Artificial Intelligence and Multiple Sclerosis

Current Neurology and Neuroscience Reports, Год журнала: 2024, Номер 24(8), С. 233 - 243

Опубликована: Июнь 28, 2024

Abstract In this paper, we analyse the different advances in artificial intelligence (AI) approaches multiple sclerosis (MS). AI applications MS range across investigation of disease pathogenesis, diagnosis, treatment, and prognosis. A subset AI, Machine learning (ML) models various data sources, including magnetic resonance imaging (MRI), genetic, clinical data, to distinguish from other conditions, predict progression, personalize treatment strategies. Additionally, have been extensively applied lesion segmentation, identification biomarkers, prediction outcomes, monitoring, management. Despite big promises solutions, model interpretability transparency remain critical for gaining clinician patient trust these methods. The future holds potential open initiatives that could feed ML increasing generalizability, implementation federated solutions training addressing sharing issues, generative address challenges interpretability, transparency. conclusion, presents an opportunity advance our understanding management MS. aid clinicians diagnosis prognosis improving outcomes quality life, however ensuring AI-generated results is going be key facilitating integration into practice.

Язык: Английский

---

Machine learning for refining interpretation of magnetic resonance imaging scans in the management of multiple sclerosis: a narrative review

Royal Society Open Science, Год журнала: 2025, Номер 12(1)

Опубликована: Янв. 1, 2025

Multiple sclerosis (MS) is an autoimmune disease of the brain and spinal cord with both inflammatory neurodegenerative features. Although advances in imaging techniques, particularly magnetic resonance (MRI), have improved process diagnosis, its cause unknown, a cure remains elusive evidence base to guide treatment lacking. Computational techniques like machine learning (ML) started be used understand MS. Published MS MRI-based computational studies can divided into five categories: automated diagnosis; differentiation between lesion types and/or stages; differential monitoring predicting progression; synthetic MRI dataset generation. Collectively, these approaches show promise assisting activity prediction future progression, all potentially contributing management. Analysis quality using ML highly dependent on size variability for training. Wider public access would mean larger datasets experimentation, resulting higher-quality analysis, permitting more conclusive research. This narrative review provides outline fundamentals pathology pathogenesis, diagnostic data as well collating literature pertaining application towards developing better understanding

Язык: Английский

---

Multiple Sclerosis Lesion Analysis in Brain Magnetic Resonance Images: Techniques and Clinical Applications

IEEE Journal of Biomedical and Health Informatics, Год журнала: 2022, Номер 26(6), С. 2680 - 2692

Опубликована: Фев. 16, 2022

Multiple sclerosis (MS) is a chronic inflammatory and degenerative disease of the central nervous system, characterized by appearance focal lesions in white gray matter that topographically correlate with an individual patient's neurological symptoms signs. Magnetic resonance imaging (MRI) provides detailed in-vivo structural information, permitting quantification categorization MS critically inform management. Traditionally, have been manually annotated on 2D MRI slices, process inefficient prone to inter-/intra-observer errors. Recently, automated statistical analysis techniques proposed detect segment based voxel intensity. However, their effectiveness limited heterogeneity both data acquisition lesions. By learning complex lesion representations directly from images, deep achieved remarkable breakthroughs segmentation task. Here, we provide comprehensive review state-of-the-art automatic deep-learning methods discuss current future clinical applications. Further, technical strategies, such as domain adaptation, enhance real-world settings.

Язык: Английский

---

Research and application progress of radiomics in neurodegenerative diseases

Meta-Radiology, Год журнала: 2024, Номер 2(1), С. 100068 - 100068

Опубликована: Фев. 22, 2024

Neurodegenerative diseases refer to degenerative of the nervous system caused by neuronal degeneration and apoptosis. Usually, onset disease is insidious, progression slow, which can last for several years decades. Clinical symptoms only appear in later stages pathological changes when degree nerve cell loss reaches or exceeds a certain threshold. Traditional electrophysiological medical imaging techniques lack valuable indicators markers. Therefore, early diagnosis differentiation are very difficult. Radiomics new technology merged recent years, extract large number invisible features from raw image data with high throughput, quantitatively analyze physiological changes. It demonstrates important potential value diagnosis, grading, prognosis evaluation NDs. This review provides an overview research progress radiomics neurodegenerative diseases, emphasizing process principles its application classification, prediction these diseases. helps deepen understanding promote treatment clinical practice.

Язык: Английский

---

Multiple sclerosis lesion segmentation: revisiting weighting mechanisms for federated learning

Frontiers in Neuroscience, Год журнала: 2023, Номер 17

Опубликована: Май 18, 2023

Background and introduction Federated learning (FL) has been widely employed for medical image analysis to facilitate multi-client collaborative without sharing raw data. Despite great success, FL's applications remain suboptimal in neuroimage tasks such as lesion segmentation multiple sclerosis (MS), due variance characteristics imparted by different scanners acquisition parameters. Methods In this work, we propose the first FL MS framework via two effective re-weighting mechanisms. Specifically, a learnable weight is assigned each local node during aggregation process, based on its performance. addition, loss function client also re-weighted according volume data training. Results The proposed method validated scenarios using public clinical datasets. case-wise voxel-wise Dice score of under dataset 65.20 74.30, respectively. On second in-house dataset, 53.66, 62.31, Discussions conclusions Comparison experiments datasets have demonstrated effectiveness significantly outperforming other methods. Furthermore, performance incorporating our mechanism can achieve comparable that from centralized training with all

Язык: Английский

---

A Radiomic “Warning Sign” of Progression on Brain MRI in Individuals with MS

American Journal of Neuroradiology, Год журнала: 2024, Номер 45(2), С. 236 - 243

Опубликована: Янв. 12, 2024

MS is a chronic progressive, idiopathic, demyelinating disorder whose diagnosis contingent on the interpretation of MR imaging. New imaging lesions are an early biomarker disease progression. We aimed to evaluate machine learning model based radiomics features in predicting progression brain individuals with MS.

Язык: Английский

---

Predicting multiple sclerosis disease progression and outcomes with machine learning and MRI-based biomarkers: a review

Journal of Neurology, Год журнала: 2024, Номер unknown

Опубликована: Сен. 12, 2024

Multiple sclerosis (MS) is a demyelinating neurological disorder with highly heterogeneous clinical presentation and course of progression. Disease-modifying therapies are the only available treatment, as there no known cure for disease. Careful selection suitable necessary, they can be accompanied by serious risks adverse effects such infection. Magnetic resonance imaging (MRI) plays central role in diagnosis management MS, though MRI lesions have displayed moderate associations MS outcomes, clinico-radiological paradox. With advent machine learning (ML) healthcare, predictive power improved leveraging both traditional advanced ML algorithms capable analyzing increasingly complex patterns within neuroimaging data. The purpose this review was to examine application MRI-based prediction disease Studies were divided into five main categories: predicting conversion clinically isolated syndrome cognitive outcome, EDSS-related disability, motor disability activity. performance models discussed along highlighting influential MRI-derived biomarkers. Overall, presents promising avenue prognosis. However, integration biomarkers other multimodal patient data shows great potential advancing personalized healthcare approaches MS.

Язык: Английский

---

Time-Dependent Deep Learning Prediction of Multiple Sclerosis Disability

Deleted Journal, Год журнала: 2024, Номер unknown

Опубликована: Июнь 13, 2024

The majority of deep learning models in medical image analysis concentrate on single snapshot timepoint circumstances, such as the identification current pathology a given or volume. This is often contrast to diagnostic methodology radiology where presumed pathologic findings are correlated prior studies and subsequent changes over time. For multiple sclerosis (MS), body literature describes various forms lesion segmentation with few analyzing disability progression purpose longitudinal time-dependent analysis, we propose combinatorial video vision transformer (ViViT) benchmarked against traditional recurrent neural network Convolutional Neural Network-Long Short-Term Memory (CNN-LSTM) architectures hybrid Vision Transformer-LSTM (ViT-LSTM) predict long-term based upon Extended Disability Severity Score (EDSS). patient cohort was procured from two-site institution 703 patients' multisequence, contrast-enhanced MRIs cervical spine between years 2002 2023. Following competitive performance VGG-16-based CNN-LSTM compared ViViT an ablation determine time-dependency models. VGG16-LSTM predicted trinary classification EDSS score 6 0.74 AUC versus 0.84 (p-value < 0.001 per 5 × 2 cross-validation F-test) 80:20 hold-out testing split. However, outperformed when patients only (n = 94) (0.75 0.72 AUC, respectively). Exact investigated for both using regression strategies but showed collectively worse performance. Our experimental results demonstrate ability MS stratification disability, mimicking clinical practice. Further work includes external validation observational trials.

Язык: Английский

---

Radiomics in radiology: What the radiologist needs to know about technical aspects and clinical impact

La radiologia medica, Год журнала: 2024, Номер unknown

Опубликована: Окт. 30, 2024

Язык: Английский

---

Combining Radiomics and Connectomics in MRI Studies of the Human Brain: A Systematic Literature Review

Computer Methods and Programs in Biomedicine, Год журнала: 2025, Номер unknown, С. 108771 - 108771

Опубликована: Апрель 1, 2025

Язык: Английский

---