arXiv:2511.14588v1 Announce Type: new 
Abstract: White matter hyperintensities (WMH) are key imaging markers in cognitive aging, Alzheimer's disease (AD), and related dementias. Although automated methods for WMH segmentation have advanced, most provide only global lesion load and overlook their spatial distribution across distinct white matter regions. We propose a deep learning framework for robust WMH segmentation and localization, evaluated across public datasets and an independent Alzheimer's Disease Neuroimaging Initiative (ADNI) cohort. Our results show that the predicted lesion loads are in line with the reference WMH estimates, confirming the robustness to variations in lesion load, acquisition, and demographics. Beyond accurate segmentation, we quantify WMH load within anatomically defined regions and combine these measures with brain structure volumes to assess diagnostic value. Regional WMH volumes consistently outperform global lesion burden for disease classification, and integration with brain atrophy metrics further improves performance, reaching area under the curve (AUC) values up to 0.97. Several spatially distinct regions, particularly within anterior white matter tracts, are reproducibly associated with diagnostic status, indicating localized vulnerability in AD. These results highlight the added value of regional WMH quantification. Incorporating localized lesion metrics alongside atrophy markers may enhance early diagnosis and stratification in neurodegenerative disorders.

تُعتبر الزيادة في كثافة المادة البيضاء (WMH) علامات تصويرية رئيسية في الشيخوخة الإدراكية ومرض الزهايمر (AD). على الرغم من التقدم في الأساليب الآلية لتجزئة WMH، فإن معظمها يوفر فقط عبء آفة عالمي ويتجاهل توزيعها المكاني عبر مناطق المادة البيضاء المختلفة. تم اقتراح إطار تعلم عميق لتجزئة WMH وتحديد موقعها بشكل موثوق، تم تقييمه عبر مجموعات بيانات عامة وعيّنة مستقلة من مبادرة تصوير مرض الزهايمر العصبي (ADNI). تُظهر النتائج أن أحمال الآفات المتوقعة تتماشى مع تقديرات WMH المرجعية، مما يؤكد القوة في مواجهة التغيرات في عبء الآفة، والاكتساب، والديموغرافيا.

Las hiperintensidades de la materia blanca (WMH) son marcadores de imagen clave en el envejecimiento cognitivo y la enfermedad de Alzheimer (EA). A pesar de los avances en los métodos automatizados para la segmentación de WMH, la mayoría solo proporcionan una carga de lesión global y pasan por alto su distribución espacial en distintas regiones de la materia blanca. Se propone un nuevo marco de aprendizaje profundo para la segmentación y localización robustas de WMH, evaluado en conjuntos de datos públicos y en una cohorte independiente de la Iniciativa de Neuroimagen de la Enfermedad de Alzhe…

Les hyperintensités de la matière blanche (WMH) sont des marqueurs d'imagerie importants dans le vieillissement cognitif et la maladie d'Alzheimer (MA). Malgré les avancées des méthodes automatisées de segmentation des WMH, la plupart ne fournissent qu'une charge lésionnelle globale et négligent la distribution spatiale dans les régions de matière blanche. Un nouveau cadre d'apprentissage profond a été proposé pour la segmentation et la localisation robustes des WMH, validé sur des ensembles de données publics et un cohort indépendant de l'Initiative d'Imagerie Neurodégénérative de la Maladie …

White matter hyperintensities (WMH) are significant imaging markers in cognitive aging and Alzheimer's disease (AD). Despite advancements in automated WMH segmentation, most methods only provide global lesion load and neglect spatial distribution across white matter regions. A new deep learning framework has been proposed for robust WMH segmentation and localization, validated on public datasets and an independent Alzheimer's Disease Neuroimaging Initiative (ADNI) cohort. The study indicates that regional WMH volumes are more effective than global lesion burden for disease classification.

Deep Learning-Based Regional White Matter Hyperintensity Mapping as a Robust Biomarker for Alzheimer's Disease

arXiv:2511.02558v2 Announce Type: replace-cross 
Abstract: Predicting future brain state from a baseline magnetic resonance image (MRI) is a central challenge in neuroimaging and has important implications for studying neurodegenerative diseases such as Alzheimer's disease (AD). Most existing approaches predict future cognitive scores or clinical outcomes, such as conversion from mild cognitive impairment to dementia. Instead, here we investigate longitudinal MRI image-to-image prediction that forecasts a participant's entire brain MRI several years into the future, intrinsically modeling complex, spatially distributed neurodegenerative patterns. We implement and evaluate five deep learning architectures (UNet, U2-Net, UNETR, Time-Embedding UNet, and ODE-UNet) on two longitudinal cohorts (ADNI and AIBL). Predicted follow-up MRIs are directly compared with the actual follow-up scans using metrics that capture global similarity and local differences. The best performing models achieve high-fidelity predictions, and all models generalize well to an independent external dataset, demonstrating robust cross-cohort performance. Our results indicate that deep learning can reliably predict participant-specific brain MRI at the voxel level, offering new opportunities for individualized prognosis.

طور الباحثون طريقة للتنبؤ بالحالات الدماغية المستقبلية باستخدام صور الرنين المغناطيسي الطولية، مع التركيز على الأنماط العصبية التنكسية المرتبطة بمرض الزهايمر. تستخدم هذه الطريقة خمس هياكل تعلم عميق للتنبؤ بصورة الرنين المغناطيسي للدماغ لدى المشاركين بعد عدة سنوات، مما يوفر رؤى حول تقدم تدهور الإدراك.

Investigadores han desarrollado un método para predecir futuros estados cerebrales utilizando escaneos de MRI longitudinales, centrándose en los patrones neurodegenerativos asociados con la enfermedad de Alzheimer. Este enfoque utiliza cinco arquitecturas de aprendizaje profundo para pronosticar la MRI cerebral de un participante varios años en el futuro, proporcionando información sobre la progresión del deterioro cognitivo.

Des chercheurs ont développé une méthode pour prédire les futurs états cérébraux à l'aide de scans IRM longitudinaux, en se concentrant sur les schémas neurodégénératifs associés à la maladie d'Alzheimer. Cette approche utilise cinq architectures d'apprentissage profond pour prévoir l'IRM cérébrale d'un participant plusieurs années à l'avance, fournissant des informations sur la progression du déclin cognitif.

Researchers have developed a method for predicting future brain states using longitudinal MRI scans, focusing on neurodegenerative patterns associated with Alzheimer's disease. This approach utilizes five deep learning architectures to forecast a participant's brain MRI several years ahead, providing insights into the progression of cognitive impairment.

Deep Learning-Based Regional White Matter Hyperintensity Mapping as a Robust Biomarker for Alzheimer's Disease

Was this article worth reading? Share it

Keywords AI

Zemith-3bda3b

WasItAI