Annotation-free prediction of treatment-specific tissue outcome from 4D CT perfusion imaging in acute ischemic stroke

Acute ischemic stroke is a critical health condition that requires timely intervention. Following admission, clinicians typically use perfusion imaging to facilitate treatment decision-making. While deep learning models leveraging perfusion data have demonstrated the ability to predict post-treatment tissue infarction for individual patients, predictions are often represented as binary or probabilistic masks that are not straightforward to interpret or easy to obtain. Moreover, these models typically rely on large amounts of subjectively segmented data and non-standard perfusion analysis techniques. To address these challenges, we propose a novel deep learning approach that directly predicts follow-up computed tomography images from full spatio-temporal 4D perfusion scans through a temporal compression. The results show that this method leads to realistic follow-up image predictions containing the infarcted tissue outcomes. The proposed compression method achieves comparable prediction results to using perfusion maps as inputs but without the need for perfusion analysis or arterial input function selection. Additionally, separate models trained on 45 patients treated with thrombolysis and 102 treated with thrombectomy showed that each model correctly captured the different patient-specific treatment effects as shown by image difference maps. The findings of this work clearly highlight the potential of our method to provide interpretable stroke treatment decision support without requiring manual annotations.

Visual Functions Are Associated with Biomarker Changes in Alzheimer's Disease

Background

While various biomarkers of Alzheimer's disease (AD) have been associated with general cognitive function, their association to visual-perceptive function across the AD spectrum warrant more attention due to its significant impact on quality of life. Thus, this study explores how AD biomarkers are associated with decline in this cognitive domain.

Objective

To explore associations between various fluid and imaging biomarkers and visual-based cognitive assessments in participants across the AD spectrum.

Methods

Data from participants (N = 1,460) in the Alzheimer's Disease Neuroimaging Initiative were analyzed, including fluid and imaging biomarkers. Along with the Mini-Mental State Examination (MMSE), three specific visual-based cognitive tests were investigated: Trail Making Test (TMT) A and TMT B, and the Boston Naming Test (BNT). Locally estimated scatterplot smoothing curves and Pearson correlation coefficients were used to examine associations.

Results

MMSE showed the strongest correlations with most biomarkers, followed by TMT-B. The p-tau181/Aβ1-42 ratio, along with the volume of the hippocampus and entorhinal cortex, had the strongest associations among the biomarkers.

Conclusions

Several biomarkers are associated with visual processing across the disease spectrum, emphasizing their potential in assessing disease severity and contributing to progression models of visual function and cognition.

Sex and age effects on gray matter volume trajectories in young children with prenatal alcohol exposure

Prenatal alcohol exposure (PAE) occurs in ~11% of North American pregnancies and is the most common known cause of neurodevelopmental disabilities such as fetal alcohol spectrum disorder (FASD; ~2-5% prevalence). PAE has been consistently associated with smaller gray matter volumes in children, adolescents, and adults. A small number of longitudinal studies show altered gray matter development trajectories in late childhood/early adolescence, but patterns in early childhood and potential sex differences have not been characterized in young children. Using longitudinal T1-weighted MRI, the present study characterized gray matter volume development in young children with PAE (N = 42, 84 scans, ages 3-8 years) compared to unexposed children (N = 127, 450 scans, ages 2-8.5 years). Overall, we observed altered global and regional gray matter development trajectories in the PAE group, wherein they had attenuated age-related increases and more volume decreases relative to unexposed children. Moreover, we found more pronounced sex differences in children with PAE; females with PAE having the smallest gray matter volumes and the least age-related changes of all groups. This pattern of altered development may indicate reduced brain plasticity and/or accelerated maturation and may underlie the cognitive/behavioral difficulties often experienced by children with PAE. In conjunction with previous research on older children, adolescents, and adults with PAE, our results suggest that gray matter volume differences associated with PAE vary by age and may become more apparent in older children.

Identifying Biases in a Multicenter MRI Database for Parkinson's Disease Classification: Is the Disease Classifier a Secret Site Classifier?

Sharing multicenter imaging datasets can be advantageous to increase data diversity and size but may lead to spurious correlations between site-related biological and non-biological image features and target labels, which machine learning (ML) models may exploit as shortcuts. To date, studies analyzing how and if deep learning models may use such effects as a shortcut are scarce. Thus, the aim of this work was to investigate if site-related effects are encoded in the feature space of an established deep learning model designed for Parkinson's disease (PD) classification based on T1-weighted MRI datasets. Therefore, all layers of the PD classifier were frozen, except for the last layer of the network, which was replaced by a linear layer that was exclusively re-trained to predict three potential bias types (biological sex, scanner type, and originating site). Our findings based on a large database consisting of 1880 MRI scans collected across 41 centers show that the feature space of the established PD model (74% accuracy) can be used to classify sex (75% accuracy), scanner type (79% accuracy), and site location (71% accuracy) with high accuracies despite this information never being explicitly provided to the PD model during original training. Overall, the results of this study suggest that trained image-based classifiers may use unwanted shortcuts that are not meaningful for the actual clinical task at hand. This finding may explain why many image-based deep learning models do not perform well when applied to data from centers not contributing to the training set.

Prediction of tissue outcome in acute ischemic stroke based on single-phase CT angiography at admission

Introduction

In acute ischemic stroke, prediction of the tissue outcome after reperfusion can be used to identify patients that might benefit from mechanical thrombectomy (MT). The aim of this work was to develop a deep learning model that can predict the follow-up infarct location and extent exclusively based on acute single-phase computed tomography angiography (CTA) datasets. In comparison to CT perfusion (CTP), CTA imaging is more widely available, less prone to artifacts, and the established standard of care in acute stroke imaging protocols. Furthermore, recent RCTs have shown that also patients with large established infarctions benefit from MT, which might not have been selected for MT based on CTP core/penumbra mismatch analysis.

Methods

All patients with acute large vessel occlusion of the anterior circulation treated at our institution between 12/2015 and 12/2020 were screened (N = 404) and 238 patients undergoing MT with successful reperfusion were included for final analysis. Ground truth infarct lesions were segmented on 24 h follow-up CT scans. Pre-processed CTA images were used as input for a U-Net-based convolutional neural network trained for lesion prediction, enhanced with a spatial and channel-wise squeeze-and-excitation block. Post-processing was applied to remove small predicted lesion components. The model was evaluated using a 5-fold cross-validation and a separate test set with Dice similarity coefficient (DSC) as the primary metric and average volume error as the secondary metric.

Results

The mean ± standard deviation test set DSC over all folds after post-processing was 0.35 ± 0.2 and the mean test set average volume error was 11.5 mL. The performance was relatively uniform across models with the best model according to the DSC achieved a score of 0.37 ± 0.2 after post-processing and the best model in terms of average volume error yielded 3.9 mL.

Conclusion

24 h follow-up infarct prediction using acute CTA imaging exclusively is feasible with DSC measures comparable to results of CTP-based algorithms reported in other studies. The proposed method might pave the way to a wider acceptance, feasibility, and applicability of follow-up infarct prediction based on artificial intelligence.

Multimodal imaging measures in the prediction of clinical response to deep brain stimulation for refractory depression: A machine learning approach

OBJECTIVES

This study compared machine learning models using unimodal imaging measures and combined multi-modal imaging measures for deep brain stimulation (DBS) outcome prediction in treatment resistant depression (TRD).

METHODS

Regional brain glucose metabolism (CMRGlu), cerebral blood flow (CBF), and grey matter volume (GMV) were measured at baseline using 18F-fluorodeoxy glucose (18F-FDG) positron emission tomography (PET), arterial spin labelling (ASL) magnetic resonance imaging (MRI), and T1-weighted MRI, respectively, in 19 patients with TRD receiving subcallosal cingulate (SCC)-DBS. Responders (n = 9) were defined by a 50% reduction in HAMD-17 at 6 months from the baseline. Using an atlas-based approach, values of each measure were determined for pre-selected brain regions. OneR feature selection algorithm and the naïve Bayes model was used for classification. Leave-out-one cross validation was used for classifier evaluation.

RESULTS

The performance accuracy of the CMRGlu classification model (84%) was greater than CBF (74%) or GMV (74%) models. The classification model using the three image modalities together led to a similar accuracy (84%0 compared to the CMRGlu classification model.

CONCLUSIONS

CMRGlu imaging measures may be useful for the development of multivariate prediction models for SCC-DBS studies for TRD. The future of multivariate methods for multimodal imaging may rest on the selection of complementing features and the developing better models.Clinical Trial Registration: ClinicalTrials.gov (#NCT01983904).

Exploiting macro- and micro-structural brain changes for improved Parkinson's disease classification from MRI data

Parkinson's disease (PD) is the second most common neurodegenerative disease. Accurate PD diagnosis is crucial for effective treatment and prognosis but can be challenging, especially at early disease stages. This study aimed to develop and evaluate an explainable deep learning model for PD classification from multimodal neuroimaging data. The model was trained using one of the largest collections of T1-weighted and diffusion-tensor magnetic resonance imaging (MRI) datasets. A total of 1264 datasets from eight different studies were collected, including 611 PD patients and 653 healthy controls (HC). These datasets were pre-processed and non-linearly registered to the MNI PD25 atlas. Six imaging maps describing the macro- and micro-structural integrity of brain tissues complemented with age and sex parameters were used to train a convolutional neural network (CNN) to classify PD/HC subjects. Explainability of the model's decision-making was achieved using SmoothGrad saliency maps, highlighting important brain regions. The CNN was trained using a 75%/10%/15% train/validation/test split stratified by diagnosis, sex, age, and study, achieving a ROC-AUC of 0.89, accuracy of 80.8%, specificity of 82.4%, and sensitivity of 79.1% on the test set. Saliency maps revealed that diffusion tensor imaging data, especially fractional anisotropy, was more important for the classification than T1-weighted data, highlighting subcortical regions such as the brainstem, thalamus, amygdala, hippocampus, and cortical areas. The proposed model, trained on a large multimodal MRI database, can classify PD patients and HC subjects with high accuracy and clinically reasonable explanations, suggesting that micro-structural brain changes play an essential role in the disease course.

A multi-center distributed learning approach for Parkinson's disease classification using the traveling model paradigm

Distributed learning is a promising alternative to central learning for machine learning (ML) model training, overcoming data-sharing problems in healthcare. Previous studies exploring federated learning (FL) or the traveling model (TM) setup for medical image-based disease classification often relied on large databases with a limited number of centers or simulated artificial centers, raising doubts about real-world applicability. This study develops and evaluates a convolution neural network (CNN) for Parkinson's disease classification using data acquired by 83 diverse real centers around the world, mostly contributing small training samples. Our approach specifically makes use of the TM setup, which has proven effective in scenarios with limited data availability but has never been used for image-based disease classification. Our findings reveal that TM is effective for training CNN models, even in complex real-world scenarios with variable data distributions. After sufficient training cycles, the TM-trained CNN matches or slightly surpasses the performance of the centrally trained counterpart (AUROC of 83% vs. 80%). Our study highlights, for the first time, the effectiveness of TM in 3D medical image classification, especially in scenarios with limited training samples and heterogeneous distributed data. These insights are relevant for situations where ML models are supposed to be trained using data from small or remote medical centers, and rare diseases with sparse cases. The simplicity of this approach enables a broad application to many deep learning tasks, enhancing its clinical utility across various contexts and medical facilities.

Challenges and Potential of Artificial Intelligence in Neuroradiology

PURPOSE

Artificial intelligence (AI) has emerged as a transformative force in medical research and is garnering increased attention in the public consciousness. This represents a critical time period in which medical researchers, healthcare providers, insurers, regulatory agencies, and patients are all developing and shaping their beliefs and policies regarding the use of AI in the healthcare sector. The successful deployment of AI will require support from all these groups. This commentary proposes that widespread support for medical AI must be driven by clear and transparent scientific reporting, beginning at the earliest stages of scientific research.

METHODS

A review of relevant guidelines and literature describing how scientific reporting plays a central role at key stages in the life cycle of an AI software product was conducted. To contextualize this principle within a specific medical domain, we discuss the current state of predictive tissue outcome modeling in acute ischemic stroke and the unique challenges presented therein.

RESULTS AND CONCLUSION

Translating AI methods from the research to the clinical domain is complicated by challenges related to model design and validation studies, medical product regulations, and healthcare providers' reservations regarding AI's efficacy and affordability. However, each of these limitations is also an opportunity for high-impact research that will help to accelerate the clinical adoption of state-of-the-art medical AI. In all cases, establishing and adhering to appropriate reporting standards is an important responsibility that is shared by all of the parties involved in the life cycle of a prospective AI software product.

An interactive atlas of three-dimensional syndromic facial morphology

Craniofacial phenotyping is critical for both syndrome delineation and diagnosis because craniofacial abnormalities occur in 30% of characterized genetic syndromes. Clinical reports, textbooks, and available software tools typically provide two-dimensional, static images and illustrations of the characteristic phenotypes of genetic syndromes. In this work, we provide an interactive web application that provides three-dimensional, dynamic visualizations for the characteristic craniofacial effects of 95 syndromes. Users can visualize syndrome facial appearance estimates quantified from data and easily compare craniofacial phenotypes of different syndromes. Our application also provides a map of morphological similarity between a target syndrome and other syndromes. Finally, users can upload 3D facial scans of individuals and compare them to our syndrome atlas estimates. In summary, we provide an interactive reference for the craniofacial phenotypes of syndromes that allows for precise, individual-specific comparisons of dysmorphology.

Providing clinical context to the spatio-temporal analysis of 4D CT perfusion to predict acute ischemic stroke lesion outcomes

Acute ischemic stroke is a leading cause of mortality and morbidity worldwide. Timely identification of the extent of a stroke is crucial for effective treatment, whereas spatio-temporal (4D) Computed Tomography Perfusion (CTP) imaging is playing a critical role in this process. Recently, the first deep learning-based methods that leverage the full spatio-temporal nature of perfusion imaging for predicting stroke lesion outcomes have been proposed. However, clinical information is typically not integrated into the learning process, which may be helpful to improve the tissue outcome prediction given the known influence of various factors (i.e., physiological, demographic, and treatment factors) on lesion growth. Cross-attention, a multimodal fusion strategy, has been successfully used to combine information from multiple sources, but it has yet to be applied to stroke lesion outcome prediction. Therefore, this work aimed to develop and evaluate a novel multimodal and spatio-temporal deep learning model that utilizes cross-attention to combine information from 4D CTP and clinical metadata simultaneously to predict stroke lesion outcomes. The proposed model was evaluated using a dataset of 70 acute ischemic stroke patients, demonstrating significantly improved volume estimates (mean error = 19 ml) compared to a baseline unimodal approach (mean error = 35 ml, p< 0.05). The proposed model allows generating attention maps and counterfactual outcome scenarios to investigate the relevance of clinical variables in predicting stroke lesion outcomes at a patient level, helping to provide a better understanding of the model's decision-making process.

Simulation of neuroplasticity in a CNN-based in-silico model of neurodegeneration of the visual system

The aim of this work was to enhance the biological feasibility of a deep convolutional neural network-based in-silico model of neurodegeneration of the visual system by equipping it with a mechanism to simulate neuroplasticity. Therefore, deep convolutional networks of multiple sizes were trained for object recognition tasks and progressively lesioned to simulate neurodegeneration of the visual cortex. More specifically, the injured parts of the network remained injured while we investigated how the added retraining steps were able to recover some of the model's object recognition baseline performance. The results showed with retraining, model object recognition abilities are subject to a smoother and more gradual decline with increasing injury levels than without retraining and, therefore, more similar to the longitudinal cognition impairments of patients diagnosed with Alzheimer's disease (AD). Moreover, with retraining, the injured model exhibits internal activation patterns similar to those of the healthy baseline model when compared to the injured model without retraining. Furthermore, we conducted this analysis on a network that had been extensively pruned, resulting in an optimized number of parameters or synapses. Our findings show that this network exhibited remarkably similar capability to recover task performance with decreasingly viable pathways through the network. In conclusion, adding a retraining step to the in-silico setup that simulates neuroplasticity improves the model's biological feasibility considerably and could prove valuable to test different rehabilitation approaches in-silico.

Image-encoded biological and non-biological variables may be used as shortcuts in deep learning models trained on multisite neuroimaging data

OBJECTIVE

This work investigates if deep learning (DL) models can classify originating site locations directly from magnetic resonance imaging (MRI) scans with and without correction for intensity differences.

MATERIAL AND METHODS

A large database of 1880 T1-weighted MRI scans collected across 41 sites originally for Parkinson's disease (PD) classification was used to classify sites in this study. Forty-six percent of the datasets are from PD patients, while 54% are from healthy participants. After preprocessing the T1-weighted scans, 2 additional data types were generated: intensity-harmonized T1-weighted scans and log-Jacobian deformation maps resulting from nonlinear atlas registration. Corresponding DL models were trained to classify sites for each data type. Additionally, logistic regression models were used to investigate the contribution of biological (age, sex, disease status) and non-biological (scanner type) variables to the models' decision.

RESULTS

A comparison of the 3 different types of data revealed that DL models trained using T1-weighted and intensity-harmonized T1-weighted scans can classify sites with an accuracy of 85%, while the model using log-Jacobian deformation maps achieved a site classification accuracy of 54%. Disease status and scanner type were found to be significant confounders.

DISCUSSION

Our results demonstrate that MRI scans encode relevant site-specific information that models could use as shortcuts that cannot be removed using simple intensity harmonization methods.

CONCLUSION

The ability of DL models to exploit site-specific biases as shortcuts raises concerns about their reliability, generalization, and deployability in clinical settings.

The causal link between cardiometabolic risk factors and gray matter atrophy: An exploratory study

Although gray matter atrophy is commonly observed with aging, it is highly variable, even among healthy people of the same age. This raises the question of what other factors may contribute to gray matter atrophy. Previous studies have reported that risk factors for cardiometabolic diseases are associated with accelerated brain aging. However, these studies were primarily based on standard correlation analyses, which do not unveil a causal relationship. While randomized controlled trials are typically required to investigate true causality, in this work, we investigated an alternative method by exploring data-driven causal discovery and inference techniques on observational data. Accordingly, this feasibility study used clinical and quantified gray matter volume data from 22,793 subjects from the UK biobank cohort without any known neurological disease. Our method identified that age, sex, body mass index (BMI), body fat percentage (BFP), and smoking exhibit a causal relationship with gray matter volume. Interventions on the causal network revealed that higher BMI and BFP values significantly increased the chance of gray matter atrophy in males, whereas this was not the case in females.

Machine learning using multimodal clinical, electroencephalographic, and magnetic resonance imaging data can predict incident depression in adults with epilepsy: A pilot study

OBJECTIVE

This study was undertaken to develop a multimodal machine learning (ML) approach for predicting incident depression in adults with epilepsy.

METHODS

We randomly selected 200 patients from the Calgary Comprehensive Epilepsy Program registry and linked their registry-based clinical data to their first-available clinical electroencephalogram (EEG) and magnetic resonance imaging (MRI) study. We excluded patients with a clinical or Neurological Disorders Depression Inventory for Epilepsy (NDDI-E)-based diagnosis of major depression at baseline. The NDDI-E was used to detect incident depression over a median of 2.4 years of follow-up (interquartile range [IQR] = 1.5-3.3 years). A ReliefF algorithm was applied to clinical as well as quantitative EEG and MRI parameters for feature selection. Six ML algorithms were trained and tested using stratified threefold cross-validation. Multiple metrics were used to assess model performances.

RESULTS

Of 200 patients, 150 had EEG and MRI data of sufficient quality for ML, of whom 59 were excluded due to prevalent depression. Therefore, 91 patients (41 women) were included, with a median age of 29 (IQR = 22-44) years. A total of 42 features were selected by ReliefF, none of which was a quantitative MRI or EEG variable. All models had a sensitivity > 80%, and five of six had an F1 score ≥ .72. A multilayer perceptron model had the highest F1 score (median = .74, IQR = .71-.78) and sensitivity (84.3%). Median area under the receiver operating characteristic curve and normalized Matthews correlation coefficient were .70 (IQR = .64-.78) and .57 (IQR = .50-.65), respectively.

SIGNIFICANCE

Multimodal ML using baseline features can predict incident depression in this population. Our pilot models demonstrated high accuracy for depression prediction. However, overall performance and calibration can be improved. This model has promise for identifying those at risk for incident depression during follow-up, although efforts to refine it in larger populations along with external validation are required.

Self-supervised-RCNN for medical image segmentation with limited data annotation

Many successful methods developed for medical image analysis based on machine learning use supervised learning approaches, which often require large datasets annotated by experts to achieve high accuracy. However, medical data annotation is time-consuming and expensive, especially for segmentation tasks. To overcome the problem of learning with limited labeled medical image data, an alternative deep learning training strategy based on self-supervised pretraining on unlabeled imaging data is proposed in this work. For the pretraining, different distortions are arbitrarily applied to random areas of unlabeled images. Next, a Mask-RCNN architecture is trained to localize the distortion location and recover the original image pixels. This pretrained model is assumed to gain knowledge of the relevant texture in the images from the self-supervised pretraining on unlabeled imaging data. This provides a good basis for fine-tuning the model to segment the structure of interest using a limited amount of labeled training data. The effectiveness of the proposed method in different pretraining and fine-tuning scenarios was evaluated based on the Osteoarthritis Initiative dataset with the aim of segmenting effusions in MRI datasets of the knee. Applying the proposed self-supervised pretraining method improved the Dice score by up to 18% compared to training the models using only the limited annotated data. The proposed self-supervised learning approach can be applied to many other medical image analysis tasks including anomaly detection, segmentation, and classification.

Expert elicitation of risk factors for progression to dementia in individuals with mild cognitive impairment

INTRODUCTION

This study assesses experts' beliefs about important predictors of developing dementia in persons with mild cognitive impairment (MCI).

METHODS

Structured expert elicitation, a methodology to quantify expert knowledge, was used to elicit the most important risk factors for developing dementia. We recruited 11 experts (6 neurologists, 3 geriatricians, and 2 psychiatrists). Ten experts fully participated in introductory meetings, two rounds of surveys, and discussion meetings. The data from these ten experts were utilized for this study.

RESULTS

The expert elicitation identified age, CSF analysis, fluorodeoxyglucose-positron emission tomography (FDG-PET) findings, hippocampal atrophy, MoCA (or MMSE) score, parkinsonism, apathy, psychosis, informant report of cognitive symptoms, and global atrophy as the ten most important predictors of progressing to dementia in persons with MCI.

DISCUSSION

Several dementia predictors are not routinely collected in existing registries, observational studies, or usual care. This might partially explain the low uptake of existing published dementia risk scores in clinical practice.

Comparing 2D and 3D representations for face-based genetic syndrome diagnosis

Human genetic syndromes are often challenging to diagnose clinically. Facial phenotype is a key diagnostic indicator for hundreds of genetic syndromes and computer-assisted facial phenotyping is a promising approach to assist diagnosis. Most previous approaches to automated face-based syndrome diagnosis have analyzed different datasets of either 2D images or surface mesh-based 3D facial representations, making direct comparisons of performance challenging. In this work, we developed a set of subject-matched 2D and 3D facial representations, which we then analyzed with the aim of comparing the performance of 2D and 3D image-based approaches to computer-assisted syndrome diagnosis. This work represents the most comprehensive subject-matched analyses to date on this topic. In our analyses of 1907 subject faces representing 43 different genetic syndromes, 3D surface-based syndrome classification models significantly outperformed 2D image-based models trained and evaluated on the same subject faces. These results suggest that the clinical adoption of 3D facial scanning technology and continued collection of syndromic 3D facial scan data may substantially improve face-based syndrome diagnosis.

Machine learning-based prediction of longitudinal cognitive decline in early Parkinson's disease using multimodal features

Patients with Parkinson's Disease (PD) often suffer from cognitive decline. Accurate prediction of cognitive decline is essential for early treatment of at-risk patients. The aim of this study was to develop and evaluate a multimodal machine learning model for the prediction of continuous cognitive decline in patients with early PD. We included 213 PD patients from the Parkinson's Progression Markers Initiative (PPMI) database. Machine learning was used to predict change in Montreal Cognitive Assessment (MoCA) score using the difference between baseline and 4-years follow-up data as outcome. Input features were categorized into four sets: clinical test scores, cerebrospinal fluid (CSF) biomarkers, brain volumes, and genetic variants. All combinations of input feature sets were added to a basic model, which consisted of demographics and baseline cognition. An iterative scheme using RReliefF-based feature ranking and support vector regression in combination with tenfold cross validation was used to determine the optimal number of predictive features and to evaluate model performance for each combination of input feature sets. Our best performing model consisted of a combination of the basic model, clinical test scores and CSF-based biomarkers. This model had 12 features, which included baseline cognition, CSF phosphorylated tau, CSF total tau, CSF amyloid-beta1-42, geriatric depression scale (GDS) scores, and anxiety scores. Interestingly, many of the predictive features in our model have previously been associated with Alzheimer's disease, showing the importance of assessing Alzheimer's disease pathology in patients with Parkinson's disease.

Elicited clinician knowledge did not improve dementia risk prediction in individuals with mild cognitive impairment

OBJECTIVES

This study aims to develop and validate a Bayesian risk prediction model that combines research cohort data with elicited expert knowledge to predict dementia progression in people with mild cognitive impairment (MCI).

STUDY DESIGN AND SETTING

This is a prognostic risk prediction modeling study based on cohort data (Alzheimer's disease neuroimaging initiative [ADNI]; n = 365) of research participants with MCI and elicited expert data. Bayesian Cox models were used to combine expert knowledge and ADNI data to predict dementia progression in people with MCI. Posterior distributions were obtained based on Gibbs sampler and the predictive performance was evaluated using ten-fold cross-validation via c-index, integrated calibration index (ICI), and integrated brier score (IBS).

RESULTS

365 people with MCI were included, mean age was 73 years (SD = 7.5), and 39% developed dementia within 3 years. When expert knowledge was incorporated, the c-index, ICI, and IBS values were 0.74 (95% CI 0.70-0.79), 0.06 (95% CI 0.05-0.08), and 0.17 (95% CI 0.14-0.19), respectively. These were similar to the model without expert knowledge data.

CONCLUSION

The addition of expert knowledge did not improve model accuracy in this ADNI sample to predict dementia progression in individuals with MCI.

Factors predicting general health concerns and atypical behaviours in children with prenatal alcohol exposure and other adverse exposures

Background

Prenatal alcohol exposure (PAE) can have significant negative consequences on the health outcomes of children. Children with PAE often experience other prenatal and postnatal adverse exposures. Increased rates of general health concerns and atypical behaviours are seen in both children with PAE as well as with other patterns of adverse exposures, although these have not been systematically described. The association between multiple adverse exposures and adverse health concerns and atypical behaviours in children with PAE is unknown.

Methods

Demographic information, medical history, adverse exposures, health concerns, and atypical behaviours were collected from children with confirmed PAE (n = 22; 14 males, age range = 7.9-15.9 years) and their caregivers. Support vector machine learning classification models were used to predict the presence of health concerns and atypical behaviours based on adverse exposures. Associations between the sums of adverse exposures, health concerns, and atypical behaviours were examined using correlation analysis.

Results

All children experienced health concerns, the most common being sensitivity to sensory inputs (64%; 14/22). Similarly, all children engaged in atypical behaviours, with atypical sensory behaviour (50%; 11/22) being the most common. Prenatal alcohol exposure was most important factor for predicting some health concerns and atypical behaviours, and alone and in combination with other factors. Simple associations between adverse exposures could not be identified for many health concerns and atypical behaviours.

Conclusion

Children with PAE and other adverse exposures experience high rates of health concerns and atypical behaviours. This study demonstrates the complex effects of multiple adverse exposures on health and behaviour in children.

Quantifying the relationship between cell proliferation and morphology during development of the face

Morphogenesis requires highly coordinated, complex interactions between cellular processes: proliferation, migration, and apoptosis, along with physical tissue interactions. How these cellular and tissue dynamics drive morphogenesis remains elusive. Three dimensional (3D) microscopic imaging poses great promise, and generates elegant images. However, generating even moderate through-put quantified images is challenging for many reasons. As a result, the association between morphogenesis and cellular processes in 3D developing tissues has not been fully explored. To address this critical gap, we have developed an imaging and image analysis pipeline to enable 3D quantification of cellular dynamics along with 3D morphology for the same individual embryo. Specifically, we focus on how 3D distribution of proliferation relates to morphogenesis during mouse facial development. Our method involves imaging with light-sheet microscopy, automated segmentation of cells and tissues using machine learning-based tools, and quantification of external morphology via geometric morphometrics. Applying this framework, we show that changes in proliferation are tightly correlated to changes in morphology over the course of facial morphogenesis. These analyses illustrate the potential of this pipeline to investigate mechanistic relationships between cellular dynamics and morphogenesis during embryonic development.

Explainable classification of Parkinson's disease using deep learning trained on a large multi-center database of T1-weighted MRI datasets

INTRODUCTION

Parkinson's disease (PD) is a severe neurodegenerative disease that affects millions of people. Early diagnosis is important to facilitate prompt interventions to slow down disease progression. However, accurate PD diagnosis can be challenging, especially in the early disease stages. The aim of this work was to develop and evaluate a robust explainable deep learning model for PD classification trained from one of the largest collections of T1-weighted magnetic resonance imaging datasets.

MATERIALS AND METHODS

A total of 2,041 T1-weighted MRI datasets from 13 different studies were collected, including 1,024 datasets from PD patients and 1,017 datasets from age- and sex-matched healthy controls (HC). The datasets were skull stripped, resampled to isotropic resolution, bias field corrected, and non-linearly registered to the MNI PD25 atlas. The Jacobian maps derived from the deformation fields together with basic clinical parameters were used to train a state-of-the-art convolutional neural network (CNN) to classify PD and HC subjects. Saliency maps were generated to display the brain regions contributing the most to the classification task as a means of explainable artificial intelligence.

RESULTS

The CNN model was trained using an 85%/5%/10% train/validation/test split stratified by diagnosis, sex, and study. The model achieved an accuracy of 79.3%, precision of 80.2%, specificity of 81.3%, sensitivity of 77.7%, and AUC-ROC of 0.87 on the test set while performing similarly on an independent test set. Saliency maps computed for the test set data highlighted frontotemporal regions, the orbital-frontal cortex, and multiple deep gray matter structures as most important.

CONCLUSION

The developed CNN model, trained on a large heterogenous database, was able to differentiate PD patients from HC subjects with high accuracy with clinically feasible classification explanations. Future research should aim to investigate the combination of multiple imaging modalities with deep learning and on validating these results in a prospective trial as a clinical decision support system.

Distinct phenotypes of multisystem inflammatory syndrome in children: a cohort study

BACKGROUND

Multisystem inflammatory syndrome in children (MIS-C) is a severe disease with an unpredictable course and a substantial risk of cardiogenic shock. Our objectives were to (a) compare MIS-C phenotypes across the COVID-19 pandemic, (b) identify features associated with intensive care need and treatment with biologic agents.

METHODS

Youth aged 0-18 years, fulfilling the World Health Organization case definition of MIS-C, and admitted to the Alberta Children's Hospital during the first four waves of the COVID-19 pandemic (May 2020-December 2021) were included in this cohort study. Demographic, clinical, biochemical, imaging, and treatment data were captured.

RESULTS

Fifty-seven MIS-C patients (median age 6 years, range 0-17) were included. Thirty patients (53%) required intensive care. Patients in the third or fourth wave (indicated as phase 2 of the pandemic) presented with higher peak ferritin (µg/l, median (IQR) = 1134 (409-1806) vs. 370 (249-629), P = 0.001), NT-proBNP (ng/l, median (IQR) = 12,217 (3013-27,161) vs. 3213 (1216-8483), P = 0.02) and D-dimer (mg/l, median (IQR) = 4.81 (2.24-5.37) vs. 2.01 (1.27-3.34), P = 0.004) levels, and higher prevalence of liver enzyme abnormalities (n(%) = 17 (68) vs. 11 (34), P = 0.02), hypoalbuminemia (n(%) = 24 (100) vs. 25 (81), P = 0.03) and thrombocytopenia (n(%) 18 (72) vs. 11 (34), P = 0.007) compared to patients in the first two waves (phase 1). These patients had a higher need of non-invasive/mechanical ventilation (n(%) 4 (16) vs. 0 (0), P = 0.03). Unsupervised clustering analyses classified 47% of the patients in the correct wave and 74% in the correct phase of the pandemic. NT-proBNP was the only significant contributor to the need for intensive care in all applied multivariate regression models. Treatment with biologic agents was significantly associated with peak CRP (mg/l (median, IQR = 240.9 (132.9-319.4) vs. 155.8 (101.0-200.7), P = 0.02) and ferritin levels (µg/l, median (IQR) = 1380 (509-1753) vs. 473 (280-296)).

CONCLUSIONS

MIS-C patients in a later stage of the pandemic displayed a more severe phenotype, reflecting the impact of distinct SARS-CoV-2 variants. NT-proBNP emerged as the most crucial feature associated with intensive care need, underscoring the importance of monitoring.

Lesion-preserving unpaired image-to-image translation between MRI and CT from ischemic stroke patients

PURPOSE

Multiple medical imaging modalities are used for clinical follow-up ischemic stroke analysis. Mixed-modality datasets are challenging, both for clinical rating purposes and for training machine learning models. While image-to-image translation methods have been applied to harmonize stroke patient images to a single modality, they have only been used for paired data so far. In the more common unpaired scenario, the standard cycle-consistent generative adversarial network (CycleGAN) method is not able to translate the stroke lesions properly. Thus, the aim of this work was to develop and evaluate a novel image-to-image translation regularization approach for unpaired 3D follow-up stroke patient datasets.

METHODS

A modified CycleGAN was used to translate images between 238 non-contrast computed tomography (NCCT) and 244 fluid-attenuated inversion recovery (FLAIR) MRI datasets, two of the most relevant follow-up modalities in clinical practice. We introduced an additional attention-guided mechanism to encourage an improved translation of the lesion and a gradient-consistency loss to preserve structural brain morphology.

RESULTS

The proposed modifications were able to preserve the overall quality provided by the CycleGAN translation. This was confirmed by the FID score and gradient correlation results. Furthermore, the lesion preservation was significantly improved compared to a standard CycleGAN. This was evaluated for location and volume with segmentation models, which were trained on real datasets and applied to the translated test images. Here, the Dice score coefficient resulted in 0.81 and 0.62 for datasets translated to FLAIR and NCCT, respectively, compared to 0.57 and 0.50 for the corresponding datasets translated using a standard CycleGAN. Finally, an analysis of the distribution of mean lesion intensities showed substantial improvements.

CONCLUSION

The results of this work show that the proposed image-to-image translation method is effective at preserving stroke lesions in unpaired modality translation, supporting its potential as a tool for stroke image analysis in real-life scenarios.

Effect of geometric distortion correction on thickness and volume measurements of cortical parcellations in 3D T1w gradient echo sequences

OBJECTIVE

Automated brain volumetric analysis based on high-resolution T1-weighted MRI datasets is a frequently used tool in neuroimaging for early detection, diagnosis, and monitoring of various neurological diseases. However, image distortions can corrupt and bias the analysis. The aim of this study was to explore the variability of brain volumetric analysis due to gradient distortions and to investigate the effect of distortion correction methods implemented on commercial scanners.

MATERIAL AND METHODS

36 healthy volunteers underwent brain imaging using a 3T magnetic resonance imaging (MRI) scanner, including a high-resolution 3D T1-weighted sequence. For all participants, each T1-weighted image was reconstructed directly on the vendor workstation with (DC) and without (nDC) distortion correction. For each participant's set of DC and nDC images, FreeSurfer was used for the determination of regional cortical thickness and volume.

RESULTS

Overall, significant differences were found in 12 cortical ROIs comparing the volumes of the DC and nDC data and in 19 cortical ROIs comparing the thickness of the DC and nDC data. The most pronounced differences for cortical thickness were found in the precentral gyrus, the lateral occipital and postcentral ROI (2.69, -2.91% and -2.79%, respectively) while cortical volumes differed most prominently in the paracentral, the pericalcarine and lateral occipital ROI (5.52%, -5.40% and -5.11%, respectively).

CONCLUSION

Correcting for gradient non-linearities can have significant influence on volumetric analysis of cortical thickness and volume. Since the distortion correction is an automatic feature of the MR scanner, it should be stated by each study that applies volumetric analysis which images were used.

Motor symptoms in Parkinson's disease are related to the interplay between cortical curvature and thickness

INTRODUCTION

Brain atrophy in Parkinson's disease occurs to varying degrees in different brain regions, even at the early stage of the disease. While cortical morphological features are often considered independently in structural brain imaging studies, research on the co-progression of different cortical morphological measurements could provide new insights regarding the progression of PD. This study's aim was to examine the interplay between cortical curvature and thickness as a function of PD diagnosis, motor symptoms, and cognitive performance.

METHODS

A total of 359 de novo PD patients and 159 healthy controls (HC) from the Parkinson's Progression Markers Initiative (PPMI) database were included in this study. Additionally, an independent cohort from four databases (182 PD, 132 HC) with longer disease durations was included to assess the effects of PD diagnosis in more advanced cases. Pearson correlation was used to determine subject-specific associations between cortical curvature and thickness estimated from T1-weighted MRI images. General linear modeling (GLM) was then used to assess the effect of PD diagnosis, motor symptoms, and cognitive performance on the curvature-thickness association. Next, longitudinal changes in the curvature-thickness correlation as well as the predictive effect of the cortical curvature-thickness association on changes in motor symptoms and cognitive performance across four years were investigated. Finally, Akaike information criterion (AIC) was used to build a GLM to model PD motor symptom severity cross-sectionally.

RESULTS

A significant interaction effect between PD motor symptoms and age on the curvature-thickness correlation was found (β_standardized = 0.11; t(350) = 2.12; p = 0.03). This interaction effect showed that motor symptoms in older patients were related to an attenuated curvature-thickness association. No significant effect of PD diagnosis was observed for the PPMI database (β = 0.03; t(510) = 0.35; p = 0.72). However, in patients with a longer disease duration, a significant effect of diagnosis on the curvature-thickness association was found (β_standardized = 0.31; t(306.7) = 3.49; p = 0.0006). Moreover, rigidity, but not tremor, in PD was significantly related to the curvature-thickness correlation (β_standardized = 0.11, t(350) = 2.24, p = 0.03; β_standardized = -0.03, t(350) = -0.58, p = 0.56, respectively). The curvature-thickness association was attenuated over time in both PD and HC, but the two groups did not show a significantly different effect (β_standardized = 0.03, t(184.7) = 0.78, p = 0.44). No predictive effects of the CC-CT correlation on longitudinal changes in cognitive performance or motor symptoms were observed (all p-values > 0.05). The best cross-sectional model for PD motor symptoms included the curvature-thickness correlation, cognitive performance, and putamen dopamine transporter (DAT) binding, which together explained 14 % of variance.

CONCLUSION

The association between cortical curvature and thickness is related to PD motor symptoms and age. This research shows the potential of modeling the curvature-thickness interplay in PD.

Aerobic exercise increases brain vessel lumen size and blood flow in young adults with elevated blood pressure. Secondary analysis of the TEPHRA randomized clinical trial

IMPORTANCE

Cerebrovascular changes are already evident in young adults with hypertension and exercise is recommended to reduce cardiovascular risk. To what extent exercise benefits the cerebrovasculature at an early stage of the disease remains unclear.

OBJECTIVE

To investigate whether structured aerobic exercise increases brain vessel lumen diameter or cerebral blood flow (CBF) and whether lumen diameter is associated with CBF.

DESIGN

Open, parallel, two-arm superiority randomized controlled (1:1) trial in the TEPHRA study on an intention-to-treat basis. The MRI sub-study was an optional part of the protocol. The outcome assessors remained blinded until the data lock.

SETTING

Single-centre trial in Oxford, UK.

PARTICIPANTS

Participants were physically inactive (<150 min/week moderate to vigorous physical activity), 18 to 35 years old, 24-hour ambulatory blood pressure 115/75 mmHg-159/99 mmHg, body mass index below 35 kg/m² and never been on prescribed hypertension medications. Out of 203 randomized participants, 135 participated in the MRI sub-study. Randomisation was stratified for sex, age (<24, 24-29, 30-35 years) and gestational age at birth (<32, 32-37, >37 weeks).

INTERVENTION

Study participants were randomised to a 16 week aerobic exercise intervention targeting 3×60 min sessions per week at 60 to 80 % peak heart rate.

MAIN OUTCOMES AND MEASURES

cerebral blood flow (CBF) maps from ASL MRI scans, internal carotid artery (ICA), middle cerebral artery (MCA) M1 and M2 segments, anterior cerebral artery (ACA), basilar artery (BA), and posterior cerebral artery (PCA) diameters extracted from TOF MRI scans.

RESULTS

Of the 135 randomized participants (median age 28 years, 58 % women) who had high quality baseline MRI data available, 93 participants also had high quality follow-up data available. The exercise group showed an increase in ICA (0.1 cm, 95 % CI 0.01 to 0.18, p =.03) and MCA M1 (0.05 cm, 95 % CI 0.01 to 0.10, p =.03) vessel diameter compared to the control group. Differences in the MCA M2 (0.03 cm, 95 % CI 0.0 to 0.06, p =.08), ACA (0.04 cm, 95 % CI 0.0 to 0.08, p =.06), BA (0.02 cm, 95 % CI -0.04 to 0.09, p =.48), and PCA (0.03 cm, 95 % CI -0.01 to 0.06, p =.17) diameters or CBF were not statistically significant. The increase in ICA vessel diameter in the exercise group was associated with local increases in CBF.

CONCLUSIONS AND RELEVANCE

Aerobic exercise induces positive cerebrovascular remodelling in young people with early hypertension, independent of blood pressure. The long-term benefit of these changes requires further study.

TRIAL REGISTRATION

Clinicaltrials.gov NCT02723552, 30 March 2016.

Dementia in Convolutional Neural Networks: Using Deep Learning Models to Simulate Neurodegeneration of the Visual System

Although current research aims to improve deep learning networks by applying knowledge about the healthy human brain and vice versa, the potential of using such networks to model and study neurodegenerative diseases remains largely unexplored. In this work, we present an in-depth feasibility study modeling progressive dementia in silico with deep convolutional neural networks. Therefore, networks were trained to perform visual object recognition and then progressively injured by applying neuronal as well as synaptic injury. After each iteration of injury, network object recognition accuracy, saliency map similarity between the intact and injured networks, and internal activations of the degenerating models were evaluated. The evaluation revealed that cognitive function of the network progressively decreased with increasing injury load whereas this effect was much more pronounced for synaptic damage. The effects of neurodegeneration found for the in silico model are especially similar to the loss of visual cognition seen in patients with posterior cortical atrophy.

Apathy and APOE in mild behavioral impairment, and risk for incident dementia

Introduction

Mild behavioral impairment (MBI) is a high-risk state for incident dementia and comprises five core domains including affective dysregulation, impulse dyscontrol, social inappropriateness, psychotic symptoms, and apathy. Apathy is among the most common neuropsychiatric symptoms (NPS) in dementia but can also develop in persons with normal cognition (NC) or mild cognitive impairment (MCI). The later-life emergence and persistence of apathy as part of the MBI syndrome may be a driving factor for dementia risk. Therefore, we investigated MBI-apathy-associated progression to dementia, and effect modification by sex, race, cognitive diagnosis, and apolipoprotein E (APOE) genotype.

Methods

Dementia-free National Alzheimer's Coordinating Center participants were stratified by persistent apathy status, based on Neuropsychiatric Inventory (NPI)-Questionnaire scores at two consecutive visits. Hazard ratios (HRs) for incident dementia for MBI-apathy and NPI-apathy relative to no NPS, and MBI-apathy relative to no apathy, were determined using Cox proportional hazards regressions, adjusted for baseline age, sex, years of education, race, cognitive diagnosis, and APOE genotype. Interactions with relevant model covariates were explored.

Results

Of the 3932 participants (3247 with NC), 354 had MBI-apathy. Of all analytic groups, MBI-apathy had the greatest dementia incidence (HR = 2.69, 95% confidence interval [CI]: 2.15-3.36, P < 0.001). Interaction effects were observed between cognitive diagnosis and APOE genotype with the NPS group. The contribution of apathy to dementia risk was greater in NC (HR = 5.91, 95% CI: 3.91-8.93) than in MCI (HR = 2.16, 95% CI: 1.69-2.77, interaction P < 0.001) and in all APOE genotypes, was greatest in APOE ɛ3 (HR = 4.25, 95% CI: 3.1-5.82, interaction P < 0.001).

Discussion

Individuals with MBI-apathy have a markedly elevated risk for future dementia, especially when symptoms emerge in those with NC. Both cognitive status and APOE genotype are important moderators in the relationship between MBI-apathy and incident dementia. MBI-apathy may represent a group in whom apathy is a preclinical or prodromal manifestation of dementia and identify a precision medicine target for preventative interventions.

Multimodal brain age prediction fusing morphometric and imaging data and association with cardiovascular risk factors

Introduction

The difference between the chronological and biological brain age, called the brain age gap (BAG), has been identified as a promising biomarker to detect deviation from normal brain aging and to indicate the presence of neurodegenerative diseases. Moreover, the BAG has been shown to encode biological information about general health, which can be measured through cardiovascular risk factors. Current approaches for biological brain age estimation, and therefore BAG estimation, either depend on hand-crafted, morphological measurements extracted from brain magnetic resonance imaging (MRI) or on direct analysis of brain MRI images. The former can be processed with traditional machine learning models while the latter is commonly processed with convolutional neural networks (CNNs). Using a multimodal setting, this study aims to compare both approaches in terms of biological brain age prediction accuracy and biological information captured in the BAG.

Methods

T1-weighted MRI, containing brain tissue information, and magnetic resonance angiography (MRA), providing information about brain arteries, from 1,658 predominantly healthy adults were used. The volumes, surface areas, and cortical thickness of brain structures were extracted from the T1-weighted MRI data, while artery density and thickness within the major blood flow territories and thickness of the major arteries were extracted from MRA data. Independent multilayer perceptron and CNN models were trained to estimate the brain age from the hand-crafted features and image data, respectively. Next, both approaches were fused to assess the benefits of combining image data and hand-crafted features for brain age prediction.

Results

The combined model achieved a mean absolute error of 4 years between the chronological and predicted biological brain age. Among the independent models, the lowest mean absolute error was observed for the CNN using T1-weighted MRI data (4.2 years). When evaluating the BAGs obtained using the different approaches and imaging modalities, diverging associations between cardiovascular risk factors were found. For example, BAGs obtained from the CNN models showed an association with systolic blood pressure, while BAGs obtained from hand-crafted measurements showed greater associations with obesity markers.

Discussion

In conclusion, the use of more diverse sources of data can improve brain age estimation modeling and capture more diverse biological deviations from normal aging.

An analysis of the effects of limited training data in distributed learning scenarios for brain age prediction

OBJECTIVE

Distributed learning avoids problems associated with central data collection by training models locally at each site. This can be achieved by federated learning (FL) aggregating multiple models that were trained in parallel or training a single model visiting sites sequentially, the traveling model (TM). While both approaches have been applied to medical imaging tasks, their performance in limited local data scenarios remains unknown. In this study, we specifically analyze FL and TM performances when very small sample sizes are available per site.

MATERIALS AND METHODS

2025 T1-weighted magnetic resonance imaging scans were used to investigate the effect of sample sizes on FL and TM for brain age prediction. We evaluated models across 18 scenarios varying the number of samples per site (1, 2, 5, 10, and 20) and the number of training rounds (20, 40, and 200).

RESULTS

Our results demonstrate that the TM outperforms FL, for every sample size examined. In the extreme case when each site provided only one sample, FL achieved a mean absolute error (MAE) of 18.9 ± 0.13 years, while the TM achieved a MAE of 6.21 ± 0.50 years, comparable to central learning (MAE = 5.99 years).

DISCUSSION

Although FL is more commonly used, our study demonstrates that TM is the best implementation for small sample sizes.

CONCLUSION

The TM offers new opportunities to apply machine learning models in rare diseases and pediatric research but also allows even small hospitals to contribute small datasets.

Detecting 3D syndromic faces as outliers using unsupervised normalizing flow models

Many genetic syndromes are associated with distinctive facial features. Several computer-assisted methods have been proposed that make use of facial features for syndrome diagnosis. Training supervised classifiers, the most common approach for this purpose, requires large, comprehensive, and difficult to collect databases of syndromic facial images. In this work, we use unsupervised, normalizing flow-based manifold and density estimation models trained entirely on unaffected subjects to detect syndromic 3D faces as statistical outliers. Furthermore, we demonstrate a general, user-friendly, gradient-based interpretability mechanism that enables clinicians and patients to understand model inferences. 3D facial surface scans of 2471 unaffected subjects and 1629 syndromic subjects representing 262 different genetic syndromes were used to train and evaluate the models. The flow-based models outperformed unsupervised comparison methods, with the best model achieving an ROC-AUC of 86.3% on a challenging, age and sex diverse data set. In addition to highlighting the viability of outlier-based syndrome screening tools, our methods generalize and extend previously proposed outlier scores for 3D face-based syndrome detection, resulting in improved performance for unsupervised syndrome detection.

Predicting the tissue outcome of acute ischemic stroke from acute 4D computed tomography perfusion imaging using temporal features and deep learning

Predicting follow-up lesions from baseline CT perfusion (CTP) datasets in acute ischemic stroke patients is important for clinical decision making. Deep convolutional networks (DCNs) are assumed to be the current state-of-the-art for this task. However, many DCN classifiers have not been validated against the methods currently used in research (random decision forests, RDF) and clinical routine (Tmax thresholding). Specialized DCNs have even been designed to extract complex temporal features directly from spatiotemporal CTP data instead of using standard perfusion parameter maps. However, the benefits of applying deep learning to source or deconvolved CTP data compared to perfusion parameter maps have not been formally investigated so far. In this work, a modular UNet-based DCN is proposed that separates temporal feature extraction from tissue outcome prediction, allowing for both model validation using perfusion parameter maps as well as end-to-end learning from spatiotemporal CTP data. 145 retrospective datasets comprising baseline CTP imaging, perfusion parameter maps, and follow-up non-contrast CT with manual lesion segmentations were assembled from acute ischemic stroke patients treated with intravenous thrombolysis alone (IV; n = 43) or intra-arterial mechanical thrombectomy (IA; n = 102) with or without combined IV. Using the perfusion parameter maps as input, the proposed DCN (mean Dice: 0.287) outperformed the RDF (0.262) and simple Tmax-thresholding (0.249). The performance of the proposed DCN was approximately equal using features optimized from the deconvolved residual curves (0.286) compared to perfusion parameter maps (0.287), while using features optimized from the source concentration-time curves (0.296) provided the best tissue outcome predictions.

Predicting treatment-specific lesion outcomes in acute ischemic stroke from 4D CT perfusion imaging using spatio-temporal convolutional neural networks

For the diagnosis and precise treatment of acute ischemic stroke, predicting the final location and volume of lesions is of great clinical interest. Current deep learning-based prediction methods mainly use perfusion parameter maps, which can be calculated from spatio-temporal (4D) CT perfusion (CTP) imaging data, to estimate the tissue outcome of an acute ischemic stroke. However, this calculation relies on a deconvolution operation, an ill-posed problem requiring strong regularization and definition of an arterial input function. Thus, improved predictions might be achievable if the deep learning models were applied directly to acute 4D CTP data rather than perfusion maps. In this work, a novel deep spatio-temporal convolutional neural network is proposed for predicting treatment-dependent stroke lesion outcomes by making full use of raw 4D CTP data. By merging a U-Net-like architecture with temporal convolutional networks, we efficiently process the spatio-temporal information available in CTP datasets to make a tissue outcome prediction. The proposed method was evaluated on 147 patients using a 10-fold cross validation, which demonstrated that the proposed 3D+time model (mean Dice=0.45) significantly outperforms both a 2D+time variant of our approach (mean Dice=0.43) and a state-of-the-art method that uses perfusion maps (mean Dice=0.38). These results show that 4D CTP datasets include more predictive information than perfusion parameter maps, and that the proposed method is an efficient approach to make use of this complex data.

Fairness-related performance and explainability effects in deep learning models for brain image analysis

Purpose: Explainability and fairness are two key factors for the effective and ethical clinical implementation of deep learning-based machine learning models in healthcare settings. However, there has been limited work on investigating how unfair performance manifests in explainable artificial intelligence (XAI) methods, and how XAI can be used to investigate potential reasons for unfairness. Thus, the aim of this work was to analyze the effects of previously established sociodemographic-related confounders on classifier performance and explainability methods. Approach: A convolutional neural network (CNN) was trained to predict biological sex from T1-weighted brain MRI datasets of 4547 9- to 10-year-old adolescents from the Adolescent Brain Cognitive Development study. Performance disparities of the trained CNN between White and Black subjects were analyzed and saliency maps were generated for each subgroup at the intersection of sex and race. Results: The classification model demonstrated a significant difference in the percentage of correctly classified White male ( 90.3 % ± 1.7 % ) and Black male ( 81.1 % ± 4.5 % ) children. Conversely, slightly higher performance was found for Black female ( 89.3 % ± 4.8 % ) compared with White female ( 86.5 % ± 2.0 % ) children. Saliency maps showed subgroup-specific differences, corresponding to brain regions previously associated with pubertal development. In line with this finding, average pubertal development scores of subjects used in this study were significantly different between Black and White females ( p < 0.001 ) and males ( p < 0.001 ). Conclusions: We demonstrate that a CNN with significantly different sex classification performance between Black and White adolescents can identify different important brain regions when comparing subgroup saliency maps. Importance scores vary substantially between subgroups within brain structures associated with pubertal development, a race-associated confounder for predicting sex. We illustrate that unfair models can produce different XAI results between subgroups and that these results may explain potential reasons for biased performance.

Sex Differences in Adult Facial Three-Dimensional Morphology: Application to Gender-Affirming Facial Surgery

Background: Gender-affirming facial surgery (GFS) is pursued by transgender individuals who desire facial features that better reflect their gender identity. Currently, there are a few objective guidelines to justify and facilitate effective surgical decision making. Objective: To quantify the effect of sex on adult facial size and shape through an analysis of three-dimensional (3D) facial surface images. Materials and Methods: Facial measurements were obtained by registering an atlas facial surface to 3D surface scans of 545 males and 1028 females older than 20 years of age. The differences between male and female faces were analyzed and visualized for a set of predefined surgically relevant facial regions. Results: On average, male faces are 7.3% larger than female faces (Cohen's D = 2.17). Sex is associated with significant facial shape differences (p < 0.0001) in the entire face as well as in each sub-region considered in this study. The facial regions in which sex has the largest effect on shape are the brow, jaw, nose, and cheek. Conclusions: These findings provide biologic data-driven anatomic guidance and justification for GFS, particularly forehead contouring cranioplasty, mandible and chin alterations, rhinoplasty, and cheek modifications.

Segmentation of Tissues and Proliferating Cells in Light-Sheet Microscopy Images of Mouse Embryos Using Convolutional Neural Networks

A variety of genetic mutations affect cell proliferation during organism development, leading to structural birth defects. However, the mechanisms by which these alterations influence the development of the face remain unclear. Cell proliferation and its relation to shape variation can be studied using Light-Sheet Microscopy (LSM) imaging across a range of developmental time points using mouse models. The aim of this work was to develop and evaluate accurate automatic methods based on convolutional neural networks (CNNs) for: (i) tissue segmentation (neural ectoderm and mesenchyme), (ii) cell segmentation in nuclear-stained images, and (iii) segmentation of proliferating cells in phospho-Histone H3 (pHH3)-stained LSM images of mouse embryos. For training and evaluation of the CNN models, 155 to 176 slices from 10 mouse embryo LSM images with corresponding manual segmentations were available depending on the segmentation task. Three U-net CNN models were trained optimizing their loss functions, among other hyper-parameters, depending on the segmentation task. The tissue segmentation achieved a macro-average F-score of 0.84, whereas the inter-observer value was 0.89. The cell segmentation achieved a Dice score of 0.57 and 0.56 for nuclear-stained and pHH3-stained images, respectively, whereas the corresponding inter-observer Dice scores were 0.39 and 0.45, respectively. The proposed pipeline using the U-net CNN architecture can accelerate LSM image analysis and together with the annotated datasets can serve as a reference for comparison of more advanced LSM image segmentation methods in future.

Invertible Modeling of Bidirectional Relationships in Neuroimaging With Normalizing Flows: Application to Brain Aging

Many machine learning tasks in neuroimaging aim at modeling complex relationships between a brain's morphology as seen in structural MR images and clinical scores and variables of interest. A frequently modeled process is healthy brain aging for which many image-based brain age estimation or age-conditioned brain morphology template generation approaches exist. While age estimation is a regression task, template generation is related to generative modeling. Both tasks can be seen as inverse directions of the same relationship between brain morphology and age. However, this view is rarely exploited and most existing approaches train separate models for each direction. In this paper, we propose a novel bidirectional approach that unifies score regression and generative morphology modeling and we use it to build a bidirectional brain aging model. We achieve this by defining an invertible normalizing flow architecture that learns a probability distribution of 3D brain morphology conditioned on age. The use of full 3D brain data is achieved by deriving a manifold-constrained formulation that models morphology variations within a low-dimensional subspace of diffeomorphic transformations. This modeling idea is evaluated on a database of MR scans of more than 5000 subjects. The evaluation results show that our bidirectional brain aging model (1) accurately estimates brain age, (2) is able to visually explain its decisions through attribution maps and counterfactuals, (3) generates realistic age-specific brain morphology templates, (4) supports the analysis of morphological variations, and (5) can be utilized for subject-specific brain aging simulation.

Intracranial Artery Morphology in Pediatric Moya Moya Disease and Moya Moya Syndrome

BACKGROUND

Moya Moya disease (MMD) and Moya Moya syndrome (MMS) are cerebrovascular disorders, which affect the internal carotid arteries (ICAs). Diagnosis and surveillance of MMD/MMS in children mostly rely on qualitative evaluation of vascular imaging, especially MR angiography (MRA).

OBJECTIVE

To quantitatively characterize arterial differences in pediatric patients with MMD/MMS compared with normal controls.

METHODS

MRA data sets from 17 presurgery MMD/MMS (10M/7F, mean age = 10.0 years) patients were retrospectively collected and compared with MRA data sets of 98 children with normal vessel morphology (49 male patients; mean age = 10.6 years). Using a level set segmentation method with anisotropic energy weights, the cerebral arteries were automatically extracted and used to compute the radius of the ICA, middle cerebral artery (MCA), anterior cerebral artery (ACA), posterior cerebral artery (PCA), and basilar artery (BA). Moreover, the density and the average radius of all arteries in the MCA, ACA, and PCA flow territories were quantified.

RESULTS

Statistical analysis revealed significant differences comparing children with MMD/MMS and those with normal vasculature (P < .001), whereas post hoc analyses identified significantly smaller radii of the ICA, MCA-M1, MCA-M2, and ACA (P < .001) in the MMD/MMS group. No significant differences were found for the radii of the PCA and BA or any artery density and average artery radius measurement in the flow territories (P > .05).

CONCLUSION

His study describes the results of an automatic approach for quantitative characterization of the cerebrovascular system in patients with MMD/MMS with promising preliminary results for quantitative surveillance in pediatric MMD/MMS management.

An exploratory causal analysis of the relationships between the brain age gap and cardiovascular risk factors

The brain age gap (BAG) has been shown to capture accelerated brain aging patterns and might serve as a biomarker for several neurological diseases. Moreover, it was also shown that it captures other biological information related to modifiable cardiovascular risk factors. Previous studies have explored statistical relationships between the BAG and cardiovascular risk factors. However, none of those studies explored causal relationships between the BAG and cardiovascular risk factors. In this work, we employ causal structure discovery techniques and define a Bayesian network to model the assumed causal relationships between the BAG, estimated using morphometric T1-weighted magnetic resonance imaging brain features from 2025 adults, and several cardiovascular risk factors. This setup allows us to not only assess observed conditional probability distributions of the BAG given cardiovascular risk factors, but also to isolate the causal effect of each cardiovascular risk factor on BAG using causal inference. Results demonstrate the feasibility of the proposed causal analysis approach by illustrating intuitive causal relationships between variables. For example, body-mass-index, waist-to-hip ratio, smoking, and alcohol consumption were found to impact the BAG, with the greatest impact for obesity markers resulting in higher chances of developing accelerated brain aging. Moreover, the findings show that causal effects differ from correlational effects, demonstrating the importance of accounting for variable relationships and confounders when evaluating the information captured by a biomarker. Our work demonstrates the feasibility and advantages of using causal analyses instead of purely correlation-based and univariate statistical analyses in the context of brain aging and related problems.

Early Onset Diffusion Abnormalities in Refractory Headache Disorders

Objective

This study sought to determine if individuals with medically refractory migraine headache have volume or diffusion abnormalities on neuroimaging compared to neurotypical individuals.

Background

Neuroimaging biomarkers in headache medicine continue to be limited. Early prediction of medically refractory headache and migraine disorders could result in earlier administration of high efficacy therapeutics.

Methods

A single-center, retrospective, case control study was performed. All patients were evaluated clinically between 2014 and 2018. Individuals with medically refractory migraine headache (defined by ICDH-3 criteria) without any other chronic medical diseases were enrolled. Patients had to have failed more than two therapeutics and aura was not exclusionary. The initial MRI study for each patient was reviewed. Multiple brain regions were analyzed for volume and apparent diffusion coefficient values. These were compared to 81 neurotypical control patients.

Results

A total of 79 patients with medically refractory migraine headache were included and compared to 74 neurotypical controls without headache disorders. Time between clinical diagnosis and neuroimaging was a median of 24 months (IQR: 12.0–37.0). Comparison of individuals with medically refractory migraine headache to controls revealed statistically significant differences in median apparent diffusion coefficient (ADC) in multiple brain subregions (p < 0.001). Post-hoc pair-wise analysis comparing individuals with medically refractory migraine headache to control patients revealed significantly decreased median ADC values for the thalamus, caudate, putamen, pallidum, amygdala, brainstem, and cerebral white matter. No volumetric differences were observed between groups.

Conclusions

In individuals with medically refractory MH, ADC changes are measurable in multiple brain structures at an early age, prior to the failure of multiple pharmacologic interventions and the diagnosis of medically refractory MH. This data supports the hypothesis that structural connectivity issues may predispose some patients toward more medically refractory pain disorders such as MH.

Association of stroke lesion shape with newly detected atrial fibrillation – Results from the MonDAFIS study

Paroxysmal Atrial fibrillation (AF) is often clinically silent and may be missed by the usual diagnostic workup after ischemic stroke. We aimed to determine whether shape characteristics of ischemic stroke lesions can be used to predict AF in stroke patients without known AF at baseline. Lesion shape quantification on brain MRI was performed in selected patients from the intervention arm of the Impact of standardized MONitoring for Detection of Atrial Fibrillation in Ischemic Stroke (MonDAFIS) study, which included patients with ischemic stroke or TIA without prior AF. Multiple morphologic parameters were calculated based on lesion segmentation in acute brain MRI data. Multivariate logistic models were used to test the association of lesion morphology, clinical parameters, and AF. A stepwise elimination regression was conducted to identify the most important variables. A total of 755 patients were included. Patients with AF detected within 2 years after stroke (n = 86) had a larger overall oriented bounding box (OBB) volume (p = 0.003) and a higher number of brain lesion components (p = 0.008) than patients without AF. In the multivariate model, OBB volume (OR 1.72, 95%CI 1.29–2.35, p < 0.001), age (OR 2.13, 95%CI 1.52–3.06, p < 0.001), and female sex (OR 2.45, 95%CI 1.41–4.31, p = 0.002) were independently associated with detected AF. Ischemic lesions in patients with detected AF after stroke presented with a more dispersed infarct pattern and a higher number of lesion components. Together with clinical characteristics, these lesion shape characteristics may help in guiding prolonged cardiac monitoring after stroke.

A Deep Invertible 3D Facial Shape Model For Interpretable Genetic Syndrome Diagnosis

One of the primary difficulties in treating patients with genetic syndromes is diagnosing their condition. Many syndromes are associated with characteristic facial features that can be imaged and utilized by computer-assisted diagnosis systems. In this work, we develop a novel 3D facial surface modeling approach with the objective of maximizing diagnostic model interpretability within a flexible deep learning framework. Therefore, an invertible normalizing flow architecture is introduced to enable both inferential and generative tasks in a unified and efficient manner. The proposed model can be used (1) to infer syndrome diagnosis and other demographic variables given a 3D facial surface scan and (2) to explain model inferences to non-technical users via multiple interpretability mechanisms. The model was trained and evaluated on more than 4700 facial surface scans from subjects with 47 different syndromes. For the challenging task of predicting syndrome diagnosis given a new 3D facial surface scan, age, and sex of a subject, the model achieves a competitive overall top-1 accuracy of 71%, and a mean sensitivity of 43% across all syndrome classes. We believe that invertible models such as the one presented in this work can achieve competitive inferential performance while greatly increasing model interpretability in the domain of medical diagnosis.

Localized Statistical Shape Models for Large-scale Problems With Few Training Data

OBJECTIVE

Statistical shape models have been successfully used in numerous biomedical image analysis applications where prior shape information is helpful such as organ segmentation or data augmentation when training deep learning models. However, training such models requires large data sets, which are often not available and, hence, shape models frequently fail to represent local details of unseen shapes. This work introduces a kernel-based method to alleviate this problem via so-called model localization. It is specifically designed to be used in large-scale shape modeling scenarios like deep learning data augmentation and fits seamlessly into the classical shape modeling framework.

METHOD

Relying on recent advances in multi-level shape model localization via distance-based covariance matrix manipulations and Grassmannian-based level fusion, this work proposes a novel and computationally efficient kernel-based localization technique. Moreover, a novel way to improve the specificity of such models via normalizing flow-based density estimation is presented.

RESULTS

The method is evaluated on the publicly available JSRT/SCR chest X-ray and IXI brain data sets. The results confirm the effectiveness of the kernelized formulation and also highlight the models' improved specificity when utilizing the proposed density estimation method.

CONCLUSION

This work shows that flexible and specific shape models from few training samples can be generated in a computationally efficient way by combining ideas from kernel theory and normalizing flows.

SIGNIFICANCE

The proposed method together with its publicly available implementation allows to build shape models from few training samples directly usable for applications like data augmentation.

Multimodal biological brain age prediction using magnetic resonance imaging and angiography with the identification of predictive regions

Biological brain age predicted using machine learning models based on high-resolution imaging data has been suggested as a potential biomarker for neurological and cerebrovascular diseases. In this work, we aimed to develop deep learning models to predict the biological brain age using structural magnetic resonance imaging and angiography datasets from a large database of 2074 adults (21-81 years). Since different imaging modalities can provide complementary information, combining them might allow to identify more complex aging patterns, with angiography data, for instance, showing vascular aging effects complementary to the atrophic brain tissue changes seen in T1-weighted MRI sequences. We used saliency maps to investigate the contribution of cortical, subcortical, and arterial structures to the prediction. Our results show that combining T1-weighted and angiography MR data led to a significantly improved brain age prediction accuracy, with a mean absolute error of 3.85 years comparing the predicted and chronological age. The most predictive brain regions included the lateral sulcus, the fourth ventricle, and the amygdala, while the brain arteries contributing the most to the prediction included the basilar artery, the middle cerebral artery M2 segments, and the left posterior cerebral artery. Our study proposes a framework for brain age prediction using multimodal imaging, which gives accurate predictions and allows identifying the most predictive regions for this task, which can serve as a surrogate for the brain regions that are most affected by aging.

A systematic comparison of generative models for medical images

Purpose

This work aims for a systematic comparison of popular shape and appearance models. Here, two statistical and four deep-learning-based shape and appearance models are compared and evaluated in terms of their expressiveness described by their generalization ability and specificity as well as further properties like input data format, interpretability and latent space distribution and dimension.

Methods

Classical shape models and their locality-based extension are considered next to autoencoders, variational autoencoders, diffeomorphic autoencoders and generative adversarial networks. The approaches are evaluated in terms of generalization ability, specificity and likeness depending on the amount of training data. Furthermore, various latent space metrics are presented in order to capture further major characteristics of the models.

Results

The experimental setup showed that locality statistical shape models yield best results in terms of generalization ability for 2D and 3D shape modeling. However, the deep learning approaches show strongly improved specificity. In the case of simultaneous shape and appearance modeling, the neural networks are able to generate more realistic and diverse appearances. A major drawback of the deep-learning models is, however, their impaired interpretability and ambiguity of the latent space.

Conclusions

It can be concluded that for applications not requiring particularly good specificity, shape modeling can be reliably established with locality-based statistical shape models, especially when it comes to 3D shapes. However, deep learning approaches are more worthwhile in terms of appearance modeling.