MapperPlus: Agnostic clustering of high-dimension data for precision medicine.

One of the goals of precision medicine is to classify patients into subgroups that differ in their susceptibility and response to a disease, thereby enabling tailored treatments for each subgroup. Therefore, there is a great need to identify distinctive clusters of patients from patient data. There are three key challenges to three key challenges of patient stratification: 1) the unknown number of clusters, 2) the need for assessing cluster validity, and 3) the clinical interpretability. We developed MapperPlus, a novel unsupervised clustering pipeline, that directly addresses these challenges. It extends the topological Mapper technique and blends it with two random-walk algorithms to automatically detect disjoint subgroups in patient data. We demonstrate that MapperPlus outperforms traditional agnostic clustering methods in key accuracy/performance metrics by testing its performance on publicly available medical and non-medical data set. We also demonstrate the predictive power of MapperPlus in a medical dataset of pediatric stem cell transplant patients where a number of cluster is unknown. Here, MapperPlus stratifies the patient population into clusters with distinctive survival rates. The MapperPlus software is open-source and publicly available.

Utilization and Cost of an Employer-Sponsored Comprehensive Primary Care Delivery Model.

Importance: Primary care is increasingly delivered at or near workplaces, yet utilization and cost of employer-sponsored primary care services remain unknown. Objective: To compare the health care utilization and cost of an employer-sponsored on-site, near-site, and virtual comprehensive primary care service delivery model with those of traditional community-based primary care. Design, Setting, and Participants: This population-based cohort study of 23 518 commercially insured employees and dependents of an engineering and manufacturing firm headquartered in southern California was performed from January 1, 2016, to July 1, 2019. A subset of the population with most (≥50%) primary care visits through employer-sponsored on-site, near-site, or virtual care clinics was matched to a subset not having most such visits through the employer-sponsored clinics using propensity score matching (n = 1983 each). In sensitivity analyses, employees were matched to dependents at neighboring firms that lacked access to the employer-sponsored primary care delivery model (additional n = 1680). Exposures: Integrated primary care, mental health, and physical therapy delivered through on-site, near-site, and virtual clinics. Main Outcomes and Measures: Utilization (inpatient, outpatient, emergency department, pharmaceutical, radiology, and laboratory visits per 1000 member-months) and spending (2019 costs per member per month in US dollars) by service type. Results: A total of 23 518 individuals (mean [SD] age, 27 [15] years; 14 604 [62.1%] male) were included in the full population sample and had been enrolled in the employer-sponsored health plan for a mean of 29 months (interquartile range, 14-48 months). Of eligible members, 5292 (22.5%) used the employer-sponsored services, with 2305 (9.8%) using them for most of their primary care. The mean (SD) cost of employer-sponsored service delivery was $87 ($32) per member month. Among the matched populations (mean [SD] age, 31 [11] years; 3349 [84.5%] male) of primary users vs control individuals, total spending was 45% lower per member per month (95% CI, 35%-55%; cost difference, -$167 per member per month; 95% CI, -$204 to -$130; P < .001) among users after adjustment. The lower spending was associated with lower spending on non-primary care services, such as emergency department (-33%; 95% CI, -44% to -22%) and hospital visits (-16%; 95% CI, -22% to -10%), despite higher spending on primary care (109%; 95% CI, 102%-116%) and mental health (20%; 95% CI, 13%-27%). Conclusions and Relevance: The findings suggest that individuals who used the models' services for most of their primary care had lower total spending despite higher primary care spending, which may be associated with self-selection of lower-risk persons to the employer-sponsored services and/or with the use of comprehensive primary care.

Mindcontrol: A web application for brain segmentation quality control.

Tissue classification plays a crucial role in the investigation of normal neural development, brain-behavior relationships, and the disease mechanisms of many psychiatric and neurological illnesses. Ensuring the accuracy of tissue classification is important for quality research and, in particular, the translation of imaging biomarkers to clinical practice. Assessment with the human eye is vital to correct various errors inherent to all currently available segmentation algorithms. Manual quality assurance becomes methodologically difficult at a large scale - a problem of increasing importance as the number of data sets is on the rise. To make this process more efficient, we have developed Mindcontrol, an open-source web application for the collaborative quality control of neuroimaging processing outputs. The Mindcontrol platform consists of a dashboard to organize data, descriptive visualizations to explore the data, an imaging viewer, and an in-browser annotation and editing toolbox for data curation and quality control. Mindcontrol is flexible and can be configured for the outputs of any software package in any data organization structure. Example configurations for three large, open-source datasets are presented: the 1000 Functional Connectomes Project (FCP), the Consortium for Reliability and Reproducibility (CoRR), and the Autism Brain Imaging Data Exchange (ABIDE) Collection. These demo applications link descriptive quality control metrics, regional brain volumes, and thickness scalars to a 3D imaging viewer and editing module, resulting in an easy-to-implement quality control protocol that can be scaled for any size and complexity of study.

Gradient nonlinearity effects on upper cervical spinal cord area measurement from 3D T1 -weighted brain MRI acquisitions.

PURPOSE: To explore (i) the variability of upper cervical cord area (UCCA) measurements from volumetric brain 3D T1 -weighted scans related to gradient nonlinearity (GNL) and subject positioning; (ii) the effect of vendor-implemented GNL corrections; and (iii) easily applicable methods that can be used to retrospectively correct data. METHODS: A multiple sclerosis patient was scanned at seven sites using 3T MRI scanners with the same 3D T1 -weighted protocol without GNL-distortion correction. Two healthy subjects and a phantom were additionally scanned at a single site with varying table positions. The 2D and 3D vendor-implemented GNL-correction algorithms and retrospective methods based on (i) phantom data fit, (ii) normalization with C2 vertebral body diameters, and (iii) the Jacobian determinant of nonlinear registrations to a template were tested. RESULTS: Depending on the positioning of the subject, GNL introduced up to 15% variability in UCCA measurements from volumetric brain T1 -weighted scans when no distortion corrections were used. The 3D vendor-implemented correction methods and the three proposed methods reduced this variability to less than 3%. CONCLUSIONS: Our results raise awareness of the significant impact that GNL can have on quantitative UCCA studies, and point the way to prospectively and retrospectively managing GNL distortions in a variety of settings, including clinical environments. Magn Reson Med 79:1595-1601, 2018. © 2017 International Society for Magnetic Resonance in Medicine.

Spinal cord grey matter segmentation challenge.

An important image processing step in spinal cord magnetic resonance imaging is the ability to reliably and accurately segment grey and white matter for tissue specific analysis. There are several semi- or fully-automated segmentation methods for cervical cord cross-sectional area measurement with an excellent performance close or equal to the manual segmentation. However, grey matter segmentation is still challenging due to small cross-sectional size and shape, and active research is being conducted by several groups around the world in this field. Therefore a grey matter spinal cord segmentation challenge was organised to test different capabilities of various methods using the same multi-centre and multi-vendor dataset acquired with distinct 3D gradient-echo sequences. This challenge aimed to characterize the state-of-the-art in the field as well as identifying new opportunities for future improvements. Six different spinal cord grey matter segmentation methods developed independently by various research groups across the world and their performance were compared to manual segmentation outcomes, the present gold-standard. All algorithms provided good overall results for detecting the grey matter butterfly, albeit with variable performance in certain quality-of-segmentation metrics. The data have been made publicly available and the challenge web site remains open to new submissions. No modifications were introduced to any of the presented methods as a result of this challenge for the purposes of this publication.

Gray matter segmentation of the spinal cord with active contours in MR images.

OBJECTIVE: Fully or partially automated spinal cord gray matter segmentation techniques for spinal cord gray matter segmentation will allow for pivotal spinal cord gray matter measurements in the study of various neurological disorders. The objective of this work was multi-fold: (1) to develop a gray matter segmentation technique that uses registration methods with an existing delineation of the cord edge along with Morphological Geodesic Active Contour (MGAC) models; (2) to assess the accuracy and reproducibility of the newly developed technique on 2D PSIR T1 weighted images; (3) to test how the algorithm performs on different resolutions and other contrasts; (4) to demonstrate how the algorithm can be extended to 3D scans; and (5) to show the clinical potential for multiple sclerosis patients. METHODS: The MGAC algorithm was developed using a publicly available implementation of a morphological geodesic active contour model and the spinal cord segmentation tool of the software Jim (Xinapse Systems) for initial estimate of the cord boundary. The MGAC algorithm was demonstrated on 2D PSIR images of the C2/C3 level with two different resolutions, 2D T2* weighted images of the C2/C3 level, and a 3D PSIR image. These images were acquired from 45 healthy controls and 58 multiple sclerosis patients selected for the absence of evident lesions at the C2/C3 level. Accuracy was assessed though visual assessment, Hausdorff distances, and Dice similarity coefficients. Reproducibility was assessed through interclass correlation coefficients. Validity was assessed through comparison of segmented gray matter areas in images with different resolution for both manual and MGAC segmentations. RESULTS: Between MGAC and manual segmentations in healthy controls, the mean Dice similarity coefficient was 0.88 (0.82-0.93) and the mean Hausdorff distance was 0.61 (0.46-0.76) mm. The interclass correlation coefficient from test and retest scans of healthy controls was 0.88. The percent change between the manual segmentations from high and low-resolution images was 25%, while the percent change between the MGAC segmentations from high and low resolution images was 13%. Between MGAC and manual segmentations in MS patients, the average Dice similarity coefficient was 0.86 (0.8-0.92) and the average Hausdorff distance was 0.83 (0.29-1.37) mm. CONCLUSION: We demonstrate that an automatic segmentation technique, based on a morphometric geodesic active contours algorithm, can provide accurate and precise spinal cord gray matter segmentations on 2D PSIR images. We have also shown how this automated technique can potentially be extended to other imaging protocols.

Association of HLA Genetic Risk Burden With Disease Phenotypes in Multiple Sclerosis.

IMPORTANCE: Although multiple HLA alleles associated with multiple sclerosis (MS) risk have been identified, genotype-phenotype studies in the HLA region remain scarce and inconclusive. OBJECTIVES: To investigate whether MS risk-associated HLA alleles also affect disease phenotypes. DESIGN, SETTING, AND PARTICIPANTS: A cross-sectional, case-control study comprising 652 patients with MS who had comprehensive phenotypic information and 455 individuals of European origin serving as controls was conducted at a single academic research site. Patients evaluated at the Multiple Sclerosis Center at University of California, San Francisco between July 2004 and September 2005 were invited to participate. Spinal cord imaging in the data set was acquired between July 2013 and March 2014; analysis was performed between December 2014 and December 2015. MAIN OUTCOMES AND MEASURES: Cumulative HLA genetic burden (HLAGB) calculated using the most updated MS-associated HLA alleles vs clinical and magnetic resonance imaging outcomes, including age at onset, disease severity, conversion time from clinically isolated syndrome to clinically definite MS, fractions of cortical and subcortical gray matter and cerebral white matter, brain lesion volume, spinal cord gray and white matter areas, upper cervical cord area, and the ratio of gray matter to the upper cervical cord area. Multivariate modeling was applied separately for each sex data set. RESULTS: Of the 652 patients with MS, 586 had no missing genetic data and were included in the HLAGB analysis. In these 586 patients (404 women [68.9%]; mean [SD] age at disease onset, 33.6 [9.4] years), HLAGB was higher than in controls (median [IQR], 0.7 [0-1.4] and 0 [-0.3 to 0.5], respectively; P = 1.8 × 10-27). A total of 619 (95.8%) had relapsing-onset MS and 27 (4.2%) had progressive-onset MS. No significant difference was observed between relapsing-onset MS and primary progressive MS. A higher HLAGB was associated with younger age at onset and the atrophy of subcortical gray matter fraction in women with relapsing-onset MS (standard ß = -1.20 × 10-1; P = 1.7 × 10-2 and standard ß = -1.67 × 10-1; P = 2.3 × 10-4, respectively), which were driven mainly by the HLA-DRB1*15:01 haplotype. In addition, we observed the distinct role of the HLA-A*24:02-B*07:02-DRB1*15:01 haplotype among the other common DRB1*15:01 haplotypes and a nominally protective effect of HLA-B*44:02 to the subcortical gray atrophy (standard ß = -1.28 × 10-1; P = 5.1 × 10-3 and standard ß = 9.52 × 10-2; P = 3.6 × 10-2, respectively). CONCLUSIONS AND RELEVANCE: We confirm and extend previous observations linking HLA MS susceptibility alleles with disease progression and specific clinical and magnetic resonance imaging phenotypic traits.

Precision medicine in chronic disease management: The multiple sclerosis BioScreen.

We present a precision medicine application developed for multiple sclerosis (MS): the MS BioScreen. This new tool addresses the challenges of dynamic management of a complex chronic disease; the interaction of clinicians and patients with such a tool illustrates the extent to which translational digital medicine-that is, the application of information technology to medicine-has the potential to radically transform medical practice. We introduce 3 key evolutionary phases in displaying data to health care providers, patients, and researchers: visualization (accessing data), contextualization (understanding the data), and actionable interpretation (real-time use of the data to assist decision making). Together, these form the stepping stones that are expected to accelerate standardization of data across platforms, promote evidence-based medicine, support shared decision making, and ultimately lead to improved outcomes.