Deep learning approach for automated segmentation of myocardium using bone scintigraphy single-photon emission computed tomography/computed tomography in patients with suspected cardiac amyloidosis.

BACKGROUND: We employed deep learning to automatically detect myocardial bone-seeking uptake as a marker of transthyretin cardiac amyloid cardiomyopathy (ATTR-CM) in patients undergoing 99mTc-pyrophosphate (PYP) or hydroxydiphosphonate (HDP) single-photon emission computed tomography (SPECT)/computed tomography (CT). METHODS: We identified a primary cohort of 77 subjects at Brigham and Women's Hospital and a validation cohort of 93 consecutive patients imaged at the University of Pennsylvania who underwent SPECT/CT with PYP and HDP, respectively, for evaluation of ATTR-CM. Global heart regions of interest (ROIs) were traced on CT axial slices from the apex of the ventricle to the carina. Myocardial images were visually scored as grade 0 (no uptake), 1 (uptakeribs). A 2D U-net architecture was used to develop whole-heart segmentations for CT scans. Uptake was determined by calculating a heart-to-blood pool (HBP) ratio between the maximal counts value of the total heart region and the maximal counts value of the most superior ROI. RESULTS: Deep learning and ground truth segmentations were comparable (p=0.63). A total of 42 (55%) patients had abnormal myocardial uptake on visual assessment. Automated quantification of the mean HBP ratio in the primary cohort was 3.1±1.4 versus 1.4±0.2 (p<0.01) for patients with positive and negative cardiac uptake, respectively. The model had 100% accuracy in the primary cohort and 98% in the validation cohort. CONCLUSION: We have developed a highly accurate diagnostic tool for automatically segmenting and identifying myocardial uptake suggestive of ATTR-CM.

Tau maturation in the clinicopathological spectrum of Lewy body and Alzheimer's disease.

OBJECTIVE: Alzheimer's disease neuropathologic change and alpha-synucleinopathy commonly co-exist and contribute to the clinical heterogeneity of dementia. Here, we examined tau epitopes marking various stages of tangle maturation to test the hypotheses that tau maturation is more strongly associated with beta-amyloid compared to alpha-synuclein, and within the context of mixed pathology, mature tau is linked to Alzheimer's disease clinical phenotype and negatively associated with Lewy body dementia. METHODS: We used digital histology to measure percent area-occupied by pathology in cortical regions among individuals with pure Alzheimer's disease neuropathologic change, pure alpha-synucleinopathy, and a co-pathology group with both Alzheimer's and alpha-synuclein pathologic diagnoses. Multiple tau monoclonal antibodies were used to detect early (AT8, MC1) and mature (TauC3) epitopes of tangle progression. We used linear/logistic regression to compare groups and test the association between pathologies and clinical features. RESULTS: There were lower levels of tau pathology (ß = 1.86-2.96, p < 0.001) across all tau antibodies in the co-pathology group compared to the pure Alzheimer's pathology group. Among individuals with alpha-synucleinopathy, higher alpha-synuclein was associated with greater early tau (AT8 ß = 1.37, p < 0.001; MC1 ß = 1.2, p < 0.001) but not mature tau (TauC3 p = 0.18), whereas mature tau was associated with beta-amyloid (ß = 0.21, p = 0.01). Finally, lower tau, particularly TauC3 pathology, was associated with lower frequency of both core clinical features and categorical clinical diagnosis of dementia with Lewy bodies. INTERPRETATION: Mature tau may be more closely related to beta-amyloidosis than alpha-synucleinopathy, and pathophysiological processes of tangle maturation may influence the clinical features of dementia in mixed Lewy-Alzheimer's pathology.

Co-Clinical Imaging Metadata Information (CIMI) for Cancer Research to Promote Open Science, Standardization, and Reproducibility in Preclinical Imaging.

Preclinical imaging is a critical component in translational research with significant complexities in workflow and site differences in deployment. Importantly, the National Cancer Institute's (NCI) precision medicine initiative emphasizes the use of translational co-clinical oncology models to address the biological and molecular bases of cancer prevention and treatment. The use of oncology models, such as patient-derived tumor xenografts (PDX) and genetically engineered mouse models (GEMMs), has ushered in an era of co-clinical trials by which preclinical studies can inform clinical trials and protocols, thus bridging the translational divide in cancer research. Similarly, preclinical imaging fills a translational gap as an enabling technology for translational imaging research. Unlike clinical imaging, where equipment manufacturers strive to meet standards in practice at clinical sites, standards are neither fully developed nor implemented in preclinical imaging. This fundamentally limits the collection and reporting of metadata to qualify preclinical imaging studies, thereby hindering open science and impacting the reproducibility of co-clinical imaging research. To begin to address these issues, the NCI co-clinical imaging research program (CIRP) conducted a survey to identify metadata requirements for reproducible quantitative co-clinical imaging. The enclosed consensus-based report summarizes co-clinical imaging metadata information (CIMI) to support quantitative co-clinical imaging research with broad implications for capturing co-clinical data, enabling interoperability and data sharing, as well as potentially leading to updates to the preclinical Digital Imaging and Communications in Medicine (DICOM) standard.

Information bottleneck-based interpretable multitask network for breast cancer classification and segmentation.

Breast cancer is one of the most common causes of death among women worldwide. Early signs of breast cancer can be an abnormality depicted on breast images (e.g., mammography or breast ultrasonography). However, reliable interpretation of breast images requires intensive labor and physicians with extensive experience. Deep learning is evolving breast imaging diagnosis by introducing a second opinion to physicians. However, most deep learning-based breast cancer analysis algorithms lack interpretability because of their black box nature, which means that domain experts cannot understand why the algorithms predict a label. In addition, most deep learning algorithms are formulated as a single-task-based model that ignores correlations between different tasks (e.g., tumor classification and segmentation). In this paper, we propose an interpretable multitask information bottleneck network (MIB-Net) to accomplish simultaneous breast tumor classification and segmentation. MIB-Net maximizes the mutual information between the latent representations and class labels while minimizing information shared by the latent representations and inputs. In contrast from existing models, our MIB-Net generates a contribution score map that offers an interpretable aid for physicians to understand the model's decision-making process. In addition, MIB-Net implements multitask learning and further proposes a dual prior knowledge guidance strategy to enhance deep task correlation. Our evaluations are carried out on three breast image datasets in different modalities. Our results show that the proposed framework is not only able to help physicians better understand the model's decisions but also improve breast tumor classification and segmentation accuracy over representative state-of-the-art models. Our code is available at https://github.com/jxw0810/MIB-Net.

DWI Metrics Differentiating Benign Intraductal Papillary Mucinous Neoplasms from Invasive Pancreatic Cancer: A Study in GEM Models.

KPC (KrasG12D:Trp53R172H:Pdx1-Cre) and CKS (KrasG12D:Smad4L/L:Ptf1a-Cre) mice are genetically engineered mouse (GEM) models that capture features of human pancreatic ductal adenocarcinoma (PDAC) and intraductal papillary mucinous neoplasms (IPMN), respectively. We compared these autochthonous tumors using quantitative imaging metrics from diffusion-weighted MRI (DW-MRI) and dynamic contrast enhanced (DCE)-MRI in reference to quantitative histological metrics including cell density, fibrosis, and microvasculature density. Our results revealed distinct DW-MRI metrics between the KPC vs. CKS model (mimicking human PDAC vs. IPMN lesion): the apparent diffusion coefficient (ADC) of CKS tumors is significantly higher than that of KPC, with little overlap (mean ± SD 2.24±0.2 vs. 1.66±0.2, p<10−10) despite intratumor and intertumor variability. Kurtosis index (KI) is also distinctively separated in the two models. DW imaging metrics are consistent with growth pattern, cell density, and the cystic nature of the CKS tumors. Coregistration of ex vivo ADC maps with H&E-stained sections allowed for regional comparison and showed a correlation between local cell density and ADC value. In conclusion, studies in GEM models demonstrate the potential utility of diffusion-weighted MRI metrics for distinguishing pancreatic cancer from benign pancreatic cysts such as IPMN.

Transparent Machine Learning Models to Diagnose Suspicious Thoracic Lesions Leveraging CT Guided Biopsy Data.

RATIONALE AND OBJECTIVES: To train and validate machine learning models capable of classifying suspicious thoracic lesions as benign or malignant and to further classify malignant lesions by pathologic subtype while quantifying feature importance for each classification. MATERIALS AND METHODS: 796 patients who had undergone CT guided thoracic biopsy for a concerning thoracic lesion (79.3% lung, 11.4% mediastinum, 6.5% pleura, 2.7% chest wall) were retrospectively enrolled. Lesions were classified as malignant or benign based on ground-truth pathology result, and malignant lesions were classified as primary or secondary cancer. Clinical variables were extracted from EMR and radiology reports. Supervised binary and multiclass classification models were trained to classify lesions based on the input features and evaluated on a held-out test set. Model specific feature analyses were performed to identify variables most predictive of each class, as well as to assess the independent importance of clinical, and imaging features. RESULTS: Binary classification models achieved a top accuracy of 80.6%, with predictive features included smoking history, age, lesion size, and lesion location. Multiclass classification models achieved a top weighted average f1-score of 0.73. Features predictive of primary cancer included smoking history, race, and age, while features predictive of secondary cancer included lesion location, and a history of cancer. CONCLUSION: Machine learning models enable classification of suspicious thoracic lesions based on clinical and imaging variables, achieving clinically useful performance while identifying importance of individual input features on a pathology-proven dataset. We believe models such as these are more likely to be trusted and adopted by clinicians.

Ex vivo MRI and histopathology detect novel iron-rich cortical inflammation in frontotemporal lobar degeneration with tau versus TDP-43 pathology.

Frontotemporal lobar degeneration (FTLD) is a heterogeneous spectrum of age-associated neurodegenerative diseases that include two main pathologic categories of tau (FTLD-Tau) and TDP-43 (FTLD-TDP) proteinopathies. These distinct proteinopathies are often clinically indistinguishable during life, posing a major obstacle for diagnosis and emerging therapeutic trials tailored to disease-specific mechanisms. Moreover, MRI-derived measures have had limited success to date discriminating between FTLD-Tau or FTLD-TDP. T2*-weighted (T2*w) ex vivo MRI has previously been shown to be sensitive to non-heme iron in healthy intracortical lamination and myelin, and to pathological iron deposits in amyloid-beta plaques and activated microglia in Alzheimer's disease neuropathologic change (ADNC). However, an integrated, ex vivo MRI and histopathology approach is understudied in FTLD. We apply joint, whole-hemisphere ex vivo MRI at 7 T and histopathology to the study autopsy-confirmed FTLD-Tau (n = 4) and FTLD-TDP (n = 3), relative to ADNC disease-control brains with antemortem clinical symptoms of frontotemporal dementia (n = 2), and an age-matched healthy control. We detect distinct laminar patterns of novel iron-laden glial pathology in both FTLD-Tau and FTLD-TDP brains. We find iron-positive ameboid and hypertrophic microglia and astrocytes largely in deeper GM and adjacent WM in FTLD-Tau. In contrast, FTLD-TDP presents prominent superficial cortical layer iron reactivity in astrocytic processes enveloping small blood vessels with limited involvement of adjacent WM, as well as more diffuse distribution of punctate iron-rich dystrophic microglial processes across all GM lamina. This integrated MRI/histopathology approach reveals ex vivo MRI features that are consistent with these pathological observations distinguishing FTLD-Tau and FTLD-TDP subtypes, including prominent irregular hypointense signal in deeper cortex in FTLD-Tau whereas FTLD-TDP showed upper cortical layer hypointense bands and diffuse cortical speckling. Moreover, differences in adjacent WM degeneration and iron-rich gliosis on histology between FTLD-Tau and FTLD-TDP were also readily apparent on MRI as hyperintense signal and irregular areas of hypointensity, respectively that were more prominent in FTLD-Tau compared to FTLD-TDP. These unique histopathological and radiographic features were distinct from healthy control and ADNC brains, suggesting that iron-sensitive T2*w MRI, adapted to in vivo application at sufficient resolution, may eventually offer an opportunity to improve antemortem diagnosis of FTLD proteinopathies using tissue-validated methods.

A Precise Method to Evaluate 360 Degree Measures of Optic Cup and Disc Morphology in an African American Cohort and Its Genetic Applications.

(1) Background: Vertical cup-to-disc ratio (CDR) is an important measure for evaluating damage to the optic nerve head (ONH) in glaucoma patients. However, this measure often does not fully capture the irregular cupping observed in glaucomatous nerves. We developed and evaluated a method to measure cup-to-disc ratio (CDR) at all 360 degrees of the ONH. (2) Methods: Non-physician graders from the Scheie Reading Center outlined the cup and disc on digital stereo color disc images from African American patients enrolled in the Primary Open-Angle African American Glaucoma Genetics (POAAGG) study. After converting the resultant coordinates into polar representation, the CDR at each 360-degree location of the ONH was obtained. We compared grader VCDR values with clinical VCDR values, using Spearman correlation analysis, and validated significant genetic associations with clinical VCDR, using grader VCDR values. (3) Results: Graders delineated outlines of the cup contour and disc boundaries twice in each of 1815 stereo disc images. For both cases and controls, the mean CDR was highest at the horizontal bisector, particularly in the temporal region, as compared to other degree locations. There was a good correlation between grader CDR at the vertical bisector and clinical VCDR (Spearman Correlation OD: r = 0.78 [95% CI: 0.76-0.79]). An SNP in the MPDZ gene, associated with clinical VCDR in a prior genome-wide association study, showed a significant association with grader VCDR (p = 0.01) and grader CDR area ratio (p = 0.02). (4) Conclusions: The CDR of both glaucomatous and non-glaucomatous eyes varies by degree location, with the highest measurements in the temporal region of the eye. This method can be useful for capturing innate eccentric ONH morphology, tracking disease progression, and identifying genetic associations.

Image- versus histogram-based considerations in semantic segmentation of pulmonary hyperpolarized gas images.

PURPOSE: To characterize the differences between histogram-based and image-based algorithms for segmentation of hyperpolarized gas lung images. METHODS: Four previously published histogram-based segmentation algorithms (ie, linear binning, hierarchical k-means, fuzzy spatial c-means, and a Gaussian mixture model with a Markov random field prior) and an image-based convolutional neural network were used to segment 2 simulated data sets derived from a public (n = 29 subjects) and a retrospective collection (n = 51 subjects) of hyperpolarized 129Xe gas lung images transformed by common MRI artifacts (noise and nonlinear intensity distortion). The resulting ventilation-based segmentations were used to assess algorithmic performance and characterize optimization domain differences in terms of measurement bias and precision. RESULTS: Although facilitating computational processing and providing discriminating clinically relevant measures of interest, histogram-based segmentation methods discard important contextual spatial information and are consequently less robust in terms of measurement precision in the presence of common MRI artifacts relative to the image-based convolutional neural network. CONCLUSIONS: Direct optimization within the image domain using convolutional neural networks leverages spatial information, which mitigates problematic issues associated with histogram-based approaches and suggests a preferred future research direction. Further, the entire processing and evaluation framework, including the newly reported deep learning functionality, is available as open source through the well-known Advanced Normalization Tools ecosystem.

Diminishing Efficacy of Prone Positioning With Late Application in Evolving Lung Injury.

OBJECTIVES: It is not known how lung injury progression during mechanical ventilation modifies pulmonary responses to prone positioning. We compared the effects of prone positioning on regional lung aeration in late versus early stages of lung injury. DESIGN: Prospective, longitudinal imaging study. SETTING: Research imaging facility at The University of Pennsylvania (Philadelphia, PA) and Medical and Surgical ICUs at Massachusetts General Hospital (Boston, MA). SUBJECTS: Anesthetized swine and patients with acute respiratory distress syndrome (acute respiratory distress syndrome). INTERVENTIONS: Lung injury was induced by bronchial hydrochloric acid (3.5 mL/kg) in 10 ventilated Yorkshire pigs and worsened by supine nonprotective ventilation for 24 hours. Whole-lung CT was performed 2 hours after hydrochloric acid (Day 1) in both prone and supine positions and repeated at 24 hours (Day 2). Prone and supine images were registered (superimposed) in pairs to measure the effects of positioning on the aeration of each tissue unit. Two patients with early acute respiratory distress syndrome were compared with two patients with late acute respiratory distress syndrome, using electrical impedance tomography to measure the effects of body position on regional lung mechanics. MEASUREMENTS AND MAIN RESULTS: Gas exchange and respiratory mechanics worsened over 24 hours, indicating lung injury progression. On Day 1, prone positioning reinflated 18.9% ± 5.2% of lung mass in the posterior lung regions. On Day 2, position-associated dorsal reinflation was reduced to 7.3% ± 1.5% (p < 0.05 vs Day 1). Prone positioning decreased aeration in the anterior lungs on both days. Although prone positioning improved posterior lung compliance in the early acute respiratory distress syndrome patients, it had no effect in late acute respiratory distress syndrome subjects. CONCLUSIONS: The effects of prone positioning on lung aeration may depend on the stage of lung injury and duration of prior ventilation; this may limit the clinical efficacy of this treatment if applied late.

In vivo imaging of canine lung deformation: effects of posture, pneumonectomy, and inhaled erythropoietin.

Mechanical stresses on the lung impose the major stimuli for developmental and compensatory lung growth and remodeling. We used computed tomography (CT) to noninvasively characterize the factors influencing lobar mechanical deformation in relation to posture, pneumonectomy (PNX), and exogenous proangiogenic factor supplementation. Post-PNX adult canines received weekly inhalations of nebulized nanoparticles loaded with recombinant human erythropoietin (EPO) or control (empty nanoparticles) for 16 wk. Supine and prone CT were performed at two transpulmonary pressures pre- and post-PNX following treatment. Lobar air and tissue volumes, fractional tissue volume (FTV), specific compliance (Cs), mechanical strains, and shear distortion were quantified. From supine to prone, lobar volume and Cs increased while strain and shear magnitudes generally decreased. From pre- to post-PNX, air volume increased less and FTV and Cs increased more in the left caudal (LCa) than in other lobes. FTV increased most in the dependent subpleural regions, and the portion of LCa lobe that expanded laterally wrapping around the mediastinum. Supine deformation was nonuniform pre- and post-PNX; strains and shear were most pronounced in LCa lobe and declined when prone. Despite nonuniform regional expansion and deformation, post-PNX lobar mechanics were well preserved compared with pre-PNX because of robust lung growth and remodeling establishing a new mechanical equilibrium. EPO treatment eliminated posture-dependent changes in FTV, accentuated the post-PNX increase in FTV, and reduced FTV heterogeneity without altering absolute air or tissue volumes, consistent with improved microvascular blood volume distribution and modestly enhanced post-PNX alveolar microvascular reserves.NEW & NOTEWORTHY Mechanical stresses on the lung impose the major stimuli for lung growth. We used computed tomography to image deformation of the lung in relation to posture, loss of lung units, and inhalational delivery of the growth promoter erythropoietin. Following loss of one lung in adult large animals, the remaining lung expanded and grew while retaining near-normal mechanical properties. Inhalation of erythropoietin promoted more uniform distribution of blood volume within the remaining lung.

Learning image-based spatial transformations via convolutional neural networks: A review.

Recent methodological innovations in deep learning and associated advancements in computational hardware have significantly impacted the various core subfields of quantitative medical image analysis. The generalizability, computational efficiency and open-source availability of deep learning algorithms and related software, particularly those utilizing convolutional neural networks, have produced paradigm shifts within the field. This impact is evident from topical prevalence in the literature, conference and workshop themes and winning methodologies in relevant competitions. In this work, we review the various state-of-the-art approaches to learning and prediction and/or optimizing image transformations using convolutional neural networks. Although of primary importance within the quantitative imaging domain, image registration algorithmic development, in the context of these deep learning strategies, has received comparatively less attention than its counterparts (e.g., image segmentation). Nevertheless, significant progress has been made in this particular subfield which has been presented in various research venues. We contextualize these contributions within the broader scope of deep learning advancements and, in so doing, attempt to facilitate the leveraging and further development of such techniques within the medical imaging research community.

Deep Learning Applications in Chest Radiography and Computed Tomography: Current State of the Art.

Deep learning is a genre of machine learning that allows computational models to learn representations of data with multiple levels of abstraction using numerous processing layers. A distinctive feature of deep learning, compared with conventional machine learning methods, is that it can generate appropriate models for tasks directly from the raw data, removing the need for human-led feature extraction. Medical images are particularly suited for deep learning applications. Deep learning techniques have already demonstrated high performance in the detection of diabetic retinopathy on fundoscopic images and metastatic breast cancer cells on pathologic images. In radiology, deep learning has the opportunity to provide improved accuracy of image interpretation and diagnosis. Many groups are exploring the possibility of using deep learning-based applications to solve unmet clinical needs. In chest imaging, there has been a large effort to develop and apply computer-aided detection systems for the detection of lung nodules on chest radiographs and chest computed tomography. The essential limitation to computer-aided detection is an inability to learn from new information. To overcome these deficiencies, many groups have turned to deep learning approaches with promising results. In addition to nodule detection, interstitial lung disease recognition, lesion segmentation, diagnosis and patient outcomes have been addressed by deep learning approaches. The purpose of this review article was to cover the current state of the art for deep learning approaches and its limitations, and some of the potential impact on the field of radiology, with specific reference to chest imaging.

Effect of Reconstruction Parameters on the Quantitative Analysis of Chest Computed Tomography.

Quantitative features obtained from computed tomography (CT) scans are being explored for clinical applications. Various classes of quantitative features exist for chest CT including radiomics features, emphysema measurements, lung nodule volumetric measurements, dual energy quantification, and perfusion parameters. A number of research articles have shown promise in diagnosis and prognosis prediction of oncologic patients or those with diffuse lung diseases using these feature classes. Nevertheless, a prerequisite for the quantification is the evaluation of variation in measurements in terms of repeatability and reproducibility, which are distinct aspects of precision but are often not separable from each other. There are well-known sources of measurement variability including patient factors, CT acquisition (scan and reconstruction) factors, and radiologist (or measurement-related) factors. The purpose of this article is to review the effects of CT reconstruction parameters on the quantitative imaging features and efforts to correct or neutralize variations induced by those parameters.

Automated data extraction and ensemble methods for predictive modeling of breast cancer outcomes after radiation therapy.

PURPOSE: The purpose of this study was to compare the effectiveness of ensemble methods (e.g., random forests) and single-model methods (e.g., logistic regression and decision trees) in predictive modeling of post-RT treatment failure and adverse events (AEs) for breast cancer patients using automatically extracted EMR data. METHODS: Data from 1967 consecutive breast radiotherapy (RT) courses at one institution between 2008 and 2015 were automatically extracted from EMRs and oncology information systems using extraction software. Over 230 variables were extracted spanning the following variable segments: patient demographics, medical/surgical history, tumor characteristics, RT treatment history, and AEs tracked using CTCAEv4.0. Treatment failure was extracted algorithmically by searching posttreatment encounters for evidence of local, nodal, or distant failure. Individual models were trained using decision trees, logistic regression, random forests, and boosted decision trees to predict treatment failures and AEs. Models were fit on 75% of the data and evaluated for probability calibration and area under the ROC curve (AUC) on the remaining test set. The impact of each variable segment was assessed by retraining without the segment and measuring change in AUC (ΔAUC). RESULTS: All AUC values were statistically significant (P < 0.05). Ensemble methods outperformed single-model methods across all outcomes. The best ensemble method outperformed decision trees and logistic regression by an average AUC of 0.053 and 0.034, respectively. Model probabilities were well calibrated as evidenced by calibration curves. Excluding the patient medical history variable segment led to the largest AUC reduction in all models (Average ΔAUC = -0.025), followed by RT treatment history (-0.021) and tumor information (-0.015). CONCLUSION: In this largest such study in breast cancer performed to date, automatically extracted EMR data provided a basis for reliable outcome predictions across multiple statistical methods. Ensemble methods provided substantial advantages over single-model methods. Patient medical history contributed the most to prediction quality.

Computational analysis in epilepsy neuroimaging: A survey of features and methods.

Epilepsy affects 65 million people worldwide, a third of whom have seizures that are resistant to anti-epileptic medications. Some of these patients may be amenable to surgical therapy or treatment with implantable devices, but this usually requires delineation of discrete structural or functional lesion(s), which is challenging in a large percentage of these patients. Advances in neuroimaging and machine learning allow semi-automated detection of malformations of cortical development (MCDs), a common cause of drug resistant epilepsy. A frequently asked question in the field is what techniques currently exist to assist radiologists in identifying these lesions, especially subtle forms of MCDs such as focal cortical dysplasia (FCD) Type I and low grade glial tumors. Below we introduce some of the common lesions encountered in patients with epilepsy and the common imaging findings that radiologists look for in these patients. We then review and discuss the computational techniques introduced over the past 10 years for quantifying and automatically detecting these imaging findings. Due to large variations in the accuracy and implementation of these studies, specific techniques are traditionally used at individual centers, often guided by local expertise, as well as selection bias introduced by the varying prevalence of specific patient populations in different epilepsy centers. We discuss the need for a multi-institutional study that combines features from different imaging modalities as well as computational techniques to definitively assess the utility of specific automated approaches to epilepsy imaging. We conclude that sharing and comparing these different computational techniques through a common data platform provides an opportunity to rigorously test and compare the accuracy of these tools across different patient populations and geographical locations. We propose that these kinds of tools, quantitative imaging analysis methods and open data platforms for aggregating and sharing data and algorithms, can play a vital role in reducing the cost of care, the risks of invasive treatments, and improve overall outcomes for patients with epilepsy.

Quantitative assessment of anatomical change using a virtual proton depth radiograph for adaptive head and neck proton therapy.

The aim of this work is to demonstrate the feasibility of using water-equivalent thickness (WET) and virtual proton depth radiographs (PDRs) of intensity corrected cone-beam computed tomography (CBCT) to detect anatomical change and patient setup error to trigger adaptive head and neck proton therapy. The planning CT (pCT) and linear accelerator (linac) equipped CBCTs acquired weekly during treatment of a head and neck patient were used in this study. Deformable image registration (DIR) was used to register each CBCT with the pCT and map Hounsfield units (HUs) from the planning CT (pCT) onto the daily CBCT. The deformed pCT is referred as the corrected CBCT (cCBCT). Two dimensional virtual lateral PDRs were generated using a ray-tracing technique to project the cumulative WET from a virtual source through the cCBCT and the pCT onto a virtual plane. The PDRs were used to identify anatomic regions with large variations in the proton range between the cCBCT and pCT using a threshold of 3 mm relative difference of WET and 3 mm search radius criteria. The relationship between PDR differences and dose distribution is established. Due to weight change and tumor response during treatment, large variations in WETs were observed in the relative PDRs which corresponded spatially with an increase in the number of failing points within the GTV, especially in the pharynx area. Failing points were also evident near the posterior neck due to setup variations. Differences in PDRs correlated spatially to differences in the distal dose distribution in the beam's eye view. Virtual PDRs generated from volumetric data, such as pCTs or CBCTs, are potentially a useful quantitative tool in proton therapy. PDRs and WET analysis may be used to detect anatomical change from baseline during treatment and trigger further analysis in adaptive proton therapy.

The DTI Challenge: Toward Standardized Evaluation of Diffusion Tensor Imaging Tractography for Neurosurgery.

BACKGROUND AND PURPOSE: Diffusion tensor imaging (DTI) tractography reconstruction of white matter pathways can help guide brain tumor resection. However, DTI tracts are complex mathematical objects and the validity of tractography-derived information in clinical settings has yet to be fully established. To address this issue, we initiated the DTI Challenge, an international working group of clinicians and scientists whose goal was to provide standardized evaluation of tractography methods for neurosurgery. The purpose of this empirical study was to evaluate different tractography techniques in the first DTI Challenge workshop. METHODS: Eight international teams from leading institutions reconstructed the pyramidal tract in four neurosurgical cases presenting with a glioma near the motor cortex. Tractography methods included deterministic, probabilistic, filtered, and global approaches. Standardized evaluation of the tracts consisted in the qualitative review of the pyramidal pathways by a panel of neurosurgeons and DTI experts and the quantitative evaluation of the degree of agreement among methods. RESULTS: The evaluation of tractography reconstructions showed a great interalgorithm variability. Although most methods found projections of the pyramidal tract from the medial portion of the motor strip, only a few algorithms could trace the lateral projections from the hand, face, and tongue area. In addition, the structure of disagreement among methods was similar across hemispheres despite the anatomical distortions caused by pathological tissues. CONCLUSIONS: The DTI Challenge provides a benchmark for the standardized evaluation of tractography methods on neurosurgical data. This study suggests that there are still limitations to the clinical use of tractography for neurosurgical decision making.

Plasticity of the human visual system after retinal gene therapy in patients with Leber's congenital amaurosis.

Much of our knowledge of the mechanisms underlying plasticity in the visual cortex in response to visual impairment, vision restoration, and environmental interactions comes from animal studies. We evaluated human brain plasticity in a group of patients with Leber's congenital amaurosis (LCA), who regained vision through gene therapy. Using non-invasive multimodal neuroimaging methods, we demonstrated that reversing blindness with gene therapy promoted long-term structural plasticity in the visual pathways emanating from the treated retina of LCA patients. The data revealed improvements and normalization along the visual fibers corresponding to the site of retinal injection of the gene therapy vector carrying the therapeutic gene in the treated eye compared to the visual pathway for the untreated eye of LCA patients. After gene therapy, the primary visual pathways (for example, geniculostriate fibers) in the treated retina were similar to those of sighted control subjects, whereas the primary visual pathways of the untreated retina continued to deteriorate. Our results suggest that visual experience, enhanced by gene therapy, may be responsible for the reorganization and maturation of synaptic connectivity in the visual pathways of the treated eye in LCA patients. The interactions between the eye and the brain enabled improved and sustained long-term visual function in patients with LCA after gene therapy.

Measuring sparse temporal-variation for accurate registration of dynamic contrast-enhanced breast MR images.

Accurate registration of dynamic contrast-enhanced (DCE) MR breast images is challenging due to the temporal variations of image intensity and the non-rigidity of breast motion. The former can cause the well-known tumor shrinking/expanding problem in registration process while the latter complicates the task by requiring an estimation of non-rigid deformation. In this paper, we treat the intensity's temporal variations as "corruptions" which spatially distribute in a sparse pattern and model them with a L1 norm and a Lorentzian norm. We show that these new image similarity measurements can characterize the non-Gaussian property of the difference between the pre-contrast and post-contrast images and help to resolve the shrinking/expanding problem by forgiving significant image variations. Furthermore, we propose an iteratively re-weighted least squares based method and a linear programming based technique for optimizing the objective functions obtained using these two novel norms. We show that these optimization techniques outperform the traditional gradient-descent approach. Experimental results with sequential DCE-MR images from 28 patients show the superior performances of our algorithms.