Mild Cognitive Impairment Staging Yields Genetic Susceptibility, Biomarker, and Neuroimaging Differences

Introduction

While Alzheimer’s disease (AD) is divided into severity stages, mild cognitive impairment (MCI) remains a solitary construct despite clinical and prognostic heterogeneity. This study aimed to characterize differences in genetic, cerebrospinal fluid (CSF), neuroimaging, and neuropsychological markers across clinician-derived MCI stages.

Methods

Vanderbilt Memory & Aging Project participants with MCI were categorized into 3 severity subtypes at screening based on neuropsychological assessment, functional assessment, and Clinical Dementia Rating interview, including mild (n = 18, 75 ± 8 years), moderate (n = 89 72 ± 7 years), and severe subtypes (n = 18, 78 ± 8 years). At enrollment, participants underwent neuropsychological testing, 3T brain magnetic resonance imaging (MRI), and optional fasting lumbar puncture to obtain CSF. Neuropsychological testing and MRI were repeated at 18-months, 3-years, and 5-years with a mean follow-up time of 3.3 years. Ordinary least square regressions examined cross-sectional associations between MCI severity and apolipoprotein E (APOE)-ε4 status, CSF biomarkers of amyloid beta (Aβ), phosphorylated tau, total tau, and synaptic dysfunction (neurogranin), baseline neuroimaging biomarkers, and baseline neuropsychological performance. Longitudinal associations between baseline MCI severity and neuroimaging and neuropsychological trajectory were assessed using linear mixed effects models with random intercepts and slopes and a follow-up time interaction. Analyses adjusted for baseline age, sex, race/ethnicity, education, and intracranial volume for MRI models.

Results

Stages differed at baseline on APOE-ε4 status (early < middle = late; p-values < 0.03) and CSF Aβ (early > middle = late), phosphorylated and total tau (early = middle < late; p-values < 0.05), and neurogranin concentrations (early = middle < late; p-values < 0.05). MCI stage related to greater longitudinal cognitive decline, hippocampal atrophy, and inferior lateral ventricle dilation (early < late; p-values < 0.03).

Discussion

Clinician staging of MCI severity yielded longitudinal cognitive trajectory and structural neuroimaging differences in regions susceptible to AD neuropathology and neurodegeneration. As expected, participants with more severe MCI symptoms at study entry had greater cognitive decline and gray matter atrophy over time. Differences are likely attributable to baseline differences in amyloidosis, tau, and synaptic dysfunction. MCI staging may provide insight into underlying pathology, prognosis, and therapeutic targets.

A fully automated pipeline for brain structure segmentation in multiple sclerosis

Accurate volume measurements of the brain structures are important for treatment evaluation and disease follow-up in multiple sclerosis (MS) patients. With the aim of obtaining reproducible measurements and avoiding the intra-/inter-rater variability that manual delineations introduce, several automated brain structure segmentation strategies have been proposed in recent years. However, most of these strategies tend to be affected by the abnormal MS lesion intensities, which corrupt the structure segmentation result. To address this problem, we recently reformulated two label fusion strategies of the state of the art, improving their segmentation performance on the lesion areas. Here, we integrate these reformulated strategies in a completely automated pipeline that includes pre-processing (inhomogeneity correction and intensity normalization), atlas selection, masked registration and label fusion, and combine them with an automated lesion segmentation method of the state of the art. We study the effect of automating the lesion mask acquisition on the structure segmentation result, analyzing the output of the proposed pipeline when used in combination with manually and automatically segmented lesion masks. We further analyze the effect of those masks on the segmentation result of the original label fusion strategies when combined with the well-established pre-processing step of lesion filling. The experiments performed show that, when the original methods are used to segment the lesion-filled images, significant structure volume differences are observed in a comparison between manually and automatically segmented lesion masks. The results indicate a mean volume decrease of 1.13%±1.93 in the cerebrospinal fluid, and a mean volume increase of 0.13%±0.14 and 0.05%±0.08 in the cerebral white matter and cerebellar gray matter, respectively. On the other hand, no significant volume differences were found when the proposed automated pipeline was used for segmentation, which demonstrates its robustness against variations in the lesion mask used.

Free-water metrics in medial temporal lobe white matter tract projections relate to longitudinal cognitive decline

Although hippocampal volume has served as a long-standing predictor of cognitive decline, diffusion magnetic resonance imaging studies of white matter have shown similar relationships. Still, it remains unclear if gray matter and white matter interact to predict cognitive impairment and longitudinal decline. Here, we investigate whether free-water (FW) and FW-corrected fractional anisotropy (FA_T) within medial temporal lobe white matter tracts provides meaningful contribution to cognition and cognitive decline beyond hippocampal volume. Using data from the Vanderbilt Memory & Aging Project (n = 319), we found that FW was associated with baseline memory and executive function beyond that of hippocampal volume and other comorbidities. Longitudinal analyses demonstrated significant interactions of hippocampal volume and inferior longitudinal fasciculus (p = 0.043) and cingulum bundle (p = 0.025) FA_T on memory decline and with fornix FA_T (p = 0.025) on decline in executive function. Results suggest that FW metrics of white matter have a unique role in cognitive decline and should be included in theoretical models of aging, cerebrovascular disease, and Alzheimer's disease.

Cerebrospinal fluid biomarkers of neurodegeneration, synaptic dysfunction, and axonal injury relate to atrophy in structural brain regions specific to Alzheimer's disease

INTRODUCTION

Patterns of atrophy can distinguish normal cognition from Alzheimer's disease (AD), but neuropathological drivers of this pattern are unknown. This study examined associations between cerebrospinal fluid biomarkers of AD pathology, synaptic dysfunction, and neuroaxonal injury with two AD imaging signatures.

METHODS

Signatures were calculated using published guidelines. Linear regressions related each biomarker to both signatures, adjusting for demographic factors. Bootstrapped analyses tested if associations were stronger with one signature versus the other.

RESULTS

Increased phosphorylated tau (p-tau), total tau, and neurofilament light (P-values <.045) related to smaller signatures (indicating greater atrophy). Diagnosis and sex modified associations between p-tau and neurogranin (P-values<.05) and signatures, such that associations were stronger among participants with mild cognitive impairment and female participants. The strength of associations did not differ between signatures.

DISCUSSION

Increased evidence of neurodegeneration, axonopathy, and tau phosphorylation relate to greater AD-related atrophy. Tau phosphorylation and synaptic dysfunction may be more prominent in AD-affected regions in females.

Structural Correlates of the Sensorimotor Cerebellum in Parkinson's Disease and Essential Tremor

BACKGROUND

Parkinson's disease (PD) and essential tremor (ET) are commonly encountered movement disorders. Pathophysiologic processes that localize to the cerebellum are described in both. There are limited studies investigating cerebellar structural changes in these conditions, largely because of inherent challenges in the efficiency of segmentation.

METHODS

We applied a novel multiatlas cerebellar segmentation method to T1-weighted images in 282 PD and 111 essential tremor patients to define 26 cerebellar lobule volumes. The severity of postural and resting tremor in both populations and gait and postural instability in PD patients were defined using subscores of the UPDRS and Washington Heights-Inwood Genetic Study motor scales. These clinical measurements were related to lobule volume size. Multiple comparisons were controlled using a false discovery rate method.

RESULTS

Group differences were identified between ET and PD patients, with reductions in deep cerebellar nucleus volume in ET versus reduced lobule VI volume in PD. In ET patients, lobule VIII was negatively correlated with the severity of postural tremor. In PD patients, lobule IV was positively correlated with resting tremor and total tremor severity. We observed differences in cerebellar structure that localized to sensorimotor lobules of the cerebellum. Lobule volumes appeared to differentially relate to clinical symptoms, suggesting important clinicopathologic distinctions between these conditions. These results emphasize the role of the cerebellum in tremor symptoms and should foster future clinical and pathologic investigations of the sensorimotor lobules of the cerebellum. © 2020 International Parkinson and Movement Disorder Society.

AUTOMATIC LABELING OF CORTICAL SULCI USING SPHERICAL CONVOLUTIONAL NEURAL NETWORKS IN A DEVELOPMENTAL COHORT

In this paper, we present the automatic labeling framework for sulci in the human lateral prefrontal cortex (PFC). We adapt an existing spherical U-Net architecture with our recent surface data augmentation technique to improve the sulcal labeling accuracy in a developmental cohort. Specifically, our framework consists of the following key components: (1) augmented geometrical features being generated during cortical surface registration, (2) spherical U-Net architecture to efficiently fit the augmented features, and (3) postrefinement of sulcal labeling by optimizing spatial coherence via a graph cut technique. We validate our method on 30 healthy subjects with manual labeling of sulcal regions within PFC. In the experiments, we demonstrate significantly improved labeling performance (0.7749) in mean Dice overlap compared to that of multi-atlas (0.6410) and standard spherical U-Net (0.7011) approaches, respectively (p < 0.05). Additionally, the proposed method achieves a full set of sulcal labels in 20 seconds in this developmental cohort.

Validation and Optimization of Multi-Organ Segmentation on Clinical Imaging Archives

Segmentation of abdominal computed tomography (CT) provides spatial context, morphological properties, and a framework for tissue-specific radiomics to guide quantitative Radiological assessment. A 2015 MICCAI challenge spurred substantial innovation in multi-organ abdominal CT segmentation with both traditional and deep learning methods. Recent innovations in deep methods have driven performance toward levels for which clinical translation is appealing. However, continued cross-validation on open datasets presents the risk of indirect knowledge contamination and could result in circular reasoning. Moreover, "real world" segmentations can be challenging due to the wide variability of abdomen physiology within patients. Herein, we perform two data retrievals to capture clinically acquired deidentified abdominal CT cohorts with respect to a recently published variation on 3D U-Net (baseline algorithm). First, we retrieved 2004 deidentified studies on 476 patients with diagnosis codes involving spleen abnormalities (cohort A). Second, we retrieved 4313 deidentified studies on 1754 patients without diagnosis codes involving spleen abnormalities (cohort B). We perform prospective evaluation of the existing algorithm on both cohorts, yielding 13% and 8% failure rate, respectively. Then, we identified 51 subjects in cohort A with segmentation failures and manually corrected the liver and gallbladder labels. We re-trained the model adding the manual labels, resulting in performance improvement of 9% and 6% failure rate for the A and B cohorts, respectively. In summary, the performance of the baseline on the prospective cohorts was similar to that on previously published datasets. Moreover, adding data from the first cohort substantively improved performance when evaluated on the second withheld validation cohort.

Deep Multi-task Prediction of Lung Cancer and Cancer-free Progression from Censored Heterogenous Clinical Imaging

Annual low dose computed tomography (CT) lung screening is currently advised for individuals at high risk of lung cancer (e.g., heavy smokers between 55 and 80 years old). The recommended screening practice significantly reduces all-cause mortality, but the vast majority of screening results are negative for cancer. If patients at very low risk could be identified based on individualized, image-based biomarkers, the health care resources could be more efficiently allocated to higher risk patients and reduce overall exposure to ionizing radiation. In this work, we propose a multi-task (diagnosis and prognosis) deep convolutional neural network to improve the diagnostic accuracy over a baseline model while simultaneously estimating a personalized cancer-free progression time (CFPT). A novel Censored Regression Loss (CRL) is proposed to perform weakly supervised regression so that even single negative screening scans can provide small incremental value. Herein, we study 2287 scans from 1433 de-identified patients from the Vanderbilt Lung Screening Program (VLSP) and Molecular Characterization Laboratories (MCL) cohorts. Using five-fold cross-validation, we train a 3D attention-based network under two scenarios: (1) single-task learning with only classification, and (2) multi-task learning with both classification and regression. The single-task learning leads to a higher AUC compared with the Kaggle challenge winner pre-trained model (0.878 v. 0.856), and multi-task learning significantly improves the single-task one (AUC 0.895, p<0.01, McNemar test). In summary, the image-based predicted CFPT can be used in follow-up year lung cancer prediction and data assessment.

MRI correlates of chronic symptoms in mild traumatic brain injury

Some veterans with a history of mild traumatic brain injury (mTBI) have reported experiencing auditory and visual dysfunction that persist beyond the acute phase of the incident. The etiology behind these symptoms is difficult to characterize, since mTBI is defined by negative imaging findings on current clinical imaging. There are several competing hypotheses that could explain functional deficits; one example is shear injury, which may manifest in diffusion-weighted magnetic resonance (MR) imaging (DWI). Herein, we explore this alternative hypothesis in a pilot study of multi-parametric MR imaging. Briefly, we consider a cohort of 8 mTBI patients relative to 22 control subjects using structural T1-weighted imaging (T1w) and connectivity with DWI. 1,344 metrics were extracted per subject from whole brain regions and connectivity patterns in sensory networks. For each set of imaging-derived metrics, the control subject metrics were embedded in a low-dimensional manifold with principal component analysis, after which mTBI subject metrics were projected into the same space. These manifolds were employed to train support vector machines (SVM) to classify subjects as controls or mTBI. Two of the SVMs trained achieved near-perfect accuracy averaged across four-fold cross-validation. Additionally, we present correlations between manifold dimensions and 22 self-reported mTBI symptoms and find that five principal components from the manifolds (one component from the T1w manifold and four components from the DWI manifold) are significantly correlated with symptoms (p<0.05, uncorrected). The novelty of this work is that the DWI and T1w imaging metrics seem to contain information critical for distinguishing between mTBI and control subjects. This work presents an analysis of the pilot phase of data collection of the Quantitative Evaluation of Visual and Auditory Dysfunction and Multi-Sensory Integration in Complex TBI Patients study and defines specific hypotheses to be tested in the full sample.

Extracting 2D weak labels from volume labels using multiple instance learning in CT hemorrhage detection

Multiple instance learning (MIL) is a supervised learning methodology that aims to allow models to learn instance class labels from bag class labels, where a bag is defined to contain multiple instances. MIL is gaining traction for learning from weak labels but has not been widely applied to 3D medical imaging. MIL is well-suited to clinical CT acquisitions since (1) the highly anisotropic voxels hinder application of traditional 3D networks and (2) patch-based networks have limited ability to learn whole volume labels. In this work, we apply MIL with a deep convolutional neural network to identify whether clinical CT head image volumes possess one or more large hemorrhages (> 20cm3), resulting in a learned 2D model without the need for 2D slice annotations. Individual image volumes are considered separate bags, and the slices in each volume are instances. Such a framework sets the stage for incorporating information obtained in clinical reports to help train a 2D segmentation approach. Within this context, we evaluate the data requirements to enable generalization of MIL by varying the amount of training data. Our results show that a training size of at least 400 patient image volumes was needed to achieve accurate per-slice hemorrhage detection. Over a five-fold cross-validation, the leading model, which made use of the maximum number of training volumes, had an average true positive rate of 98.10%, an average true negative rate of 99.36%, and an average precision of 0.9698. The models have been made available along with source code1 to enabled continued exploration and adaption of MIL in CT neuroimaging.

Generalizing Deep Whole Brain Segmentation for Pediatric and Post- Contrast MRI with Augmented Transfer Learning

Generalizability is an important problem in deep neural networks, especially in the context of the variability of data acquisition in clinical magnetic resonance imaging (MRI). Recently, the Spatially Localized Atlas Network Tiles (SLANT) approach has been shown to effectively segment whole brain non-contrast T1w MRI with 132 volumetric labels. Enhancing generalizability of SLANT would enable broader application of volumetric assessment in multi-site studies. Transfer learning (TL) is commonly to update neural network weights for local factors; yet, it is commonly recognized to risk degradation of performance on the original validation/test cohorts. Here, we explore TL by data augmentation to address these concerns in the context of adapting SLANT to anatomical variation (e.g., adults versus children) and scanning protocol (e.g., non-contrast research T1w MRI versus contrast-enhanced clinical T1w MRI). We consider two datasets: First, 30 T1w MRI of young children with manually corrected volumetric labels, and accuracy of automated segmentation defined relative to the manually provided truth. Second, 36 paired datasets of pre- and post-contrast clinically acquired T1w MRI, and accuracy of the post-contrast segmentations assessed relative to the pre-contrast automated assessment. For both studies, we augment the original TL step of SLANT with either only the new data or with both original and new data. Over baseline SLANT, both approaches yielded significantly improved performance (pediatric: 0.89 vs. 0.82 DSC, p<0.001; contrast: 0.80 vs 0.76, p<0.001). The performance on the original test set decreased with the new-data only transfer learning approach, so data augmentation was superior to strict transfer learning.

Internal-transfer Weighting of Multi-task Learning for Lung Cancer Detection

Deep learning has achieved many successes in medical imaging, including lung nodule segmentation and lung cancer prediction on computed tomography (CT). Recently, multi-task networks have shown to both offer additional estimation capabilities, and, perhaps more importantly, increased performance over single-task networks on a "main/primary" task. However, balancing the optimization criteria of multi-task networks across different tasks is an area of active exploration. Here, we extend a previously proposed 3D attention-based network with four additional multi-task subnetworks for the detection of lung cancer and four auxiliary tasks (diagnosis of asthma, chronic bronchitis, chronic obstructive pulmonary disease, and emphysema). We introduce and evaluate a learning policy, Periodic Focusing Learning Policy (PFLP), that alternates the dominance of tasks throughout the training. To improve performance on the primary task, we propose an Internal-Transfer Weighting (ITW) strategy to suppress the loss functions on auxiliary tasks for the final stages of training. To evaluate this approach, we examined 3386 patients (single scan per patient) from the National Lung Screening Trial (NLST) and de-identified data from the Vanderbilt Lung Screening Program, with a 2517/277/592 (scans) split for training, validation, and testing. Baseline networks include a single-task strategy and a multi-task strategy without adaptive weights (PFLP/ITW), while primary experiments are multi-task trials with either PFLP or ITW or both. On the test set for lung cancer prediction, the baseline single-task network achieved prediction AUC of 0.8080 and multi-task baseline failed to converge (AUC 0.6720). However, applying PFLP helped multi-task network clarify and achieved test set lung cancer prediction AUC of 0.8402. Furthermore, our ITW technique boosted the PFLP enabled multi-task network and achieved an AUC of 0.8462 (McNemar test, p < 0.01). In conclusion, adaptive consideration of multi-task learning weights is important, and PFLP and ITW are promising strategies.

Contrast Phase Classification with a Generative Adversarial Network

Dynamic contrast enhanced computed tomography (CT) is an imaging technique that provides critical information on the relationship of vascular structure and dynamics in the context of underlying anatomy. A key challenge for image processing with contrast enhanced CT is that phase discrepancies are latent in different tissues due to contrast protocols, vascular dynamics, and metabolism variance. Previous studies with deep learning frameworks have been proposed for classifying contrast enhancement with networks inspired by computer vision. Here, we revisit the challenge in the context of whole abdomen contrast enhanced CTs. To capture and compensate for the complex contrast changes, we propose a novel discriminator in the form of a multi-domain disentangled representation learning network. The goal of this network is to learn an intermediate representation that separates contrast enhancement from anatomy and enables classification of images with varying contrast time. Briefly, our unpaired contrast disentangling GAN(CD-GAN) Discriminator follows the ResNet architecture to classify a CT scan from different enhancement phases. To evaluate the approach, we trained the enhancement phase classifier on 21060 slices from two clinical cohorts of 230 subjects. The scans were manually labeled with three independent enhancement phases (non-contrast, portal venous and delayed). Testing was performed on 9100 slices from 30 independent subjects who had been imaged with CT scans from all contrast phases. Performance was quantified in terms of the multi-class normalized confusion matrix. The proposed network significantly improved correspondence over baseline UNet, ResNet50 and StarGAN's performance of accuracy scores 0.54. 0.55, 0.62 and 0.91, respectively (p-value<0.0001 paired t-test for ResNet versus CD-GAN). The proposed discriminator from the disentangled network presents a promising technique that may allow deeper modeling of dynamic imaging against patient specific anatomies.

Deep Learning Estimation of Multi-Tissue Constrained Spherical Deconvolution with Limited Single Shell DW-MRI

Diffusion-weighted magnetic resonance imaging (DW-MRI) is the only non-invasive approach for estimation of intra-voxel tissue microarchitecture and reconstruction of in vivo neural pathways for the human brain. With improvement in accelerated MRI acquisition technologies, DW-MRI protocols that make use of multiple levels of diffusion sensitization have gained popularity. A well-known advanced method for reconstruction of white matter microstructure that uses multi-shell data is multi-tissue constrained spherical deconvolution (MT-CSD). MT-CSD substantially improves the resolution of intra-voxel structure over the traditional single shell version, constrained spherical deconvolution (CSD). Herein, we explore the possibility of using deep learning on single shell data (using the b=1000 s/mm² from the Human Connectome Project (HCP)) to estimate the information content captured by 8^th order MT-CSD using the full three shell data (b=1000, 2000, and 3000 s/mm² from HCP). Briefly, we examine two network architectures: 1.) Sequential network of fully connected dense layers with a residual block in the middle (ResDNN), 2.) Patch based convolutional neural network with a residual block (ResCNN). For both networks an additional output block for estimation of voxel fraction was used with a modified loss function. Each approach was compared against the baseline of using MT-CSD on all data on 15 subjects from the HCP divided into 5 training, 2 validation, and 8 testing subjects with a total of 6.7 million voxels. The fiber orientation distribution function (fODF) can be recovered with high correlation (0.77 vs 0.74 and 0.65) and low root mean squared error ResCNN:0.0124, ResDNN:0.0168 and sCSD:0.0323 as compared to the ground truth of MT-CST, which was derived from the multi-shell DW-MRI acquisitions. The mean squared error between the MT-CSD estimates for white matter tissue fraction and for the predictions are ResCNN:0.0249 vs ResDNN:0.0264. We illustrate the applicability of high definition fiber tractography on a single testing subject with arcuate and corpus callosum Tractography. In summary, the proposed approach provides a promising framework to estimate MT-CSD with limited single shell data. Source code and models have been made publicly available.

Semi-Supervised Multi-Organ Segmentation through Quality Assurance Supervision

Human in-the-loop quality assurance (QA) is typically performed after medical image segmentation to ensure that the systems are performing as intended, as well as identifying and excluding outliers. By performing QA on large-scale, previously unlabeled testing data, categorical QA scores (e.g. "successful" versus "unsuccessful") can be generated. Unfortunately, the precious use of resources for human in-the-loop QA scores are not typically reused in medical image machine learning, especially to train a deep neural network for image segmentation. Herein, we perform a pilot study to investigate if the QA labels can be used as supplementary supervision to augment the training process in a semi-supervised fashion. In this paper, we propose a semi-supervised multi-organ segmentation deep neural network consisting of a traditional segmentation model generator and a QA involved discriminator. An existing 3-D abdominal segmentation network is employed, while the pre-trained ResNet-18 network is used as discriminator. A large-scale dataset of 2027 volumes are used to train the generator, whose 2-D montage images and segmentation mask with QA scores are used to train the discriminator. To generate the QA scores, the 2-D montage images were reviewed manually and coded 0 (success), 1 (errors consistent with published performance), and 2 (gross failure). Then, the ResNet-18 network was trained with 1623 montage images in equal distribution of all three code labels and achieved an accuracy 94% for classification predictions with 404 montage images withheld for the test cohort. To assess the performance of using the QA supervision, the discriminator was used as a loss function in a multi-organ segmentation pipeline. The inclusion of QA-loss function boosted performance on the unlabeled test dataset from 714 patients to 951 patients over the baseline model. Additionally, the number of failures decreased from 606 (29.90%) to 402 (19.83%). The contributions of the proposed method are three-fold: We show that (1) the QA scores can be used as a loss function to perform semi-supervised learning for unlabeled data, (2) the well trained discriminator is learnt by QA score rather than traditional "true/false", and (3) the performance of multi-organ segmentation on unlabeled datasets can be fine-tuned with more robust and higher accuracy than the original baseline method. The use of QA-inspired loss functions represents a promising area of future research and may permit tighter integration of supervised and semi-supervised learning.

Tractography reproducibility challenge with empirical data (TraCED): The 2017 ISMRM diffusion study group challenge

BACKGROUND

Fiber tracking with diffusion-weighted MRI has become an essential tool for estimating in vivo brain white matter architecture. Fiber tracking results are sensitive to the choice of processing method and tracking criteria.

PURPOSE

To assess the variability for an algorithm in group studies reproducibility is of critical context. However, reproducibility does not assess the validity of the brain connections. Phantom studies provide concrete quantitative comparisons of methods relative to absolute ground truths, yet do no capture variabilities because of in vivo physiological factors. The ISMRM 2017 TraCED challenge was created to fulfill the gap.

STUDY TYPE

A systematic review of algorithms and tract reproducibility studies.

SUBJECTS

Single healthy volunteers.

FIELD STRENGTH/SEQUENCE

3.0T, two different scanners by the same manufacturer. The multishell acquisition included b-values of 1000, 2000, and 3000 s/mm2 with 20, 45, and 64 diffusion gradient directions per shell, respectively.

ASSESSMENT

Nine international groups submitted 46 tractography algorithm entries each consisting 16 tracts per scan. The algorithms were assessed using intraclass correlation (ICC) and the Dice similarity measure.

STATISTICAL TESTS

Containment analysis was performed to assess if the submitted algorithms had containment within tracts of larger volume submissions. This also serves the purpose to detect if spurious submissions had been made.

RESULTS

The top five submissions had high ICC and Dice >0.88. Reproducibility was high within the top five submissions when assessed across sessions or across scanners: 0.87-0.97. Containment analysis shows that the top five submissions are contained within larger volume submissions. From the total of 16 tracts as an outcome relatively the number of tracts with high, moderate, and low reproducibility were 8, 4, and 4.

DATA CONCLUSION

The different methods clearly result in fundamentally different tract structures at the more conservative specificity choices. Data and challenge infrastructure remain available for continued analysis and provide a platform for comparison.

LEVEL OF EVIDENCE

5 Technical Efficacy Stage: 1 J. Magn. Reson. Imaging 2020;51:234-249.

Distributed deep learning across multisite datasets for generalized CT hemorrhage segmentation

PURPOSE

As deep neural networks achieve more success in the wide field of computer vision, greater emphasis is being placed on the generalizations of these models for production deployment. With sufficiently large training datasets, models can typically avoid overfitting their data; however, for medical imaging it is often difficult to obtain enough data from a single site. Sharing data between institutions is also frequently nonviable or prohibited due to security measures and research compliance constraints, enforced to guard protected health information (PHI) and patient anonymity.

METHODS

In this paper, we implement cyclic weight transfer with independent datasets from multiple geographically disparate sites without compromising PHI. We compare results between single-site learning (SSL) and multisite learning (MSL) models on testing data drawn from each of the training sites as well as two other institutions.

RESULTS

The MSL model attains an average dice similarity coefficient (DSC) of 0.690 on the holdout institution datasets with a volume correlation of 0.914, respectively corresponding to a 7% and 5% statistically significant improvement over the average of both SSL models, which attained an average DSC of 0.646 and average correlation of 0.871.

CONCLUSIONS

We show that a neural network can be efficiently trained on data from two physically remote sites without consolidating patient data to a single location. The resulting network improves model generalization and achieves higher average DSCs on external datasets than neural networks trained on data from a single source.

Outlier Guided Optimization of Abdominal Segmentation

Abdominal multi-organ segmentation of computed tomography (CT) images has been the subject of extensive research interest. It presents a substantial challenge in medical image processing, as the shape and distribution of abdominal organs can vary greatly among the population and within an individual over time. While continuous integration of novel datasets into the training set provides potential for better segmentation performance, collection of data at scale is not only costly, but also impractical in some contexts. Moreover, it remains unclear what marginal value additional data have to offer. Herein, we propose a single-pass active learning method through human quality assurance (QA). We built on a pre-trained 3D U-Net model for abdominal multi-organ segmentation and augmented the dataset either with outlier data (e.g., exemplars for which the baseline algorithm failed) or inliers (e.g., exemplars for which the baseline algorithm worked). The new models were trained using the augmented datasets with 5-fold cross-validation (for outlier data) and withheld outlier samples (for inlier data). Manual labeling of outliers increased Dice scores with outliers by 0.130, compared to an increase of 0.067 with inliers (p<0.001, two-tailed paired t-test). By adding 5 to 37 inliers or outliers to training, we find that the marginal value of adding outliers is higher than that of adding inliers. In summary, improvement on single-organ performance was obtained without diminishing multi-organ performance or significantly increasing training time. Hence, identification and correction of baseline failure cases present an effective and efficient method of selecting training data to improve algorithm performance.

Using deep learning for a diffusion-based segmentation of the dentate nucleus and its benefits over atlas-based methods

The dentate nucleus (DN) is a gray matter structure deep in the cerebellum involved in motor coordination, sensory input integration, executive planning, language, and visuospatial function. The DN is an emerging biomarker of disease, informing studies that advance pathophysiologic understanding of neurodegenerative and related disorders. The main challenge in defining the DN radiologically is that, like many deep gray matter structures, it has poor contrast in T1-weighted magnetic resonance (MR) images and therefore requires specialized MR acquisitions for visualization. Manual tracing of the DN across multiple acquisitions is resource-intensive and does not scale well to large datasets. We describe a technique that automatically segments the DN using deep learning (DL) on common imaging sequences, such as T1-weighted, T2-weighted, and diffusion MR imaging. We trained a DL algorithm that can automatically delineate the DN and provide an estimate of its volume. The automatic segmentation achieved higher agreement to the manual labels compared to template registration, which is the current common practice in DN segmentation or multiatlas segmentation of manual labels. Across all sequences, the FA maps achieved the highest mean Dice similarity coefficient (DSC) of 0.83 compared to T1 imaging ( DSC = 0.76 ), T2 imaging ( DSC = 0.79 ), or a multisequence approach ( DSC = 0.80 ). A single atlas registration approach using the spatially unbiased atlas template of the cerebellum and brainstem template achieved a DSC of 0.23, and multi-atlas segmentation achieved a DSC of 0.33. Overall, we propose a method of delineating the DN on clinical imaging that can reproduce manual labels with higher accuracy than current atlas-based tools.

Disruption of Neural Homeostasis as a Model of Relapse and Recurrence in Late-Life Depression

The significant public health burden associated with late-life depression (LLD) is magnified by the high rates of recurrence. In this manuscript, we review what is known about recurrence risk factors, conceptualize recurrence within a model of homeostatic disequilibrium, and discuss the potential significance and challenges of new research into LLD recurrence. The proposed model is anchored in the allostatic load theory of stress. We review the allostatic response characterized by neural changes in network function and connectivity and physiologic changes in the hypothalamic-pituitary-adrenal axis, autonomic nervous system, immune system, and circadian rhythm. We discuss the role of neural networks' instability following treatment response as a source of downstream disequilibrium, triggering and/or amplifying abnormal stress response, cognitive dysfunction and behavioral changes, ultimately precipitating a full-blown recurrent episode of depression. We propose strategies to identify and capture early change points that signal recurrence risk through mobile technology to collect ecologically measured symptoms, accompanied by automated algorithms that monitor for state shifts (persistent worsening) and variance shifts (increased variability) relative to a patient's baseline. Identifying such change points in relevant sensor data could potentially provide an automated tool that could alert clinicians to at-risk individuals or relevant symptom changes even in a large practice.

Discovering novel disease comorbidities using electronic medical records

Increasing reliance on electronic medical records at large medical centers provides unique opportunities to perform population level analyses exploring disease progression and etiology. The massive accumulation of diagnostic, procedure, and laboratory codes in one place has enabled the exploration of co-occurring conditions, their risk factors, and potential prognostic factors. While most of the readily identifiable associations in medical records are (now) well known to the scientific community, there is no doubt many more relationships are still to be uncovered in EMR data. In this paper, we introduce a novel finding index to help with that task. This new index uses data mined from real-time PubMed abstracts to indicate the extent to which empirically discovered associations are already known (i.e., present in the scientific literature). Our methods leverage second-generation p-values, which better identify associations that are truly clinically meaningful. We illustrate our new method with three examples: Autism Spectrum Disorder, Alzheimer’s Disease, and Optic Neuritis. Our results demonstrate wide utility for identifying new associations in EMR data that have the highest priority among the complex web of correlations and causalities. Data scientists and clinicians can work together more effectively to discover novel associations that are both empirically reliable and clinically understudied.

Montage based 3D Medical Image Retrieval from Traumatic Brain Injury Cohort using Deep Convolutional Neural Network

Brain imaging analysis on clinically acquired computed tomography (CT) is essential for the diagnosis, risk prediction of progression, and treatment of the structural phenotypes of traumatic brain injury (TBI). However, in real clinical imaging scenarios, entire body CT images (e.g., neck, abdomen, chest, pelvis) are typically captured along with whole brain CT scans. For instance, in a typical sample of clinical TBI imaging cohort, only ~15% of CT scans actually contain whole brain CT images suitable for volumetric brain analyses; the remaining are partial brain or non-brain images. Therefore, a manual image retrieval process is typically required to isolate the whole brain CT scans from the entire cohort. However, the manual image retrieval is time and resource consuming and even more difficult for the larger cohorts. To alleviate the manual efforts, in this paper we propose an automated 3D medical image retrieval pipeline, called deep montage-based image retrieval (dMIR), which performs classification on 2D montage images via a deep convolutional neural network. The novelty of the proposed method for image processing is to characterize the medical image retrieval task based on the montage images. In a cohort of 2000 clinically acquired TBI scans, 794 scans were used as training data, 206 scans were used as validation data, and the remaining 1000 scans were used as testing data. The proposed achieved accuracy=1.0, recall=1.0, precision=1.0, f1=1.0 for validation data, while achieved accuracy=0.988, recall=0.962, precision=0.962, f1=0.962 for testing data. Thus, the proposed dMIR is able to perform accurate CT whole brain image retrieval from large-scale clinical cohorts.

Quantitative Spatial Analysis of Metabolic Heterogeneity Across in vivo and in vitro Tumor Models

Metabolic preferences of tumor cells vary within a single tumor, contributing to tumor heterogeneity, drug resistance, and patient relapse. However, the relationship between tumor treatment response and metabolically distinct tumor cell populations is not well-understood. Here, a quantitative approach was developed to characterize spatial patterns of metabolic heterogeneity in tumor cell populations within in vivo xenografts and 3D in vitro cultures (i.e., organoids) of head and neck cancer. Label-free images of cell metabolism were acquired using two-photon fluorescence lifetime microscopy of the metabolic co-enzymes NAD(P)H and FAD. Previous studies have shown that NAD(P)H mean fluorescence lifetimes can identify metabolically distinct cells with varying drug response. Thus, density-based clustering of the NAD(P)H mean fluorescence lifetime was used to identify metabolic sub-populations of cells, then assessed in control, cetuximab-, cisplatin-, and combination-treated xenografts 13 days post-treatment and organoids 24 h post-treatment. Proximity analysis of these metabolically distinct cells was designed to quantify differences in spatial patterns between treatment groups and between xenografts and organoids. Multivariate spatial autocorrelation and principal components analyses of all autofluorescence intensity and lifetime variables were developed to further improve separation between cell sub-populations. Spatial principal components analysis and Z-score calculations of autofluorescence and spatial distribution variables also visualized differences between models. This analysis captures spatial distributions of tumor cell sub-populations influenced by treatment conditions and model-specific environments. Overall, this novel spatial analysis could provide new insights into tumor growth, treatment resistance, and more effective drug treatments across a range of microscopic imaging modalities (e.g., immunofluorescence, imaging mass spectrometry).

Cortical Surface Parcellation using Spherical Convolutional Neural Networks

We present cortical surface parcellation using spherical deep convolutional neural networks. Traditional multi-atlas cortical surface parcellation requires inter-subject surface registration using geometric features with slow processing speed on a single subject (2-3 hours). Moreover, even optimal surface registration does not necessarily produce optimal cortical parcellation as parcel boundaries are not fully matched to the geometric features. In this context, a choice of training features is important for accurate cortical parcellation. To utilize the networks efficiently, we propose cortical parcellation-specific input data from an irregular and complicated structure of cortical surfaces. To this end, we align ground-truth cortical parcel boundaries and use their resulting deformation fields to generate new pairs of deformed geometric features and parcellation maps. To extend the capability of the networks, we then smoothly morph cortical geometric features and parcellation maps using the intermediate deformation fields. We validate our method on 427 adult brains for 49 labels. The experimental results show that our method outperforms traditional multi-atlas and naive spherical U-Net approaches, while achieving full cortical parcellation in less than a minute.