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.

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.

An Analysis of the Vulnerability of Two Common Deep Learning-Based Medical Image Segmentation Techniques to Model Inversion Attacks.

Recent research in computer vision has shown that original images used for training of deep learning models can be reconstructed using so-called inversion attacks. However, the feasibility of this attack type has not been investigated for complex 3D medical images. Thus, the aim of this study was to examine the vulnerability of deep learning techniques used in medical imaging to model inversion attacks and investigate multiple quantitative metrics to evaluate the quality of the reconstructed images. For the development and evaluation of model inversion attacks, the public LPBA40 database consisting of 40 brain MRI scans with corresponding segmentations of the gyri and deep grey matter brain structures were used to train two popular deep convolutional neural networks, namely a U-Net and SegNet, and corresponding inversion decoders. Matthews correlation coefficient, the structural similarity index measure (SSIM), and the magnitude of the deformation field resulting from non-linear registration of the original and reconstructed images were used to evaluate the reconstruction accuracy. A comparison of the similarity metrics revealed that the SSIM is best suited to evaluate the reconstruction accuray, followed closely by the magnitude of the deformation field. The quantitative evaluation of the reconstructed images revealed SSIM scores of 0.73±0.12 and 0.61±0.12 for the U-Net and the SegNet, respectively. The qualitative evaluation showed that training images can be reconstructed with some degradation due to blurring but can be correctly matched to the original images in the majority of the cases. In conclusion, the results of this study indicate that it is possible to reconstruct patient data used for training of convolutional neural networks and that the SSIM is a good metric to assess the reconstruction accuracy.

Semantically Guided Large Deformation Estimation with Deep Networks.

Deformable image registration is still a challenge when the considered images have strong variations in appearance and large initial misalignment. A huge performance gap currently remains for fast-moving regions in videos or strong deformations of natural objects. We present a new semantically guided and two-step deep deformation network that is particularly well suited for the estimation of large deformations. We combine a U-Net architecture that is weakly supervised with segmentation information to extract semantically meaningful features with multiple stages of nonrigid spatial transformer networks parameterized with low-dimensional B-spline deformations. Combining alignment loss and semantic loss functions together with a regularization penalty to obtain smooth and plausible deformations, we achieve superior results in terms of alignment quality compared to previous approaches that have only considered a label-driven alignment loss. Our network model advances the state of the art for inter-subject face part alignment and motion tracking in medical cardiac magnetic resonance imaging (MRI) sequences in comparison to the FlowNet and Label-Reg, two recent deep-learning registration frameworks. The models are compact, very fast in inference, and demonstrate clear potential for a variety of challenging tracking and/or alignment tasks in computer vision and medical image analysis.

Fully Automatic Landmarking of Syndromic 3D Facial Surface Scans Using 2D Images.

3D facial landmarks are known to be diagnostically relevant biometrics for many genetic syndromes. The objective of this study was to extend a state-of-the-art image-based 2D facial landmarking algorithm for the challenging task of 3D landmark identification on subjects with genetic syndromes, who often have moderate to severe facial dysmorphia. The automatic 3D facial landmarking algorithm presented here uses 2D image-based facial detection and landmarking models to identify 12 landmarks on 3D facial surface scans. The landmarking algorithm was evaluated using a test set of 444 facial scans with ground truth landmarks identified by two different human observers. Three hundred and sixty nine of the subjects in the test set had a genetic syndrome that is associated with facial dysmorphology. For comparison purposes, the manual landmarks were also used to initialize a non-linear surface-based registration of a non-syndromic atlas to each subject scan. Compared to the average intra- and inter-observer landmark distances of 1.1 mm and 1.5 mm respectively, the average distance between the manual landmark positions and those produced by the automatic image-based landmarking algorithm was 2.5 mm. The average error of the registration-based approach was 3.1 mm. Comparing the distributions of Procrustes distances from the mean for each landmarking approach showed that the surface registration algorithm produces a systemic bias towards the atlas shape. In summary, the image-based automatic landmarking approach performed well on this challenging test set, outperforming a semi-automatic surface registration approach, and producing landmark errors that are comparable to state-of-the-art 3D geometry-based facial landmarking algorithms evaluated on non-syndromic subjects.

Foundation model-driven distributed learning for enhanced retinal age prediction.

OBJECTIVES: The retinal age gap (RAG) is emerging as a potential biomarker for various diseases of the human body, yet its utility depends on machine learning models capable of accurately predicting biological retinal age from fundus images. However, training generalizable models is hindered by potential shortages of diverse training data. To overcome these obstacles, this work develops a novel and computationally efficient distributed learning framework for retinal age prediction. MATERIALS AND METHODS: The proposed framework employs a memory-efficient 8-bit quantized version of RETFound, a cutting-edge foundation model for retinal image analysis, to extract features from fundus images. These features are then used to train an efficient linear regression head model for predicting retinal age. The framework explores federated learning (FL) as well as traveling model (TM) approaches for distributed training of the linear regression head. To evaluate this framework, we simulate a client network using fundus image data from the UK Biobank. Additionally, data from patients with type 1 diabetes from the UK Biobank and the Brazilian Multilabel Ophthalmological Dataset (BRSET) were utilized to explore the clinical utility of the developed methods. RESULTS: Our findings reveal that the developed distributed learning framework achieves retinal age prediction performance on par with centralized methods, with FL and TM providing similar performance (mean absolute error of 3.57 ± 0.18 years for centralized learning, 3.60 ± 0.16 years for TM, and 3.63 ± 0.19 years for FL). Notably, the TM was found to converge with fewer local updates than FL. Moreover, patients with type 1 diabetes exhibited significantly higher RAG values than healthy controls in all models, for both the UK Biobank and BRSET datasets (P < .001). DISCUSSION: The high computational and memory efficiency of the developed distributed learning framework makes it well suited for resource-constrained environments. CONCLUSION: The capacity of this framework to integrate data from underrepresented populations for training of retinal age prediction models could significantly enhance the accessibility of the RAG as an important disease biomarker.

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.

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.

Towards objective and systematic evaluation of bias in artificial intelligence for medical imaging.

OBJECTIVE: Artificial intelligence (AI) models trained using medical images for clinical tasks often exhibit bias in the form of subgroup performance disparities. However, since not all sources of bias in real-world medical imaging data are easily identifiable, it is challenging to comprehensively assess their impacts. In this article, we introduce an analysis framework for systematically and objectively investigating the impact of biases in medical images on AI models. MATERIALS AND METHODS: Our framework utilizes synthetic neuroimages with known disease effects and sources of bias. We evaluated the impact of bias effects and the efficacy of 3 bias mitigation strategies in counterfactual data scenarios on a convolutional neural network (CNN) classifier. RESULTS: The analysis revealed that training a CNN model on the datasets containing bias effects resulted in expected subgroup performance disparities. Moreover, reweighing was the most successful bias mitigation strategy for this setup. Finally, we demonstrated that explainable AI methods can aid in investigating the manifestation of bias in the model using this framework. DISCUSSION: The value of this framework is showcased in our findings on the impact of bias scenarios and efficacy of bias mitigation in a deep learning model pipeline. This systematic analysis can be easily expanded to conduct further controlled in silico trials in other investigations of bias in medical imaging AI. CONCLUSION: Our novel methodology for objectively studying bias in medical imaging AI can help support the development of clinical decision-support tools that are robust and responsible.

Analysis and visualization of the effect of multiple sclerosis on biological brain age.

Introduction: The rate of neurodegeneration in multiple sclerosis (MS) is an important biomarker for disease progression but can be challenging to quantify. The brain age gap, which quantifies the difference between a patient's chronological and their estimated biological brain age, might be a valuable biomarker of neurodegeneration in patients with MS. Thus, the aim of this study was to investigate the value of an image-based prediction of the brain age gap using a deep learning model and compare brain age gap values between healthy individuals and patients with MS. Methods: A multi-center dataset consisting of 5,294 T1-weighted magnetic resonance images of the brain from healthy individuals aged between 19 and 89 years was used to train a convolutional neural network (CNN) for biological brain age prediction. The trained model was then used to calculate the brain age gap in 195 patients with relapsing remitting MS (20-60 years). Additionally, saliency maps were generated for healthy subjects and patients with MS to identify brain regions that were deemed important for the brain age prediction task by the CNN. Results: Overall, the application of the CNN revealed accelerated brain aging with a larger brain age gap for patients with MS with a mean of 6.98 ± 7.18 years in comparison to healthy test set subjects (0.23 ± 4.64 years). The brain age gap for MS patients was weakly to moderately correlated with age at disease onset (ρ = -0.299, p < 0.0001), EDSS score (ρ = 0.206, p = 0.004), disease duration (ρ = 0.162, p = 0.024), lesion volume (ρ = 0.630, p < 0.0001), and brain parenchymal fraction (ρ = -0.718, p < 0.0001). The saliency maps indicated significant differences in the lateral ventricle (p < 0.0001), insula (p < 0.0001), third ventricle (p < 0.0001), and fourth ventricle (p = 0.0001) in the right hemisphere. In the left hemisphere, the inferior lateral ventricle (p < 0.0001) and the third ventricle (p < 0.0001) showed significant differences. Furthermore, the Dice similarity coefficient showed the highest overlap of salient regions between the MS patients and the oldest healthy subjects, indicating that neurodegeneration is accelerated in this patient cohort. Discussion: In conclusion, the results of this study show that the brain age gap is a valuable surrogate biomarker to measure disease progression in patients with multiple sclerosis.

HarmonyTM: multi-center data harmonization applied to distributed learning for Parkinson's disease classification.

Purpose: Distributed learning is widely used to comply with data-sharing regulations and access diverse datasets for training machine learning (ML) models. The traveling model (TM) is a distributed learning approach that sequentially trains with data from one center at a time, which is especially advantageous when dealing with limited local datasets. However, a critical concern emerges when centers utilize different scanners for data acquisition, which could potentially lead models to exploit these differences as shortcuts. Although data harmonization can mitigate this issue, current methods typically rely on large or paired datasets, which can be impractical to obtain in distributed setups. Approach: We introduced HarmonyTM, a data harmonization method tailored for the TM. HarmonyTM effectively mitigates bias in the model's feature representation while retaining crucial disease-related information, all without requiring extensive datasets. Specifically, we employed adversarial training to "unlearn" bias from the features used in the model for classifying Parkinson's disease (PD). We evaluated HarmonyTM using multi-center three-dimensional (3D) neuroimaging datasets from 83 centers using 23 different scanners. Results: Our results show that HarmonyTM improved PD classification accuracy from 72% to 76% and reduced (unwanted) scanner classification accuracy from 53% to 30% in the TM setup. Conclusion: HarmonyTM is a method tailored for harmonizing 3D neuroimaging data within the TM approach, aiming to minimize shortcut learning in distributed setups. This prevents the disease classifier from leveraging scanner-specific details to classify patients with or without PD-a key aspect for deploying ML models for clinical applications.

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.

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.

Cerebroventricular deformation and vector mapping, a topographic visualizer for surgical interventions in pediatric hydrocephalus.

OBJECTIVE: Hydrocephalus is a challenging neurosurgical condition due to nonspecific symptoms and complex brain-fluid pressure dynamics. Typically, the assessment of hydrocephalus in children requires radiographic or invasive pressure monitoring. There is usually a qualitative focus on the ventricular spaces even though stress and shear forces extend across the brain. Here, the authors present an MRI-based vector approach for voxelwise brain and ventricular deformation visualization and analysis. METHODS: Twenty pediatric patients (mean age 7.7 years, range 6 months-18 years; 14 males) with acute, newly diagnosed hydrocephalus requiring surgical intervention for symptomatic relief were randomly identified after retrospective chart review. Selection criteria included acquisition of both pre- and posttherapy paired 3D T1-weighted volumetric MRI (3D T1-MRI) performed on 3T MRI systems. Both pre- and posttherapy 3D T1-MRI pairs were aligned using image registration, and subsequently, voxelwise nonlinear transformations were performed to derive two exemplary visualizations of compliance: 1) a whole-brain vector map projecting the resulting deformation field on baseline axial imaging; and 2) a 3D heat map projecting the volumetric changes along ventricular boundaries and the brain periphery. RESULTS: The patients underwent the following interventions for treatment of hydrocephalus: endoscopic third ventriculostomy (n = 6); external ventricular drain placement and/or tumor resection (n = 10); or ventriculoperitoneal shunt placement (n = 4). The mean time between pre- and postoperative imaging was 36.5 days. Following intervention, the ventricular volumes decreased significantly (mean pre- and posttherapy volumes of 151.9 cm3 and 82.0 cm3, respectively; p < 0.001, paired t-test). The largest degree of deformation vector changes occurred along the lateral ventricular spaces, relative to the genu and splenium. There was a significant correlation between change in deformation vector magnitudes within the cortical layer and age (p = 0.011, Pearson), as well as between the ventricle size and age (p = 0.014, Pearson), suggesting higher compliance among infants and younger children. CONCLUSIONS: This study highlights an approach for deformation analysis and vector mapping that may serve as a topographic visualizer for therapeutic interventions in patients with hydrocephalus. A future study that correlates the degree of cerebroventricular deformation or compliance with intracranial pressures could clarify the potential role of this technique in noninvasive pressure monitoring or in cases of noncompliant ventricles.

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.

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.

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.

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.

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.

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.