Academic literature on the topic 'Mri sequence optimization'

Magnetic Resonance Imaging (MRI) examination of the brain at high resolution will be able to detect abnormalities in the brain that are not detected before. The MRI machine is equipped with a very strong magnetic force, therefore metal objects can interfere with the workings of the machine . Some patients, there may be a magnet in the body unnoticed, for example: orthodontic bracket. The purpose of this study was to analyze the effectiveness of sequences to reduce metal artifacts due to the installation of the Orthodontic Bracket. This type of research is a quantitative analytic with a quasi-experimental research design. The research design used was a one group pretest-posttest design to determine the optimization of the application of the Slice Encoding Metal Artifact Correction (SEMAC) and View Angle Tilting (VAT) sequences on Brain MRI with Orthodontic Bracket. The results showed that the SEMAC sequence combined with T2 TSE was able to reduce metal artifacts well. VAT sequences combined with T2 TSE were able to reduce metal artifacts quite well. A more optimal sequence to reduce metal artifacts is T2 TSE SEMAC, where the sequence is able to reveal thin structures that are not visible in the T2 TSE or T2 TSE VAT sequences.

Abstract Objective: To emphasize the most appropriate magnetic resonance imaging (MRI) diffusion protocol for the detection of lesions that cause transient global amnesia, in order to perform an accurate examination, as well as to determine the ideal time point after the onset of symptoms to perform the examination. Materials and Methods: We evaluated five patients with a diagnosis of transient global amnesia treated between 2012 and 2015. We analyzed demographic characteristics, clinical data, symptom onset, diffusion techniques, and radiological findings. Examination techniques included a standard diffusion sequence (b value = 1000 s/mm2; slice thickness = 5 mm) and a optimized diffusion sequence (b value = 2000 s/mm2; slice thickness = 3 mm). Results: Brain MRI was performed at 24 h or 36 h after symptom onset, except in one patient, in whom it was performed at 12 h after (at which point no changes were seen) and repeated at 36 h after symptom onset (at which point it showed alterations in the right hippocampus). The standard and optimized diffusion sequences were both able to demonstrate focal changes in the hippocampi in all of the patients but one, in whom the changes were demonstrated only in the optimized sequence. Conclusion: MRI can confirm a clinical hypothesis of transient global amnesia. Knowledge of the optimal diffusion parameters and the ideal timing of diffusion-weighted imaging (> 24 h after symptom onset) are essential to improving diagnostic efficiency.

Magnetic resonance imaging (MRI) is known to provide a useful approach for the exploration of the chemistry and dynamics of a wide range of soft condensed materials. However, its application to solids has been limited to those materials with relatively narrow resonances. The time needed to obtain an image of a solid with a given resolution and signal-to-noise ratio (SNR) is directly proportional to the line width of the resonance. For MRI to become practical for the imaging of solids it will have to rely on the development and use of MR sequences that avoid the issues raised by line broadening of the resonance. In this paper we review the latest contributions towards MR imaging of solids from our laboratory, in particular, applications using optimized gradient waveforms. Acoustic noise reduction and SNR improvement obtained with modifications of the standard single-point imaging sequence are presented and discussed using examples.

Background Magnetic resonance imaging (MRI) is widely used to diagnose acute abdominal pain; however, it remains unclear which pulse sequence has priority in acute abdominal pain. Purpose To investigate the diagnostic accuracy of MRI and to assess the conspicuity of each pulse sequence for the diagnosis of acute abdominal pain due to gastrointestinal diseases Material and Methods We retrospectively enrolled 60 patients with acute abdominal pain who underwent MRI for axial and coronal T2-weighted (T2W) imaging, fat-suppressed (FS)-T2W imaging, and true-fast imaging with steady-state precession (True-FISP) and axial T1-weighted (T1W) imaging and investigated the diagnosis with endoscopy, surgery, histopathology, computed tomography, and clinical follow-up as standard references. Two radiologists determined the diagnosis with MRI and rated scores of the respective sequences in assessing intraluminal, intramural, and extramural abnormality using a 5-point scale after one month. Diagnostic accuracy was calculated and scores were compared by Wilcoxon-signed rank test with Bonferroni correction. Results Diagnostic accuracy was 90.0% and 93.3% for readers 1 and 2, respectively. Regarding intraluminal abnormality, T2W, FS-T2W, and True-FISP imaging were superior to T1W imaging in both readers. FS-T2W imaging was superior to True-FISP in reader 2 ( P < 0.0083). For intramural findings, there was no significant difference in reader 1, whereas T2W, FS-T2W, and True-FISP imaging were superior to T1W imaging in reader 2 ( P < 0.0083). For extramural findings, FS-T2W imaging was superior to T2W, T1W, and True-FISP imaging in both readers ( P < 0.0083). Conclusion T2W and FS-T2W imaging are pivotal pulse sequences and should be obtained before T1W and True-FISP imaging.

BACKGROUND: Nowadays, the total knee arthroplasty (TKA) technique plays an important role in surgical treatment for patients with severe knee osteoarthritis (OA). However, there are still several key issues such as promotion of osteotomy accuracy and prosthesis matching degree that need to be addressed. OBJECTIVE: It is significant to construct an accurate three-dimensional (3D) geometric anatomy structure model of subject-specific human knee joint with major bone and soft tissue structures, which greatly contributes to obtaining personalized osteotomy guide plate and suitable size of prosthesis. METHODS: Considering different soft tissue structures, magnetic resonance imaging (MRI) scanning sequences involving two-dimensional (2D) spin echo (SE) sequence T1 weighted image (T1WI) and 3D SE sequence T2 weighted image (T2WI) fat suppression (FS) are selected. A 3D modeling methodology based on computed tomography (CT) and two sets of MRI images is proposed. RESULTS: According to the proposed methods of image segmentation and 3D model registration, a novel 3D knee joint model with high accuracy is finally constructed. Furthermore, remeshing is used to optimize the established model by adjusting the relevant parameters. CONCLUSIONS: The modeling results demonstrate that reconstruction and optimization model of 3D knee joint can clearly and accurately reflect the key characteristics, including anatomical structure and geometric morphology for each component.

Radiotherapy plays an increasingly important role in cancer treatment, and medical imaging plays an increasingly important role in radiotherapy. Magnetic resonance imaging (MRI) is poised to be a major component in the development towards more effective radiotherapy treatments with fewer side effects. This thesis attempts to contribute in realizing this potential. Radiotherapy planning requires simulation of radiation transport. The necessary physical properties are typically derived from CT images, but in some cases only MR images are available. In such a case, a crude but common approach is to approximate all tissue properties as equivalent to those of water. In this thesis we propose two methods to improve upon this approximation. The first uses a machine learning algorithm to automatically identify bone tissue in MR. The second, which we refer to as atlas-based regression, can be used to generate a realistic, patient-specific, pseudo-CT directly from anatomical MR images. Atlas-based regression uses deformable registration to estimate a pseudo-CT of a new patient based on a database of aligned MR and CT pairs. Cancerous tissue has a dierent structure from normal tissue. This affects molecular diusion, which can be measured using MRI. The prototypical diusion encoding sequence has recently been challenged with the introduction of more general  waveforms. To take full advantage of their capabilities it is, however, imperative to respect the constraints imposed by the hardware while at the same time maximizing the diffusion encoding strength. In this thesis we formulate this as a constrained optimization problem that is easily adaptable to various hardware constraints.

Diffusion weighted imaging (DWI) is a branch within the field of magnetic resonance imaging (MRI) that relies on the diffusion of water molecules for its contrast. Its clinical applications include the early diagnosis of ischemic stroke and mapping of the nerve tracts of the brain. The recent development of filter exchange imaging (FEXI) and the introduction of the apparent exchange rate (AXR) present a new DWI based technique that uses the exchange of water between compartments as contrast. FEXI could offer new clinical possibilities in diagnosis, differentiation and treatment follow-up of conditions involving edema or altered membrane permeability, such as tumors, cerebral edema, multiple sclerosis and stroke. Necessary steps in determining the potential of AXR as a new biomarker include running comparative studies between controls and different patient groups, looking for conditions showing large AXR-changes. However, before designing such studies, the experimental protocol of FEXI should be optimized to minimize the experimental variance. Such optimization would improve the data quality, shorten the scan time and keep the required study group sizes smaller.  Here, optimization was done using an active imaging approach and the Cramer-Rao lower bound (CRLB) of Fisher information theory. Three optimal protocols were obtained, each specialized at different tissue types, and the CRLB method was verified by bootstrapping. A test-retest study of 18 volunteers was conducted in order to investigate the reproducibility of the AXR as measured by one of the protocols, adapted for the scanner. Group sizes required were calculated based on both CRLB and the variability of the test-retest data, as well as choices in data analysis such as region of interest (ROI) size. The result of this study is new protocols offering a reduction in coefficient of variation (CV) of around 30%, as compared to previously presented protocols. Calculations of group sizes required showed that they can be used to decide whether any patient group, in a given brain region, has large alterations of AXR using as few as four individuals per group, on average, while still keeping the scan time below 15 minutes. The test-retest study showed a larger than expected variability however, and uncovered artifact like changes in AXR between measurements. Reproducibility of AXR values ranged from modest to acceptable, depending on the brain region. Group size estimations based on the collected data showed that it is still possible to detect AXR difference larger than 50% in most brain regions using fewer than ten individuals. Limitations of this study include an imprecise knowledge of model priors and a possibly suboptimal modeling of the bias caused by weak signals. Future studies on FEXI methodology could improve the method further by addressing these matters and possibly also the unknown source of variability. For minimal variability, comparative studies of AXR in patient groups could use a protocol among those presented here, while choosing large ROI sizes and calculating the AXR based on averaged signals.

Contexto: Existe, actualmente, uma incessante necessidade por exames imagiológicos melhores e mais rápidos no campo da Ressonância Magnética Imagiológica (RMI) clinica. Isto é particularmente verdade em RMI músculo-esquelética (MSK), na qual longas listas de espera são um problema constante. Para além disto, fundos para adquirir ou actualizar equipamentos imagiológicos encontram-se totalmente ou parcialmente cortados, devido à difícil situação financeira na qual a maioria dos Sistemas de Saúde se encontra. Este projecto pretende avaliar se e até que ponto a Optimização de Sequências de RMI pode ser uma resposta ao desafio de melhorar a qualidade de imagem (QI) e reduzir o Tempo de Aquisição (TA) sem recurso a investimento financeiro. Metodologia: Foram criadas Sequências Optimizadas (SO), para scanners de 1,5 e 3 Tesla, que se focaram em obter a melhor relação QI/TA. As SO foram desenvolvidas pegando em sequências RMI genéricas, já disponibilizadas pelos fabricantes nos scanners, e depois manipulando os seus parâmetros RMI através de um processo iterativo, em conjunção com várias antenas RMI receptoras e fantomas de RMI biológicos e não biológicos. Após a melhor relação QI/TA ter sido estabelecida, para cada sequência, estas foram usadas para criar réplicas dos protocolos padrão do Departamento de RMI. A diferença em TA entre os protocolos velhos e os novos foi calculada para medir a redução de TA obtida. A mudança em QI foi avaliada através análise visual retrospectiva efectuada por avaliadores cegos, os quais tinham extensa experiência em RMI MSK. As respostas dos avaliadores foram registadas através de um questionário padrão que se focou na QI geral e em aspectos técnicos específicos (exemplo: Resolução Espacial, Contraste, Artefactos, etc.) Resultados: A redução média no TA foi de 6 minutos e 48 segundos por exame. Esta redução intensificou-se em protocolos com um maior número de sequências. T1 foi a ponderação que demonstrou maior redução do TA. No geral os avaliadores consideraram que as imagens obtidas com SO tinham melhor ou muito melhor QI que as imagens obtidas com sequências não optimizadas. Esta tendência esteve também presente para todos os aspectos técnicos de QI avaliados (p <0.05), exceptuando: Relação Sinal-Ruído e Artefactos. T1 foi a ponderação na qual houve maior melhoria da QI. Foi notado que as SO produziam mais barulho e maior Taxa de Absorção Especifica que as sequências não optimizadas, embora os níveis de segurança tenham sido respeitados. Conclusões: A Optimização de Sequências é um método útil para melhorar a QI e reduzir o TA, que não precisa de investimento monetário significativo. Naturalmente tem os seus limites mas, caso seja empregue correctamente por técnicos entendidos em RMI, é uma ferramenta versátil que pode fazer a diferença na qualidade de imagem e na carga de trabalho expedida pelo scanner.
Background: There is currently a nonstop necessity for faster and improved imaging, in the field of clinical Magnetic Resonance Imaging (MRI). This is particularly true in Musculoskeletal (MSK) MRI, were long waiting lists are a constant problem. In addition, funds to acquire or upgrade imaging equipment have been fully or partially cut, due to the difficult financial situation that most Healthcare Systems find themselves in. The project developed aims to evaluate if and to what extent MRI Sequence Optimization can be an answer to the challenge of improving Image Quality (IQ) and decreasing scan time duration without financial investment. Methodology: Optimized Sequences (OS) were created, for both 1.5 and 3 Tesla MRI scanners, which focused on providing the best Image Quality (IQ)/Time of Acquisition (TA) relationship. OS were developed by taking generic MRI sequences, already available in the scanners by the manufacturer, and then manipulating their MRI parameters using an iterative process in conjunction with several receiving Coils, and both non-biological and biological MRI phantoms. After the best IQ/TA relation was establish, for each sequence, they were used to create replicas of the MRI Department’s standard MRI protocols. The difference in TA between the old and new MRI protocols was calculated to measure the reduction in TA obtained. IQ change was assessed through retrospective visual analysis by several blinded assessors, which had extensive experience in MSK MRI. The assessor’s answers were recorded using a standard questionnaire that focused on overall IQ and also on specific technical aspects (e.g. Spatial Resolution, Contrast, Artefacts, etc.) Results: The average reduction in TA measured was 6 minutes & 48 seconds per examination. TA reduction was more marked for protocols with a higher number of sequences. T1 was the weight type that showed a more marked TA reduction. Overall the assessors deemed that images produced from OS had either better or significantly better IQ than images produced with non-OS sequences. This trend was also present for all IQ’s technical aspect assessed (p<0.05), with exception on Signal-to-Noise Ratio and Artefacts. T1 was considered the weight type were the most IQ improvement was observed. It was also noted that OS sequences produced higher audio noise and Specific Absorption Rate compared to non-OS, but that the safety levels were respected. Conclusion: Sequence Optimization is indeed a useful method to improve IQ and reduce TA, without requiring any significant monetary investment. Obviously it has its limits but, if employed correctly by MRI knowledgeable technicians, it is a versatile tool that can make a significant improvement in scanner workload output and image quality.

Tese de mestrado integrado em Engenharia Biomédica e Biofísica (Sinais e Imagens Médicas), Universidade de Lisboa, Faculdade de Ciências, 2021
O líquido cefalorraquidiano (LCR) tem um papel essencial na drenagem dos resíduos resultantes do metabolismo cerebral e o constante movimento a que este fluido está sujeito é vital para manter a homeostasia do cérebro. Com feito, alterações neste movimento, geralmente associadas com o envelhecimento ou com doença, levam a perturbações fisiológicas, como a doença de Alzheimer ou a hidrocefalia. Por esta razão, é fundamental consolidar e aprofundar o conhecimento referente a este fluido, nomeadamente perceber como varia a sua velocidade e deslocamento, pois só desta forma será possível desenvolver e aperfeiçoar a prevenção e tratamento destas perturbações. Com efeito, este fluido está em constante movimento e o seu comportamento está intimamente ligado ao ciclo cardíaco. Apesar de estudos anteriores sobre a velocidade e o deslocamento do líquido cefalorraquidiano através de métodos de Ressonância Magnética (RM), ainda não existe uma descrição completa sobre o comportamento deste fluido. O objetivo principal deste estudo, consistiu em obter uma descrição detalhada da velocidade e do deslocamento do LCR através da aquisição de imagens de ressonância magnética obtidas com contraste de fase, um método de referência no que toca ao estudo da velocidade de fluidos No entanto, utilizar RM de contraste de fase para adquirir velocidades mais baixas, como as do LCR, requer tempos de aquisição mais longos e, consequentemente, as imagens obtidas estão mais sujeitas a distorções. Assim, a segunda parte deste projecto partiu dos resultados de deslocamento obtidos através da RM com contraste de fase para otimizar os parâmetros de uma segunda sequência de MR. Esta sequência é relativamente recente e possibilita o estudo do deslocamento sub-milimétrico do LCR associado ao movimento do cérebro através da aplicação de gradientes sucessivos (DENSE). Porém, é necessária uma escolha rigorosa dos parâmetros utilizados de forma a obter resultados que retratem o deslocamento do LCR de uma forma rigorosa e exata. Na primeira parte deste projecto, quatro voluntários foram estudados utilizando RM com contraste de fase, entre outubro de 2019 e fevereiro de 2020, em concordância com as diretrizes éticas da University Medical Center em Utrecth, Países Baixos. As aquisições foram realizadas utilizando um scanner de RM Philips 7 T e dois tipos de contraste foram utilizados: contraste de fase com 1mm de resolução isotrópica e com uma codificação de velocidade de 5m/s, e imagens 3D com ponderação em T1 com 1mm de resolução isotrópica. As imagens foram obtidas para três orientações distintas: anterior posterior, inferior-superior, e direita-esquerda. Na segunda parte deste projecto, dois voluntários foram estudados, de janeiro a fevereiro de 2020, utilizando seis contrastes: contraste de fase com 1mm de resolução isotrópica, e imagens 3D com ponderação em T1 com 1mm de resolução isotrópica, uma sequência básica DENSE com 2mm de resolução isotrópica, uma sequência básica DENSE com 3mm de resolução isotrópica, uma sequência DENSE com uma preparação T2 com 3mm de resolução isotrópica e, finalmente, uma sequência DENSE com tempo de eco prolongado com 3mm de resolução isotrópica. No entanto, e ao contrário das imagens adquiridas na primeira parte deste projecto, as imagens da segunda parte foram obtidas apenas para a orientação inferior-superior. Todas as imagens adquiridas no decorrer desta dissertação foram obtidas com gating cardíaco. O gating cardíaco foi realizado através da utilização de um eletrocardiograma e de um oxímetro de pulso de modo a relacionar o evolução da velocidade e do deslocamento com o ciclo cardíaco. Neste projecto foi também desenvolvida uma pipeline que permite que a partir das imagens adquiridas seja possível estudar a velocidade e o deslocamento do LCR. Esta pipeline inclui diversos passos. O primeiro passo consistiu em realinhar e co-registar as imagens obtidas de forma a permitir uma análise voxel a voxel. Seguidamente, as imagens foram segmentas em três tipos de tecidos: LCR, substância cinzenta, e substância branca. Adicionalmente, as primeiras etapas foram realizadas através da utilização de toolboxs disponíveis no MATLAB como o SPM e o CAT12. Posteriormente, os artefactos presentes nas imagens, tais como as correntes-eddy, foram corrigidos. No decorrer deste projecto diversas regiões foram analisadas e foram divididas em dois grupos: regiões do sistema ventricular, nas quais se incluíram os ventrículos laterais, o terceiro e quarto ventrículo, o aqueduto de Sylvius e a Cisterna Magna; e regiões mais abrangentes, como a região anterior e posterior do cérebro. Estas áreas do cérebro foram selecionadas através da segmentação das imagens anatómicas. Finalmente, a velocidade de cada uma destas regiões foi extraída e integrada ao longo do ciclo cardíaco de maneira a calcular o deslocamento do LCR. Os resultados obtidos relativamente à velocidade mostraram consistência para os quatro voluntários deste projecto. Verificou-se que as regiões do sistema ventricular demonstram valores de velocidade consideravelmente mais elevados do que as regiões mais abrangentes. Com efeito, a região que apresentou valores absolutos de velocidade mais elevados foi o aqueducto de Sylvius. Adicionalmente, verificou-se que as velocidades são superiores na orientação caudal-cranial e inferiores na orientação direita-esquerda. Concluiu-se também que o valor de velocidade escolhido não foi o mais indicado para as regiões mais abrangentes pois a velocidade destas regiões é significativamente inferior e, desta forma, poderá ter existido perda de sinal do LCR. Posteriormente, ao integrar a velocidade obtida através da RM com contraste fase obtiveram-se mapas de deslocamento para as mesmas regiões cerebrais. Estes resultados mostraram-se consistentes e, tal como anteriormente observado, o deslocamento é consideravelmente superior para as regiões do sistema ventricular. A região inferior do cérebro foi a que apresentou valores de deslocamento mais elevados, o que pode ser justificado pelo facto desta região se encontrar mais próxima do coração e, desta maneira, o LCR ser ejetado das regiões que ocupa com maior velocidade. Adicionalmente, verificou-se que as maiores alterações do deslocamento ocorrem imediatamente após a sístole cardíaca. Seguidamente, foi possível, a partir dos valores de deslocamento obtidos, determinar um valor ótimo para a sensibilidade, relativamente ao deslocamento, da sequência DENSE. Contrariamente à primeira parte deste projecto, os resultados obtidos utilizando as sequências DENSE dizem respeito exclusivamente às regiões mais abrangentes. De facto, esta exclusão das regiões do sistema ventricular foi causada pela baixa resolução das imagens obtidas que, desta forma, não permitiram uma segmentação de áreas tão reduzidas com fiabilidade razoável. Os resultados desta análise mostram que a sequência utilizada cujos resultados de deslocamento se assemelham mais aos resultados obtidos através do contraste de fase foi a sequência que utilizou a preparação T2. Por oposição, as sequências básicas utilizadas mostraram semelhança reduzida com o método de comparação. Esta diferença observada foi justifica pela baixa resolução das imagens adquiridas, o que contribui para que não fosse possível eliminar o efeito de volume parcial. Adicionalmente, concluiu-se que o valor de sensibilidade para o deslocamento utilizado não foi o correto para estas regiões e, desta forma, houve perda de sinal adquirido justificando assim às diferenças encontradas entre os dois métodos. Concluindo, esta dissertação cumpriu o objetivo principal proposto, nomeadamente fazer uma descrição completa e quantificar a evolução da velocidade e do deslocamento do líquido cefalorraquidiano ao longo do ciclo cardíaco. Adicionalmente, o método de RM com contraste de fase mostrou ser um método fiável para o estudo do comportamento do LCR mesmo em regiões com velocidades mais lentas. Os resultados de deslocamento obtidos através da utilização do método DENSE permitiram confirmar o potencial desta técnica para medir deslocamentos sub-milimétricos. No entanto, este método ainda necessita de ser otimizado de forma a ser uma alternativa viável ao contraste de fase. Finalmente, os resultados obtidos neste estudo permitem que estudos futuros utilizem os valores máximos de cada região obtida de forma a otimizar futuras sequências.
Cerebrospinal fluid (CSF) plays an essential role in the drainage of cerebral waste, and its continuous motion is vital to maintain the brain’s homeostasis. Variations in this motion, associated with aging and disease, are observed in physical and physiological disorders, such as Alzheimer’s Disease. Therefore, a deep understating of this fluid motion, such as its velocity and displacement, is fundamental to strengthen our knowledge of these diseases and might be vital to their prevention and treatment. Despite previous studies reporting CSF velocity and displacement using magnetic resonance imaging techniques, a complete picture of this fluid motion has not yet been obtained. The aim of this study was to, first and foremost, obtain a general picture of CSF velocity and displacement using Phase Contrast (PC) MRI, a method of reference for velocity acquisition. Furthermore, this sequence was also used to optimize the parameters for an MRI technique called Displacement Encoding with Stimulated Echoes (DENSE), a sequence that was modified in order to be capable of measuring small displacements. Four healthy subjects were studied using whole-brain ultra-high field (UHF) MRI at 7 Tesla (T). The volunteers were scanned using two different MRI imaging sequences: Phase Contrast MRI at 1 mm isotropic resolution and 3D T1-weighted (T1w) at 1 mm isotropic resolution. Additionally, two healthy subjects were scanned using PC and four different DENSE acquisitions. Firstly, two basic DENSE sequences with 2mm and 3mm isotropic resolution were acquired. Next, a DENSE acquisition with a T2 prepared magnetization, and a DENSE sequence with a long echo time were acquired to avoid confounding effects from partial volume between tissue and CSF. The image processing pipeline included coregistration, segmentation, eddy current correction. Moreover, mean velocity and displacement maps were calculated for regions of interest previously selected. The results in this study obtained from the PC acquisitions show consistent velocity and displacement values across all subjects. Furthermore, CSF shows higher values for the ventricular regions, such as the aqueduct, and predominant motion in the anterior and feet direction. Comparatively, regions in the periphery of the brain display slower velocities and smaller displacements. The displacement values obtained with PC were used to optimize the displacement sensitivity used in the DENSE acquisition. The DENSE sequence acquired with a T2 magnetization preparation showed the most consistent results when compared to the Phase Contrast. In conclusion, this project managed to study and quantify CSF behavior in the brain, which allows for the optimization of future sequences that desire a more detailed study of this fluid’s in specific brain regions.

Modularity is an approach to manage the design of complex systems by partitioning and assigning elements of a concept to simpler subsystems according to a planned architecture. Functional-flow heuristics suggest possible modules that have been demonstrated in past products, but using them still leaves it to the designer to choose which heuristics make sense in a certain architecture. This constitutes an opportunity for a designer to take other constraints and objectives into account. With large complex systems, the number of alternative groupings of elements into modular chunks becomes exponentially large and some form of automation would be beneficial to accomplish this task. Clustering algorithms using the design structure matrix (DSM) representation search the space of alternative relative positioning of elements and present one ideal outcome ordering which “optimizes” a modularity metric. Beyond the problems of lack of interactive exploration around the optimized result, such approaches also partition the elements in an unconstrained manner. Yet, typical complex products are subject to constraints which invalidate the unconstrained optimization. Such architectural partitioning constraints include those associated with external force fields including electric, magnetic, or pressure fields that constrain some functions to perform or not perform in different regions of the field. There are also supplier constraints where some components cannot be easily provided with others. Overall, it is difficult to simply embed all objectives of modular thinking into one metric to optimize. We develop a new type of interactive clustering algorithm approach considering multiple objectives and partitioning constraints. Partitioning options are offered to a designer interactively as a sequence of clustering choices between elements in the architecture. A designer can incorporate constraints that determine the compatibility or incompatibility of elements by choosing among alternative groupings progressively. Our aim is to combine computational capability of clustering algorithms with the flexibility of manual approaches. Through applying these algorithms to a MRI machine injector, we demonstrate the benefits of interactive cooperation between a designer and modularity algorithms, where constraints can be naturally considered.