the OCTA images than the traditional CNN. The present model may measure the DMI index more accurately, thereby assisting in the diagnosis and prognosis of retinal vascular diseases such as diabetic retinopathy.
28
Edelman, J. L., H. Lin, and S. S. Miller. "Acidification stimulates chloride and fluid absorption across frog retinal pigment epithelium." American Journal of Physiology-Cell Physiology 266, no. 4 (April 1, 1994): C946—C956. http://dx.doi.org/10.1152/ajpcell.1994.266.4.c946.
Full textAbstract:
Radioactive tracers and a modified capacitance-probe technique were used to characterize the mechanisms that mediate Cl and fluid absorption across the bullfrog retinal pigment epithelium (RPE)-choroid. In control (HCO3/CO2) Ringer solution, 36Cl was actively absorbed (retina to choroid) at a mean rate of 0.34 mu eq.cm-2.h-1 (n = 34) and accounted for approximately 25% of the short-circuit current. Apical bumetanide (100 microM) or basal 4,4'-diisothiocyanostilbene-2,2'-disulfonic acid (DIDS; 1 mM) inhibited active Cl transport by 70 and 62%, respectively. Active Cl absorption was doubled, either by removing HCO3 from the bathing media or by elevating CO2 from 5 to 13%, and the increased flux was inhibited by apical bumetanide or basal DIDS. Open-circuit measurements of fluid absorption rate (Jv) and the net fluxes of 36Cl, 22Na, and 86Rb (K substitute) indicated that CO2-induced acidification stimulated NaCl and fluid absorption across the RPE. During acidification, bumetanide produced a twofold larger inhibition of Jv compared with control. Stimulation of net Cl absorption was most likely caused by inhibition of the the basolateral membrane intracellular pH-dependent Cl-HCO3 exchanger.
29
Prakash M, Madhura, Deepthi K Prasad, Meghna S Kulkarni, Spoorthi K, and Venkatakrishnan S. "A Systematic Study of Deep Learning Architectures for Analysis of Glaucoma and Hypertensive Retinopathy." International Journal of Artificial Intelligence & Applications 13, no. 6 (November 30, 2022): 33–49. http://dx.doi.org/10.5121/ijaia.2022.13603.
Full textAbstract:
Deep learning models are applied seamlessly across various computer vision tasks like object detection, object tracking, scene understanding and further. The application of cutting-edge deep learning (DL) models like U-Net in the classification and segmentation of medical images on different modalities has established significant results in the past few years. Ocular diseases like Diabetic Retinopathy (DR), Glaucoma, Age-Related Macular Degeneration (AMD / ARMD), Hypertensive Retina (HR), Cataract, and dry eyes can be detected at the early stages of disease onset by capturing the fundus image or the anterior image of the subject’s eye. Early detection is key to seeking early treatment and thereby preventing the disease progression, which in some cases may lead to blindness. There is a plethora of deep learning models available which have established significant results in medical image processing and specifically in ocular disease detection. A given task can be solved by using a variety of models and or a combination of them. Deep learning models can be computationally expensive and deploying them on an edge device may be a challenge. This paper provides a comprehensive report and critical evaluation of the various deep learning architectures that can be used to segment and classify ocular diseases namely Glaucoma and Hypertensive Retina on the posterior images of the eye. This review also compares the models based on complexity and edge deployability.
30
Arsalan, Owais, Mahmood, Cho, and Park. "Aiding the Diagnosis of Diabetic and Hypertensive Retinopathy Using Artificial Intelligence-Based Semantic Segmentation." Journal of Clinical Medicine 8, no. 9 (September 11, 2019): 1446. http://dx.doi.org/10.3390/jcm8091446.
Full textAbstract:
Automatic segmentation of retinal images is an important task in computer-assisted medical image analysis for the diagnosis of diseases such as hypertension, diabetic and hypertensive retinopathy, and arteriosclerosis. Among the diseases, diabetic retinopathy, which is the leading cause of vision detachment, can be diagnosed early through the detection of retinal vessels. The manual detection of these retinal vessels is a time-consuming process that can be automated with the help of artificial intelligence with deep learning. The detection of vessels is difficult due to intensity variation and noise from non-ideal imaging. Although there are deep learning approaches for vessel segmentation, these methods require many trainable parameters, which increase the network complexity. To address these issues, this paper presents a dual-residual-stream-based vessel segmentation network (Vess-Net), which is not as deep as conventional semantic segmentation networks, but provides good segmentation with few trainable parameters and layers. The method takes advantage of artificial intelligence for semantic segmentation to aid the diagnosis of retinopathy. To evaluate the proposed Vess-Net method, experiments were conducted with three publicly available datasets for vessel segmentation: digital retinal images for vessel extraction (DRIVE), the Child Heart Health Study in England (CHASE-DB1), and structured analysis of retina (STARE). Experimental results show that Vess-Net achieved superior performance for all datasets with sensitivity (Se), specificity (Sp), area under the curve (AUC), and accuracy (Acc) of 80.22%, 98.1%, 98.2%, and 96.55% for DRVIE; 82.06%, 98.41%, 98.0%, and 97.26% for CHASE-DB1; and 85.26%, 97.91%, 98.83%, and 96.97% for STARE dataset.
31
Viedma, Ignacio A., David Alonso-Caneiro, Scott A. Read, and Michael J. Collins. "OCT Retinal and Choroidal Layer Instance Segmentation Using Mask R-CNN." Sensors 22, no. 5 (March 4, 2022): 2016. http://dx.doi.org/10.3390/s22052016.
Full textAbstract:
Optical coherence tomography (OCT) of the posterior segment of the eye provides high-resolution cross-sectional images that allow visualization of individual layers of the posterior eye tissue (the retina and choroid), facilitating the diagnosis and monitoring of ocular diseases and abnormalities. The manual analysis of retinal OCT images is a time-consuming task; therefore, the development of automatic image analysis methods is important for both research and clinical applications. In recent years, deep learning methods have emerged as an alternative method to perform this segmentation task. A large number of the proposed segmentation methods in the literature focus on the use of encoder–decoder architectures, such as U-Net, while other architectural modalities have not received as much attention. In this study, the application of an instance segmentation method based on region proposal architecture, called the Mask R-CNN, is explored in depth in the context of retinal OCT image segmentation. The importance of adequate hyper-parameter selection is examined, and the performance is compared with commonly used techniques. The Mask R-CNN provides a suitable method for the segmentation of OCT images with low segmentation boundary errors and high Dice coefficients, with segmentation performance comparable with the commonly used U-Net method. The Mask R-CNN has the advantage of a simpler extraction of the boundary positions, especially avoiding the need for a time-consuming graph search method to extract boundaries, which reduces the inference time by 2.5 times compared to U-Net, while segmenting seven retinal layers.
32
Mohammedhasan, Mali, and Harun Uğuz. "A New Early Stage Diabetic Retinopathy Diagnosis Model Using Deep Convolutional Neural Networks and Principal Component Analysis." Traitement du Signal 37, no. 5 (November 25, 2020): 711–22. http://dx.doi.org/10.18280/ts.370503.
Full textAbstract:
Diabetic retinopathy (DR) is a disease of the retina, which leads over time to vision problems such retinal detachment, vitreous hemorrhage, glaucoma, and in worse cases leads to blindness, which can initially be controlled by periodic DR-screening. Early diagnosis will lead to greater control of the disease, whereas performing retinal examinations on all diabetic patients is an unattainable need, as diabetes is a chronic disease and its global prevalence has been steadily increasing over the past few decades. According to recent World Health Organization statistics, about 422 million people worldwide have diabetes, the majority living in low-and middle-income countries. This paper proposes a new strategy that brings the strength of convolutional neural networks (CNNs) to the diagnosis of DR. Coupled with using principal component analysis (PCA) that performs dimension reduction to improve the diagnostic accuracy, the proposed model exploiting edge-preserving guided image filtering (E-GIF) that performs as a contrast enhancement mechanism, and in addition to smoothing low gradient areas, it also accentuates strong edges. Diabetic retinopathy causes progressive damage to the blood vessels in the retina to the extent that it leaves traces and lesions in the tissues of the retina. These lesions appear in the form of edges and when processing retinal images, we seek to accentuate these edges to enable better diagnosis of diabetic retinopathy symptoms. A new CNN architecture with residual connections is used, which performs very well in diagnosing DR. The proposed model is named with RUnet-PCA: Residual U-net Deep CNN with Principal Component Analysis. The well-known AlexNet, VggNet-s, VggNet-16, VggNet-19, GoogleNet, and ResNet models were adopted for comparison with the proposed model. Publicly available Kaggle dataset was employed for training exploring the DR diagnosis accuracy. Experimental results show that the proposed RUnet-PCA model achieved a diagnosis accuracy of 98.44% and it was extremely robust and promising in comparison to other diagnosis methods.
33
Zhao, Jie, and Qianjin Feng. "Deep Att-ResGAN: A Retinal Vessel Segmentation Network for Color Fundus Images." Traitement du Signal 38, no. 5 (October 31, 2021): 1309–17. http://dx.doi.org/10.18280/ts.380505.
Full textAbstract:
Retinal vessel segmentation plays a significant role in the diagnosis and treatment of ophthalmological diseases. Recent studies have proved that deep learning can effectively segment the retinal vessel structure. However, the existing methods have difficulty in segmenting thin vessels, especially when the original image contains lesions. Based on generative adversarial network (GAN), this paper proposes a deep network with residual module and attention module (Deep Att-ResGAN). The network consists of four identical subnetworks. The output of each subnetwork is imported to the next subnetwork as contextual features that guide the segmentation. Firstly, the problems of the original image, namely, low contrast, uneven illumination, and data insufficiency, were solved through image enhancement and preprocessing. Next, an improved U-Net was adopted to serve as the generator, which stacks the residual and attention modules. These modules optimize the weight of the generator, and enhance the generalizability of the network. Further, the segmentation was refined iteratively by the discriminator, which contributes to the performance of vessel segmentation. Finally, comparative experiments were carried out on two public datasets: Digital Retinal Images for Vessel Extraction (DRIVE) and Structured Analysis of the Retina (STARE). The experimental results show that Deep Att-ResGAN outperformed the equivalent models like U-Net and GAN in most metrics. Our network achieved accuracy of 0.9565 and F1 of 0.829 on DRIVE, and accuracy of 0.9690 and F1 of 0.841 on STARE.
34
Xu, Jing-Jing, Qi-Jie Wei, Kang Li, Zhen-Ping Li, Tian Yu, Jian-Chun Zhao, Da-Yong Ding, Xi-Rong Li, Guang-Zhi Wang, and Hong Dai. "Three-dimensional diabetic macular edema thickness maps based on fluid segmentation and fovea detection using deep learning." International Journal of Ophthalmology 15, no. 3 (March 18, 2022): 495–501. http://dx.doi.org/10.18240/ijo.2022.03.19.
Full textAbstract:
AIM: To explore a more accurate quantifying diagnosis method of diabetic macular edema (DME) by displaying detailed 3D morphometry beyond the gold-standard quantification indicator-central retinal thickness (CRT) and apply it in follow-up of DME patients. METHODS: Optical coherence tomography (OCT) scans of 229 eyes from 160 patients were collected. We manually annotated cystoid macular edema (CME), subretinal fluid (SRF) and fovea as ground truths. Deep convolution neural networks (DCNNs) were constructed including U-Net, sASPP, HRNetV2-W48, and HRNetV2-W48+Object-Contextual Representation (OCR) for fluid (CME+SRF) segmentation and fovea detection respectively, based on which the thickness maps of CME, SRF and retina were generated and divided by Early Treatment Diabetic Retinopathy Study (ETDRS) grid. RESULTS: In fluid segmentation, with the best DCNN constructed and loss function, the dice similarity coefficients (DSC) of segmentation reached 0.78 (CME), 0.82 (SRF), and 0.95 (retina). In fovea detection, the average deviation between the predicted fovea and the ground truth reached 145.7±117.8 μm. The generated macular edema thickness maps are able to discover center-involved DME by intuitive morphometry and fluid volume, which is ignored by the traditional definition of CRT>250 μm. Thickness maps could also help to discover fluid above or below the fovea center ignored or underestimated by a single OCT B-scan. CONCLUSION: Compared to the traditional unidimensional indicator-CRT, 3D macular edema thickness maps are able to display more intuitive morphometry and detailed statistics of DME, supporting more accurate diagnoses and follow-up of DME patients.
35
Arsalan, Muhammad, Adnan Haider, Jiho Choi, and Kang Ryoung Park. "Diabetic and Hypertensive Retinopathy Screening in Fundus Images Using Artificially Intelligent Shallow Architectures." Journal of Personalized Medicine 12, no. 1 (December 23, 2021): 7. http://dx.doi.org/10.3390/jpm12010007.
Full textAbstract:
Retinal blood vessels are considered valuable biomarkers for the detection of diabetic retinopathy, hypertensive retinopathy, and other retinal disorders. Ophthalmologists analyze retinal vasculature by manual segmentation, which is a tedious task. Numerous studies have focused on automatic retinal vasculature segmentation using different methods for ophthalmic disease analysis. However, most of these methods are computationally expensive and lack robustness. This paper proposes two new shallow deep learning architectures: dual-stream fusion network (DSF-Net) and dual-stream aggregation network (DSA-Net) to accurately detect retinal vasculature. The proposed method uses semantic segmentation in raw color fundus images for the screening of diabetic and hypertensive retinopathies. The proposed method’s performance is assessed using three publicly available fundus image datasets: Digital Retinal Images for Vessel Extraction (DRIVE), Structured Analysis of Retina (STARE), and Children Heart Health Study in England Database (CHASE-DB1). The experimental results revealed that the proposed method provided superior segmentation performance with accuracy (Acc), sensitivity (SE), specificity (SP), and area under the curve (AUC) of 96.93%, 82.68%, 98.30%, and 98.42% for DRIVE, 97.25%, 82.22%, 98.38%, and 98.15% for CHASE-DB1, and 97.00%, 86.07%, 98.00%, and 98.65% for STARE datasets, respectively. The experimental results also show that the proposed DSA-Net provides higher SE compared to the existing approaches. It means that the proposed method detected the minor vessels and provided the least false negatives, which is extremely important for diagnosis. The proposed method provides an automatic and accurate segmentation mask that can be used to highlight the vessel pixels. This detected vasculature can be utilized to compute the ratio between the vessel and the non-vessel pixels and distinguish between diabetic and hypertensive retinopathies, and morphology can be analyzed for related retinal disorders.
36
Ayoub, George S., and Gary Matthews. "Substance P modulates calcium current in retinal bipolar neurons." Visual Neuroscience 8, no. 6 (June 1992): 539–44. http://dx.doi.org/10.1017/s0952523800005630.
Full textAbstract:
AbstractRetinal bipolar cells are non-spiking interneurons that relay information from photoreceptors to amacrine and ganglion cells. In turn, bipolar cells receive extensive synaptic feedback from amacrine cells, some of which contain neuropeptides, including substance P. We have examined the effect of substance P on single bipolar neurons isolated from goldfish retina and find that substance P (0.1–1 nM) produced a voltage-dependent inhibition of calcium current in these cells. The inhibition was strongest at negative potentials, with the peak suppression occurring at –20 to –30 mV; at potentials positive to 0 mV, there was little effect on calcium current. Thus, the net effect was to shift the voltage range of activation of calcium current toward more positive potentials. The inhibition of calcium current by substance P required GTP in the patch pipette and was blocked by internal GDP-β-S. Similar effects on calcium current were observed with somatostatin and metenkephalin, which are also found in amacrine cells.
37
Bleckert, Adam, Chi Zhang, Maxwell H. Turner, David Koren, David M. Berson, Silvia J. H. Park, Jonathan B. Demb, Fred Rieke, Wei Wei, and Rachel O. Wong. "GABA release selectively regulates synapse development at distinct inputs on direction-selective retinal ganglion cells." Proceedings of the National Academy of Sciences 115, no. 51 (December 3, 2018): E12083—E12090. http://dx.doi.org/10.1073/pnas.1803490115.
Full textAbstract:
Synaptic inhibition controls a neuron’s output via functionally distinct inputs at two subcellular compartments, the cell body and the dendrites. It is unclear whether the assembly of these distinct inhibitory inputs can be regulated independently by neurotransmission. In the mammalian retina, γ-aminobutyric acid (GABA) release from starburst amacrine cells (SACs) onto the dendrites of on–off direction-selective ganglion cells (ooDSGCs) is essential for directionally selective responses. We found that ooDSGCs also receive GABAergic input on their somata from other amacrine cells (ACs), including ACs containing the vasoactive intestinal peptide (VIP). When net GABAergic transmission is reduced, somatic, but not dendritic, GABAA receptor clusters on the ooDSGC increased in number and size. Correlative fluorescence imaging and serial electron microscopy revealed that these enlarged somatic receptor clusters are localized to synapses. By contrast, selectively blocking vesicular GABA release from either SACs or VIP ACs did not alter dendritic or somatic receptor distributions on the ooDSGCs, showing that neither SAC nor VIP AC GABA release alone is required for the development of inhibitory synapses in ooDSGCs. Furthermore, a reduction in net GABAergic transmission, but not a selective reduction from SACs, increased excitatory drive onto ooDSGCs. This increased excitation may drive a homeostatic increase in ooDSGC somatic GABAA receptors. Differential regulation of GABAA receptors on the ooDSGC’s soma and dendrites could facilitate homeostatic control of the ooDSGC’s output while enabling the assembly of the GABAergic connectivity underlying direction selectivity to be indifferent to altered transmission.
38
Alryalat, Saif Aldeen, Mohammad Al-Antary, Yasmine Arafa, Babak Azad, Cornelia Boldyreff, Tasneem Ghnaimat, Nada Al-Antary, Safa Alfegi, Mutasem Elfalah, and Mohammed Abu-Ameerh. "Deep Learning Prediction of Response to Anti-VEGF among Diabetic Macular Edema Patients: Treatment Response Analyzer System (TRAS)." Diagnostics 12, no. 2 (January 26, 2022): 312. http://dx.doi.org/10.3390/diagnostics12020312.
Full textAbstract:
Diabetic macular edema (DME) is the most common cause of visual impairment among patients with diabetes mellitus. Anti-vascular endothelial growth factors (Anti-VEGFs) are considered the first line in its management. The aim of this research has been to develop a deep learning (DL) model for predicting response to intravitreal anti-VEGF injections among DME patients. The research included treatment naive DME patients who were treated with anti-VEGF. Patient’s pre-treatment and post-treatment clinical and macular optical coherence tomography (OCT) were assessed by retina specialists, who annotated pre-treatment images for five prognostic features. Patients were also classified based on their response to treatment in their post-treatment OCT into either good responder, defined as a reduction of thickness by >25% or 50 µm by 3 months, or poor responder. A novel modified U-net DL model for image segmentation, and another DL EfficientNet-B3 model for response classification were developed and implemented for predicting response to anti-VEGF injections among patients with DME. Finally, the classification DL model was compared with different levels of ophthalmology residents and specialists regarding response classification accuracy. The segmentation deep learning model resulted in segmentation accuracy of 95.9%, with a specificity of 98.9%, and a sensitivity of 87.9%. The classification accuracy of classifying patients’ images into good and poor responders reached 75%. Upon comparing the model’s performance with practicing ophthalmology residents, ophthalmologists and retina specialists, the model’s accuracy is comparable to ophthalmologist’s accuracy. The developed DL models can segment and predict response to anti-VEGF treatment among DME patients with comparable accuracy to general ophthalmologists. Further training on a larger dataset is nonetheless needed to yield more accurate response predictions.
39
MITRA, PRATIP, and ROBERT F. MILLER. "Normal and rebound impulse firing in retinal ganglion cells." Visual Neuroscience 24, no. 1 (January 2007): 79–90. http://dx.doi.org/10.1017/s0952523807070101.
Full textAbstract:
Given that the action potential output of retinal ganglion cells (RGCs) determines the nature of the visual information that is transmitted from the retina, an understanding of their intrinsic impulse firing characteristics is critical for an appreciation of the overall processing of visual information. Recordings from RGCs within an isolated whole-mount retina preparation showed that their normal impulse firing from the resting membrane potential (RMP) was linearly correlated in its frequency with the stimulus intensity. In addition to describing the relationship between the magnitude of the current injection and the resulting impulse frequency (F/I relationship), we have characterized the properties of individual action potentials when they are elicited from the RMP. In contrast, hyperpolarizing below the RMP revealed that RGCs displayed a time dependent anomalous rectification, manifested by the appearance of a depolarizing sag in their voltage response. When an adequate period of hyperpolarization was terminated, a fast phasic period of “rebound excitation” was observed, characterized by a brief phasic burst of impulse activity. When compared to equivalent action potential firing evoked by depolarizing from the RMP, rebound spiking was associated with a lower threshold and shorter latency for impulse activation as well as a prominent, phasic, burst-like doublet, or triplet of impulses. The rebound action potential had a more positive voltage overshoot and displayed a higher peak rate of rise in its upstroke than those correspondingly generated by depolarizing current pulses from the RMP. Blocking sodium spikes with TTX confirmed that the preceding hyperpolarization led to the recruitment and subsequent generation of a transient depolarizing voltage overshoot, which we have termed the net depolarizing overshoot (NDO). We propose that the NDO boosts the generation of sodium spikes by triggering rebound spikes on its upstroke and crest, thus accounting for the observed voltage dependent change in the firing pattern of RGCs.
40
Khoroshevsky, Faina, Stanislav Khoroshevsky, and Aharon Bar-Hillel. "Parts-per-Object Count in Agricultural Images: Solving Phenotyping Problems via a Single Deep Neural Network." Remote Sensing 13, no. 13 (June 26, 2021): 2496. http://dx.doi.org/10.3390/rs13132496.
Full textAbstract:
Solving many phenotyping problems involves not only automatic detection of objects in an image, but also counting the number of parts per object. We propose a solution in the form of a single deep network, tested for three agricultural datasets pertaining to bananas-per-bunch, spikelets-per-wheat-spike, and berries-per-grape-cluster. The suggested network incorporates object detection, object resizing, and part counting as modules in a single deep network, with several variants tested. The detection module is based on a Retina-Net architecture, whereas for the counting modules, two different architectures are examined: the first based on direct regression of the predicted count, and the other on explicit parts detection and counting. The results are promising, with the mean relative deviation between estimated and visible part count in the range of 9.2% to 11.5%. Further inference of count-based yield related statistics is considered. For banana bunches, the actual banana count (including occluded bananas) is inferred from the count of visible bananas. For spikelets-per-wheat-spike, robust estimation methods are employed to get the average spikelet count across the field, which is an effective yield estimator.
41
LI, HONGYAN, ALICE Z. CHUANG, and JOHN O’BRIEN. "Regulation of photoreceptor gap junction phosphorylation by adenosine in zebrafish retina." Visual Neuroscience 31, no. 3 (January 22, 2014): 237–43. http://dx.doi.org/10.1017/s095252381300062x.
Full textAbstract:
AbstractElectrical coupling of photoreceptors through gap junctions suppresses voltage noise, routes rod signals into cone pathways, expands the dynamic range of rod photoreceptors in high scotopic and mesopic illumination, and improves detection of contrast and small stimuli. In essentially all vertebrates, connexin 35/36 (gene homologs Cx36 in mammals, Cx35 in other vertebrates) is the major gap junction protein observed in photoreceptors, mediating rod–cone, cone–cone, and possibly rod–rod communication. Photoreceptor coupling is dynamically controlled by the day/night cycle and light/dark adaptation, and is directly correlated with phosphorylation of Cx35/36 at two sites, serine110 and serine 276/293 (homologous sites in teleost fish and mammals, respectively). Activity of protein kinase A (PKA) plays a key role during this process. Previous studies have shown that activation of dopamine D4 receptors on photoreceptors inhibits adenylyl cyclase, down-regulates cAMP and PKA activity, and leads to photoreceptor uncoupling, imposing the daytime/light condition. In this study, we explored the role of adenosine, a nighttime signal with a high extracellular concentration at night and a low concentration in the day, in regulating photoreceptor coupling by examining photoreceptor Cx35 phosphorylation in zebrafish retina. Adenosine enhanced photoreceptor Cx35 phosphorylation in daytime, but with a complex dose–response curve. Selective pharmacological manipulations revealed that adenosine A2a receptors provide a potent positive drive to phosphorylate photoreceptor Cx35 under the influence of endogenous adenosine at night. A2a receptors can be activated in the daytime as well by micromolar exogenous adenosine. However, the higher affinity adenosine A1 receptors are also present and have an antagonistic though less potent effect. Thus, the nighttime/darkness signal adenosine provides a net positive drive on Cx35 phosphorylation at night, working in opposition to dopamine to regulate photoreceptor coupling via a push–pull mechanism. However, the lower concentration of adenosine present in the daytime actually reinforces the dopamine signal through action on the A1 receptor.
42
Kenyon, E., A. Maminishkis, D. P. Joseph, and S. S. Miller. "Apical and basolateral membrane mechanisms that regulate pHi in bovine retinal pigment epithelium." American Journal of Physiology-Cell Physiology 273, no. 2 (August 1, 1997): C456—C472. http://dx.doi.org/10.1152/ajpcell.1997.273.2.c456.
Full textAbstract:
pH regulation was studied in fresh explant bovine retinal pigment epithelium-choroid using the pH-sensitive dye 2',7'-bis(carboxyethyl)-5(6)-carboxyfluorescein and intracellular microelectrodes. Acid recovery was HCO3 dependent, inhibited by apical amiloride and apical or basal 4,4'-diisothiocyanostilbene-2,2'-disulfonic acid (DIDS), and required apical and basal Na. Alkali recovery was HCO3 dependent and inhibitable by apical or basal DIDS. Three apical and two basolateral transporters were identified. Four contribute to acid extrusion, i.e., apical Na/H exchange, apical H-lactate cotransport, and apical Na-HCO3 cotransport and basolateral Na-HCO3 cotransport. At least two contribute to alkali extrusion, i.e., apical Na-HCO3 cotransport and a basolateral HCO3-dependent, DIDS-inhibitable mechanism, possibly Na-HCO3 cotransport, Cl/HCO3 exchange, or both. The apical Na-HCO3 cotransporter is electrogenic, carrying net negative charge inward. Basal Cl removal or addition of basal HCO3 caused HCO3- and Cl-dependent alkalinizations, respectively. Apical DIDS increased both responses. These cytosolic pH (pHi) regulatory mechanisms are so tightly coupled that changes in pHi can only occur after two or more of them are inhibited. In addition, these mechanisms help provide pathways for transport of Na and HCO3 across the retinal pigment epithelium between the blood and the distal retina.
43
Tian, N. M., T. Pratt, and D. J. Price. "Foxg1 regulates retinal axon pathfinding by repressing an ipsilateral program in nasal retina and by causing optic chiasm cells to exert a net axonal growth-promoting activity." Development 135, no. 24 (December 15, 2008): 4081–89. http://dx.doi.org/10.1242/dev.023572.
Full text
44
Staurenghi, Giovanni, Jose Cunha Vaz, and Jean-Francois Korobelnik. "Optical Coherence Tomography Angiography of the Retinal Microvasculature using the Zeiss AngioPlex." European Ophthalmic Review 09, no. 02 (2015): 147. http://dx.doi.org/10.17925/eor.2015.09.02.147.
Full textAbstract:
AngioPlex™ optical coherence tomography (OCT) is a new approach to diagnostic imaging in retinal disease. This technology uses amplitude and phase aspects of the OCT signal and novel algorithms to enable highly detailed visualisation of the retinal microvasculature. This system detects capillary flow rather than the presence of an injected dye. Unlike fluorescein angiography (FA), AngioPlex OCT imaging can discriminate capillaries at the superficial and deep retina and can also generate a colour map of the retinal vessels providing unparalleled views of diseased capillaries and net structures. Retinal and choroidal pathologies that are amenable to AngioPlex OCT examination include diabetic retinopathy (DR), vein occlusions, age-related macular degeneration and pathological myopia. In addition, AngioPlex OCT can be used to explore the vasculature of the optic nerve. These applications have considerable potential for improved retinal and choroidal disease diagnosis, but further experience and debate are needed to fully interpret the images. AngioPlex OCT is rapid, non-invasive and improves the patient experience and has been granted US Food and Drug Administration (FDA) 510(k) clearance for clinical use. The convenience and the range of retinal disease applications may encourage widespread adoption of the technology. AngioPlex OCT may therefore become a standard initial examination step for patients with retinal disease prior to and during treatment and may also indicate where further investigations including FA or indocyanine green are necessary.
45
Ghnemat, Rawan. "Hybrid Framework for Diabetic Retinopathy Stage Measurement Using Convolutional Neural Network and a Fuzzy Rules Inference System." Applied System Innovation 5, no. 5 (October 14, 2022): 102. http://dx.doi.org/10.3390/asi5050102.
Full textAbstract:
Diabetic retinopathy (DR) is an increasingly common eye disorder that gradually damages the retina. Identification at the early stage can significantly reduce the severity of vision loss. Deep learning techniques provide detection for retinal images based on data size and quality, as the error rate increases with low-quality images and unbalanced data classes. This paper proposes a hybrid intelligent framework of a conventional neural network and a fuzzy inference system to measure the stages of DR automatically, Diabetic Retinopathy Stage Measurement using Conventional Neural Network and Fuzzy Inference System (DRSM-CNNFIS). The fuzzy inference used human experts’ rules to overcome data dependency problems. At first, the Conventional Neural Network (CNN) model was used for feature extraction, and then fuzzy rules were used to measure diabetic retinopathy stage percentage. The framework is trained using images from Kaggle datasets (Diabetic Retinopathy Detection, 2022). The efficacy of this framework outperformed the other models with regard to accuracy, macro average precision, macro average recall, and macro average F1 score: 0.9281, 0.7142, 0.7753, and 0.7301, respectively. The evaluation results indicate that the proposed framework, without any segmentation process, has a similar performance for all the classes, while the other classification models (Dense-Net-201, Inception-ResNet ResNet-50, Xception, and Ensemble methods) have different levels of performance for each class classification.
46
Adam, Marwan, Sulaeman Martasuganda, and Eko Sri Wiyono. "ANALISIS PENGGUNAAN LIGHT FISHING DAN UNDERWATER LIGHT FISHING PADA BAGAN PERAHU DI PERAIRAN BOTANG LOMAN HALMAHERA SELATAN." ALBACORE Jurnal Penelitian Perikanan Laut 2, no. 1 (July 27, 2018): 29–42. http://dx.doi.org/10.29244/core.2.1.29-42.
Full textAbstract:
Bagan perahu adalah alat penangkapan ikan yang digolongkan ke dalam jaring angkat (lift net). Nelayan Desa Bajo menggunakan bagan perahu untuk menangkapan ikan pelagis kecil. Bagan perahu adalah salah satu alat tangkap yang dioperasikan pada malam hari dengan menggunakan cahaya. Nelayan Desa Bajo pada saat pengoperasian masih berdasarkan pengetahuan dan kebiasaan nelayan. Penelitian ini untuk mengetahui efektivitas dan efisiensi operasi penangkapan ikan maka diperlukan metode lain dalam penentuan daerah penangkapan ikan dengan menggunakan fish finder dan underwater light fishing, guna pengoperasian bagan perahu dapat berjalan dengan baik. Menganalisis hasil tangkapan berdasarkan bulan dan perlakuan. Hasil tangkapan selama penelitian menunjukkan bahwa total hasil tangkapan baik pada bulan gelap maupun bulan terang, jenis ikan layang dan ikan teri lebih mendominasi berada pada perlakuan kedua, sedangkan komposisi hasil tangkapan setelah tengah malam lebih banyak dibandingkan sebelum tengah malam, hal ini didasarkan dengan penelitian Sudirman et al. (2003) pada penelitiannya tentang adaptasi retina mata ikan layang (Decapterus ruselli) bahwa ikan yang dominan tertangkap adalah ikan layang karena teradaptasi sempurna oleh cahaya. Hasil penelitian dapat disimpulkan bahwa: Hasil tangkapan oleh enam jenis ikan yang dominan tertangkap adalah, spesies layang (Decapterus russelli) mendominasi selama penelitian. Fase bulan yang terbaik adalah pada saat bulan gelap. Hasil tangkapan pada perlakuan kedua menggunakan lampu normal, fish finder dan underwater light fishing mendominasi selama penelitian.Kata kunci: bagan perahu, light fishing, fish finder dan underwater light fishing.
47
Ravasio, Claudio S., Theodoros Pissas, Edward Bloch, Blanca Flores, Sepehr Jalali, Danail Stoyanov, Jorge M. Cardoso, Lyndon Da Cruz, and Christos Bergeles. "Learned optical flow for intra-operative tracking of the retinal fundus." International Journal of Computer Assisted Radiology and Surgery 15, no. 5 (April 22, 2020): 827–36. http://dx.doi.org/10.1007/s11548-020-02160-9.
Full textAbstract:
Abstract Purpose Sustained delivery of regenerative retinal therapies by robotic systems requires intra-operative tracking of the retinal fundus. We propose a supervised deep convolutional neural network to densely predict semantic segmentation and optical flow of the retina as mutually supportive tasks, implicitly inpainting retinal flow information missing due to occlusion by surgical tools. Methods As manual annotation of optical flow is infeasible, we propose a flexible algorithm for generation of large synthetic training datasets on the basis of given intra-operative retinal images. We evaluate optical flow estimation by tracking a grid and sparsely annotated ground truth points on a benchmark of challenging real intra-operative clips obtained from an extensive internally acquired dataset encompassing representative vitreoretinal surgical cases. Results The U-Net-based network trained on the synthetic dataset is shown to generalise well to the benchmark of real surgical videos. When used to track retinal points of interest, our flow estimation outperforms variational baseline methods on clips containing tool motions which occlude the points of interest, as is routinely observed in intra-operatively recorded surgery videos. Conclusions The results indicate that complex synthetic training datasets can be used to specifically guide optical flow estimation. Our proposed algorithm therefore lays the foundation for a robust system which can assist with intra-operative tracking of moving surgical targets even when occluded.
48
THORESON, WALLACE B., SALVATORE L. STELLA, ERIC J. BRYSON, JOHN CLEMENTS, and PAUL WITKOVSKY. "D2-like dopamine receptors promote interactions between calcium and chloride channels that diminish rod synaptic transfer in the salamander retina." Visual Neuroscience 19, no. 3 (May 2002): 235–47. http://dx.doi.org/10.1017/s0952523802192017.
Full textAbstract:
Activation of D2-like dopamine receptors in rods with quinpirole stimulates L-type calcium currents (ICa). This result appears inconsistent with studies showing that D2-like dopamine receptor activation diminishes rod signals in second-order retinal neurons. Since small reductions in [Cl−]i can inhibit photoreceptor ICa, we tested the hypothesis that enhancement of ICa with the D2/D4 receptor agonist, quinpirole, increases calcium-activated chloride currents (ICl(Ca)) causing an efflux of Cl− from rods that would provide a negative feedback inhibition of ICa. In agreement with studies from Xenopus, quinpirole reduced rod input to second-order neurons of tiger salamander retina without significantly altering rod voltage responses. Quinpirole also diminished the amplitude of depolarization-evoked increases in [Ca2+]i measured with Fura-2 in rods, a finding consistent with inhibition of synaptic transmission from rods. Electrophysiological and Cl−-imaging experiments indicated ECl in rods is ∼ −20 mV. Quinpirole enhanced ICl(Ca) and elicited an efflux of Cl− at the resting potential. A similar Cl− efflux was produced by extracellular replacement of 24 mM Cl− with CH3SO4− and this low Cl− solution inhibited Ca2+responses to a similar degree as quinpirole did. When ICl(Ca) was inhibited with niflumic acid, quinpirole enhanced both ICa and depolarization-evoked increases in [Ca2+]i. Furthermore, with niflumic acid, quinpirole no longer inhibited rod inputs into horizontal and bipolar cells. These results suggest an initial enhancement of ICa by quinpirole is followed by a stimulation of Cl− currents, including ICl(Ca). The net result is a Cl− efflux that inhibits depolarization-evoked increases in [Ca2+]i and synaptic transmission from rods.
49
Magera, Lukáš, Jan Krásný, Patrik Pluhovský, and Lucie Holubová. "Changes of the Foveal Avascular Zone and Macular Microvasculature within the Framework of OCT Angiography Examination in Young Patients with Type 1 Diabetes (Pilot Study)." Czech and Slovak Ophthalmology 76, no. 3 (October 10, 2020): 111–17. http://dx.doi.org/10.31348/2020/19.
Full textAbstract:
Aim: Learn about the development and changes in foveal avascular zone (FAZ) and vascularity of retina in the surrounding zone, depending on the duration in young diabetic patients type 1 (T1DM). Methods: As part of regular one-year examinations of young T1DM patients at the Eye Clinic of the University Hospital Královské Vinohrady in Prague (Czech Republic, EU) from January to December 2019, OCT angiography using the device Spectralis (Heidelberg Engineering) was included. Forty patients aged 18 to 30 years were examined, median 21 years. T1DM was diagnosed in childhood and lasted for more than 10 years. At the same time, a control group of forty individuals of similar age, without metabolic and other general disease was examined, normal visual acuity and physiological fundoscopic finding were obligatory. The FAZ size was evaluated in both groups (using built-in function "Draw Region"), also its shape, density decrease and change in character of vascularity of the retina was assessed. Results: In the control group, the FAZ area ranged from 0.06 to 0.4 mm², with an average of 0.253 ± 0.092 mm² and a median of 0.27 mm². It was not affected by a fundamental change in its round shape and the surrounding capillary netting was regular and reasonably dense. In T1DM patients, the FAZ area was in a wider range, from 0.05 to 0.64 mm², an average of 0.300 ± 0.132 mm², and a median of 0.31 mm². The difference in FAZ across-the-board evaluation was statistically significant (p = 0, 009). Diabetic preretinopathy (DpR) was defined by the irregularity of the capillary density up to the manifestation of non-perfusion, in 61% of cases the size of the FAZ was changed. In diabetic retinopathy (DR) there was always an irregularity of the FAZ shape with its enlargement, manifestation of non-perfusion, capillary dilatation and rare microaneurysms. Conclusion: Changes in FAZ size corresponded to the stage of T1DM on the fundoscopic finding of the eye depending on its duration. The initial increased amount of foveal capillaries, which resulted in decreased FAZ area, was followed by a gradual decrease in capillaries and increased FAZ area, consistent with the manifestations of DpR. It was accompanied by a change in capillary density in macula to eventual non-perfusion. On the contrary, the increase in the FAZ area and its irregularity accompanied by non-perfusion of the capillary net and microaneurysms corresponded to the development of DR already.
50
Yue, Chen, Mingquan Ye, Peipei Wang, Daobin Huang, and Xiaojie Lu. "Generative Adversarial Network Combined with SE-ResNet and Dilated Inception Block for Segmenting Retinal Vessels." Computational Intelligence and Neuroscience 2022 (August 28, 2022): 1–13. http://dx.doi.org/10.1155/2022/3585506.
Full textAbstract:
This study develops an accurate method based on the generative adversarial network (GAN) that targets the issue of the current discontinuity of micro vessel segmentation in the retinal segmentation images. The processing of images has become increasingly efficient since the advent of deep learning method. We have proposed an improved GAN combined with SE-ResNet and dilated inception block for the segmenting retinal vessels (SAD-GAN). The GAN model has been improved with respect to the following points. (1) In the generator, the original convolution block is replaced with SE-ResNet module. Furthermore, SE-Net can extract the global channel information, while concomitantly strengthening and weakening the key features and invalid features, respectively. The residual structure can alleviate the issue of gradient disappearance. (2) The inception block and dilated convolution are introduced into the discriminator, which enhance the transmission of features and expand the acceptance domain for improved extraction of the deep network features. (3) We have included the attention mechanism in the discriminator for combining the local features with the corresponding global dependencies, and for highlighting the interdependent channel mapping. SAD-GAN performs satisfactorily on public retina datasets. On DRIVE dataset, ROC_AUC and PR_AUC reach 0.9813 and 0.8928, respectively. On CHASE_DB1 dataset, ROC_AUC and PR_AUC reach 0.9839 and 0.9002, respectively. Experimental results demonstrate that the generative adversarial model, combined with deep convolutional neural network, enhances the segmentation accuracy of the retinal vessels far above that of certain state-of-the-art methods.