We present a MR brain atlas for structure and function (diffusion weighted images). The atlas is ... more We present a MR brain atlas for structure and function (diffusion weighted images). The atlas is customizable for contrast and orientation to match the current patient images. In addition, the atlas also provides normative values of MR parameters (T1, T2 and ADC values). The atlas is designed on informatics principles to provide context sensitive decision support at the time of primary image interpretation. Additional support for diagnostic interpretation is provided by a list of expert created most relevant 'Image Finding Descriptors' that will serve as cues to the user. The architecture of the atlas module is integrated into the image workflow of a radiology department to provide support at the time of primary diagnosis.
Standardized medical terminologies are gaining importance in the representation of medical data. ... more Standardized medical terminologies are gaining importance in the representation of medical data. In this paper, we present the evaluation of the SNOMED3.5 medical terminology to code concepts routinely used in chest radiology reports. Integration of this terminology mapper into a radiology reporting workstation that incorporates a speech recognition system and a natural language processor is also discussed. A total of 700 anatomical location terms (including synonyms) were tested and 72% of the terms had corresponding SNOMED terms. Of the 28% that did not result in a match, 16% were either morphological variants of SNOMED terms or could be found from a combination of terms from two or more SNOMED axes. Only 12% of the terms (primarily specialized radiology terms) were concepts not actually included in the SNOMED terminology.
We show how to generate case-based explanations for non-case-based learning methods such as artif... more We show how to generate case-based explanations for non-case-based learning methods such as artificial neural nets or decision trees. The method uses the trained model (e.g., the neural net or the decision tree) as a distance metric to determine which cases in the training set are most similar to the case that needs to be explained. This approach is well suited to medical domains, where it is important to understand predictions made by complex machine learning models, and where training and clinical practice makes users adept at case interpretation.
A multi-tiered telemedicine system based on Java and object-oriented database technology has yiel... more A multi-tiered telemedicine system based on Java and object-oriented database technology has yielded a number of practical insights and experiences on their effectiveness and suitability as implementation bases for a health care infrastructure. The advantages and drawbacks to their use, as seen within the context of the telemedicine system's development, are discussed. Overall, these technologies deliver on their early promise, with a few remaining issues that are due primarily to their relative newness.
Purpose: Magnetic Resonance Imaging is a challenging area of Medical Physics. Concepts presented ... more Purpose: Magnetic Resonance Imaging is a challenging area of Medical Physics. Concepts presented in didactic lectures can be reinforced if hands‐on laboratory procedures can be simultaneously performed by students. This can be accomplished on a clinical scanner if access is unlimited. However, the constraints of scanner availability usually preclude students from experimenting with different procedures. Further, it is not feasible to allow students to experiment extensively with clinical systems since scanner performance cannot be compromised. Methods: To address the need for hands‐on training as well as the limited availability of MRscanners, we have integrated a desktop Earth's field MRI system to reinforce concepts in NMR and MRI and a video presentation of a comprehensive MR Quality Control procedure performed by a MR physicist. The desktop Earth's Field MRI system (TerraNova) allows students to learn NMR and MRI fundamentals. The video records a MR QC procedure and takes students through the different steps to perform the comprehensive QC exam for ABR. In addition, the video highlights procedural errors and simulates conditions that result in commonly encountered artifacts. The video and MRimages are provided to students so that they can independently analyze the images. The sequence of presentation was as follows: didactic lecture (prior semester), desktop experiments, video of MRQC culminating with the actual clinical rotation. Results: A one hour lecture and 4 hour lab was developed with the desktop model and students successfully completed the lab in groups of two. Video and image analysis were completed individually by each student prior to the clinical lab rotation. Surveys conducted at each step confirmed the additional learning value of the desktop model and the video. Conclusions: In addition to enhancing the learning of MRI, the novel approach also resulted in an improved efficiency of the MR clinical rotation. Funding Support: P1 16V090024, Department of Education, FIPSE.
Recent advances in imaging have lead to increases in the number of images/study. Automated method... more Recent advances in imaging have lead to increases in the number of images/study. Automated methods to select relevant images are critical to effectively convey study results. The proposed method combines natural language processing (NLP) and automatic structure localization to identify relevant images of a MR brain study. NLP extracts relevant locations of findings. Two algorithms were implemented and evaluated for structure localization. The first method involves registration of patient dataset to a labeled atlas. The second method involves an eigenimage search using a training set of images. A prototype was developed and tested on MR brain studies of nine patients. With the registration method, slices containing the relevant structure agreed with expert selection in 98% of cases. Structure localization by eigenimage search was able to locate the lateral ventricles correctly in all the test cases. The proposed method provides an accurate method for identifying relevant slices of an...
International Journal for Numerical Methods in Biomedical Engineering
Passive materials in human skeletal muscle tissues play an important role in force output of skel... more Passive materials in human skeletal muscle tissues play an important role in force output of skeletal muscles. This paper introduces a multiscale modeling framework to investigate how age-associated variations in micro-scale passive muscle components, including microstructural geometry (e.g., connective tissue thickness) and material properties (e.g., anisotropy), influence the force output and deformations of the continuum skeletal muscle. We first define a representative volume element (RVE) for the microstructure of muscle and determine the homogenized macro-scale mechanical properties of the RVE from the separate mechanical properties of the individual components of the RVE, including muscle fibers and connective tissue with its associated collagen fibers. The homogenized properties of the RVE are then used to define the elements of the continuum muscle model to evaluate the force output and deformations of the whole muscle. Conversely, the regional deformations of the continuum model are fed back to the RVE model to determine the responses of the individual micro-scale components. Simulations of muscle isometric contractions at a range of muscle lengths are performed to investigate the effects of muscle architectural changes (e.g., pennation angles) due to ageing on force output and muscle deformation. The correlations between the pennation angle, the shear deformation in the micro-scale connective tissue (an indicator for the lateral force transmission), the angle difference between the fiber direction and principal strain direction (PSD) and the resulting shear deformation at the continuum scale, as well as the force output of the skeletal muscle are also discussed. This article is protected by copyright. All rights reserved.
Background: The aim was to compare spin-lattice relaxation (T1) mapping from sequences with no fa... more Background: The aim was to compare spin-lattice relaxation (T1) mapping from sequences with no fat suppression and three fat suppression methods and Magnetization Transfer Saturation (MTsat) mapping, to identify regional and age-related differences in calf muscle. These differences may be of clinical significance in age-related loss of muscle force. Methods: Ten young and seven senior subjects were imaged on a 3T MRI scanner using a 3D Fast Low Angle Shot sequence without and with different fat suppression and with MT saturation pulse. Bland–Altman plots were used to assess T1 maps using the fat unsuppressed sequence as the reference image. Age and regional differences in T1 and in MTsat were assessed using two-way factorial analyses of variance (ANOVAs) with Bonferroni-adjusted independent sample t-tests for post hoc analyses. Results: A significant age-related increase in T1 and decrease in MTsat was seen in the calf muscles. The largest size effect was observed in the T1 sequence...
— We present a new method for automatic segmentation of heterogeneous image data that takes a ste... more — We present a new method for automatic segmentation of heterogeneous image data that takes a step toward bridging the gap between bottom-up affinity-based segmentation methods and top-down generative model based approaches. The main contribution of the paper is a Bayesian formulation for incorporating soft model assignments into the calculation of affinities, which are conventionally model free. We integrate the resulting model-aware affinities into the multilevel segmentation by weighted aggregation algorithm, and apply the technique to the task of detecting and segmenting brain tumor and edema in multichannel MR volumes. The computationally efficient method runs orders of magnitude faster than current state-ofthe-art techniques giving comparable or improved results. Our quantitative results indicate the benefit of incorporating modelaware affinities into the segmentation process for the difficult case of brain tumor. Index Terms — Multilevel segmentation, normalized cuts, Bayesia...
Proceedings of the Second Joint 24th Annual Conference and the Annual Fall Meeting of the Biomedical Engineering Society] [Engineering in Medicine and Biology
A fully automated, fast and accurate method for segmenting contrast enhanced T1 weighted MR head ... more A fully automated, fast and accurate method for segmenting contrast enhanced T1 weighted MR head images into brain and non-brain has been developed. The algorithm employs information acquired from the smoothed MR intensity histogram to define thresholds that can be used to first remove the background noise and second segment head mask into smaller regions. A fuzzy clustering technique was
We present a MR brain atlas for structure and function (diffusion weighted images). The atlas is ... more We present a MR brain atlas for structure and function (diffusion weighted images). The atlas is customizable for contrast and orientation to match the current patient images. In addition, the atlas also provides normative values of MR parameters (T1, T2 and ADC values). The atlas is designed on informatics principles to provide context sensitive decision support at the time of primary image interpretation. Additional support for diagnostic interpretation is provided by a list of expert created most relevant 'Image Finding Descriptors' that will serve as cues to the user. The architecture of the atlas module is integrated into the image workflow of a radiology department to provide support at the time of primary diagnosis.
Standardized medical terminologies are gaining importance in the representation of medical data. ... more Standardized medical terminologies are gaining importance in the representation of medical data. In this paper, we present the evaluation of the SNOMED3.5 medical terminology to code concepts routinely used in chest radiology reports. Integration of this terminology mapper into a radiology reporting workstation that incorporates a speech recognition system and a natural language processor is also discussed. A total of 700 anatomical location terms (including synonyms) were tested and 72% of the terms had corresponding SNOMED terms. Of the 28% that did not result in a match, 16% were either morphological variants of SNOMED terms or could be found from a combination of terms from two or more SNOMED axes. Only 12% of the terms (primarily specialized radiology terms) were concepts not actually included in the SNOMED terminology.
We show how to generate case-based explanations for non-case-based learning methods such as artif... more We show how to generate case-based explanations for non-case-based learning methods such as artificial neural nets or decision trees. The method uses the trained model (e.g., the neural net or the decision tree) as a distance metric to determine which cases in the training set are most similar to the case that needs to be explained. This approach is well suited to medical domains, where it is important to understand predictions made by complex machine learning models, and where training and clinical practice makes users adept at case interpretation.
A multi-tiered telemedicine system based on Java and object-oriented database technology has yiel... more A multi-tiered telemedicine system based on Java and object-oriented database technology has yielded a number of practical insights and experiences on their effectiveness and suitability as implementation bases for a health care infrastructure. The advantages and drawbacks to their use, as seen within the context of the telemedicine system's development, are discussed. Overall, these technologies deliver on their early promise, with a few remaining issues that are due primarily to their relative newness.
Purpose: Magnetic Resonance Imaging is a challenging area of Medical Physics. Concepts presented ... more Purpose: Magnetic Resonance Imaging is a challenging area of Medical Physics. Concepts presented in didactic lectures can be reinforced if hands‐on laboratory procedures can be simultaneously performed by students. This can be accomplished on a clinical scanner if access is unlimited. However, the constraints of scanner availability usually preclude students from experimenting with different procedures. Further, it is not feasible to allow students to experiment extensively with clinical systems since scanner performance cannot be compromised. Methods: To address the need for hands‐on training as well as the limited availability of MRscanners, we have integrated a desktop Earth's field MRI system to reinforce concepts in NMR and MRI and a video presentation of a comprehensive MR Quality Control procedure performed by a MR physicist. The desktop Earth's Field MRI system (TerraNova) allows students to learn NMR and MRI fundamentals. The video records a MR QC procedure and takes students through the different steps to perform the comprehensive QC exam for ABR. In addition, the video highlights procedural errors and simulates conditions that result in commonly encountered artifacts. The video and MRimages are provided to students so that they can independently analyze the images. The sequence of presentation was as follows: didactic lecture (prior semester), desktop experiments, video of MRQC culminating with the actual clinical rotation. Results: A one hour lecture and 4 hour lab was developed with the desktop model and students successfully completed the lab in groups of two. Video and image analysis were completed individually by each student prior to the clinical lab rotation. Surveys conducted at each step confirmed the additional learning value of the desktop model and the video. Conclusions: In addition to enhancing the learning of MRI, the novel approach also resulted in an improved efficiency of the MR clinical rotation. Funding Support: P1 16V090024, Department of Education, FIPSE.
Recent advances in imaging have lead to increases in the number of images/study. Automated method... more Recent advances in imaging have lead to increases in the number of images/study. Automated methods to select relevant images are critical to effectively convey study results. The proposed method combines natural language processing (NLP) and automatic structure localization to identify relevant images of a MR brain study. NLP extracts relevant locations of findings. Two algorithms were implemented and evaluated for structure localization. The first method involves registration of patient dataset to a labeled atlas. The second method involves an eigenimage search using a training set of images. A prototype was developed and tested on MR brain studies of nine patients. With the registration method, slices containing the relevant structure agreed with expert selection in 98% of cases. Structure localization by eigenimage search was able to locate the lateral ventricles correctly in all the test cases. The proposed method provides an accurate method for identifying relevant slices of an...
International Journal for Numerical Methods in Biomedical Engineering
Passive materials in human skeletal muscle tissues play an important role in force output of skel... more Passive materials in human skeletal muscle tissues play an important role in force output of skeletal muscles. This paper introduces a multiscale modeling framework to investigate how age-associated variations in micro-scale passive muscle components, including microstructural geometry (e.g., connective tissue thickness) and material properties (e.g., anisotropy), influence the force output and deformations of the continuum skeletal muscle. We first define a representative volume element (RVE) for the microstructure of muscle and determine the homogenized macro-scale mechanical properties of the RVE from the separate mechanical properties of the individual components of the RVE, including muscle fibers and connective tissue with its associated collagen fibers. The homogenized properties of the RVE are then used to define the elements of the continuum muscle model to evaluate the force output and deformations of the whole muscle. Conversely, the regional deformations of the continuum model are fed back to the RVE model to determine the responses of the individual micro-scale components. Simulations of muscle isometric contractions at a range of muscle lengths are performed to investigate the effects of muscle architectural changes (e.g., pennation angles) due to ageing on force output and muscle deformation. The correlations between the pennation angle, the shear deformation in the micro-scale connective tissue (an indicator for the lateral force transmission), the angle difference between the fiber direction and principal strain direction (PSD) and the resulting shear deformation at the continuum scale, as well as the force output of the skeletal muscle are also discussed. This article is protected by copyright. All rights reserved.
Background: The aim was to compare spin-lattice relaxation (T1) mapping from sequences with no fa... more Background: The aim was to compare spin-lattice relaxation (T1) mapping from sequences with no fat suppression and three fat suppression methods and Magnetization Transfer Saturation (MTsat) mapping, to identify regional and age-related differences in calf muscle. These differences may be of clinical significance in age-related loss of muscle force. Methods: Ten young and seven senior subjects were imaged on a 3T MRI scanner using a 3D Fast Low Angle Shot sequence without and with different fat suppression and with MT saturation pulse. Bland–Altman plots were used to assess T1 maps using the fat unsuppressed sequence as the reference image. Age and regional differences in T1 and in MTsat were assessed using two-way factorial analyses of variance (ANOVAs) with Bonferroni-adjusted independent sample t-tests for post hoc analyses. Results: A significant age-related increase in T1 and decrease in MTsat was seen in the calf muscles. The largest size effect was observed in the T1 sequence...
— We present a new method for automatic segmentation of heterogeneous image data that takes a ste... more — We present a new method for automatic segmentation of heterogeneous image data that takes a step toward bridging the gap between bottom-up affinity-based segmentation methods and top-down generative model based approaches. The main contribution of the paper is a Bayesian formulation for incorporating soft model assignments into the calculation of affinities, which are conventionally model free. We integrate the resulting model-aware affinities into the multilevel segmentation by weighted aggregation algorithm, and apply the technique to the task of detecting and segmenting brain tumor and edema in multichannel MR volumes. The computationally efficient method runs orders of magnitude faster than current state-ofthe-art techniques giving comparable or improved results. Our quantitative results indicate the benefit of incorporating modelaware affinities into the segmentation process for the difficult case of brain tumor. Index Terms — Multilevel segmentation, normalized cuts, Bayesia...
Proceedings of the Second Joint 24th Annual Conference and the Annual Fall Meeting of the Biomedical Engineering Society] [Engineering in Medicine and Biology
A fully automated, fast and accurate method for segmenting contrast enhanced T1 weighted MR head ... more A fully automated, fast and accurate method for segmenting contrast enhanced T1 weighted MR head images into brain and non-brain has been developed. The algorithm employs information acquired from the smoothed MR intensity histogram to define thresholds that can be used to first remove the background noise and second segment head mask into smaller regions. A fuzzy clustering technique was
Uploads
Papers by Usha Sinha