Single-timepoint low-dimensional characterization and classification of acute versus chronic multiple sclerosis lesions using machine learning

Deep learning for discrimination of active and inactive lesions in multiple sclerosis using non-contrast FLAIR MRI: A multicenter study

BACKGROUND

Within the domain of multiple sclerosis (MS), the precise discrimination between active and inactive lesions bears immense significance. Active lesions are enhanced on T1-weighted MRI images after administration of gadolinium-based contrast agents, which brings about associated complexities. This study investigates the potential of deep learning to differentiate between active and inactive lesions in MS using non-contrast FLAIR-type MRI data, presenting a non-invasive alternative to conventional gadolinium-based MRI methods.

METHODS

The dataset encompasses 9097 lesion images collected from 130 MS patients across four distinct imaging centers, with post-contrast T1-weighted images as the benchmark reference. We initially identified and labeled the lesions and carefully selected corresponding regions of interest (ROIs). These ROIs were employed as inputs for a convolutional neural network (CNN) to predict lesion status. Also, transfer learning was utilized, incorporating 12 pre-trained CNN models. Subsequently, an ensemble technique was applied to 3 of best models, followed by a systematic comparison of the results.

RESULTS

Through a 5-fold cross-validation, our custom designed network exhibited an average accuracy of 85 %, a sensitivity of 95 %, a specificity of 75 %, and an AUC value of 0.90. Among the pre-trained models, ResNet50 emerged as the most effective, achieving a specificity of 58 %, an accuracy of 75 %, a sensitivity of 91 %, and an AUC value of 0.81. Our comprehensive evaluations encompassed the receiver operating characteristic curve, precision-recall curve, and confusion matrix analyses.

CONCLUSION

The findings underscore the efficacy of the proposed CNN, trained on FLAIR MRI data ROIs, in accurately discerning active and inactive lesions without reliance on contrast agents. Our multicenter study of 130 patients with diverse imaging devices outperforms the other single-center studies, achieving superior sensitivity and specificity. Unlike studies using multiple modalities, our exclusive use of FLAIR images streamlines the process, and our streamlined approach, excluding conventional pre-processing, demonstrates efficiency. The external validation conducted on diverse datasets, coupled with the analysis of dilated masks, underscores the adaptability and efficacy of our custom CNN model in discerning between active and inactive lesions.

Artificial intelligence applied to MRI data to tackle key challenges in multiple sclerosis

Artificial intelligence (AI) is the branch of science aiming at creating algorithms able to carry out tasks that typically require human intelligence. In medicine, there has been a tremendous increase in AI applications thanks to increasingly powerful computers and the emergence of big data repositories. Multiple sclerosis (MS) is a chronic autoimmune condition affecting the central nervous system with a complex pathogenesis, a challenging diagnostic process strongly relying on magnetic resonance imaging (MRI) and a high and largely unexplained variability across patients. Therefore, AI applications in MS have the great potential of helping us better support the diagnosis, find markers for prognosis to eventually design more powerful randomised clinical trials and improve patient management in clinical practice and eventually understand the mechanisms of the disease. This topical review aims to summarise the recent advances in AI applied to MRI data in MS to illustrate its achievements, limitations and future directions.

An update on the role of magnetic resonance imaging in predicting and monitoring multiple sclerosis progression

INTRODUCTION

While magnetic resonance imaging (MRI) is established in diagnosing and monitoring disease activity in multiple sclerosis (MS), its utility in predicting and monitoring disease progression is less clear.

AREAS COVERED

The authors consider changing concepts in the phenotypic classification of MS, including progression independent of relapses; pathological processes underpinning progression; advances in MRI measures to assess them; how well MRI features explain and predict clinical outcomes, including models that assess disease effects on neural networks, and the potential role for machine learning.

EXPERT OPINION

Relapsing-remitting and progressive MS have evolved from being viewed as mutually exclusive to having considerable overlap. Progression is likely the consequence of several pathological elements, each important in building more holistic prognostic models beyond conventional phenotypes. MRI is well placed to assess pathogenic processes underpinning progression, but we need to bridge the gap between MRI measures and clinical outcomes. Mapping pathological effects on specific neural networks may help and machine learning methods may be able to optimize predictive markers while identifying new, or previously overlooked, clinically relevant features. The ever-increasing ability to measure features on MRI raises the dilemma of what to measure and when, and the challenge of translating research methods into clinically useable tools.

Exemplar MobileNetV2-Based Artificial Intelligence for Robust and Accurate Diagnosis of Multiple Sclerosis

Multiple sclerosis (MS) is a chronic autoimmune disease of the central nervous system that prominently affects young adults due to its debilitating nature. The pathogenesis of the disease is focused on the inflammation and neurodegeneration processes. Inflammation is associated with relapses, while neurodegeneration emerges in the progressive stages of the disease. Magnetic resonance imaging (MRI) is commonly used for the diagnosis of MS, and guidelines such as the McDonald criteria are available. MRI is an essential tool to demonstrate the spatial distribution and changes over time in the disease. This study discusses the use of image processing techniques for the diagnosis of MS and specifically combines the MobileNetV2 network with exemplar-based learning, IMrMr feature selection, and K-Nearest Neighbors (KNN) classification methods. Experiments conducted on two different datasets (Dataset 1 and Dataset 2) demonstrate that these methods provide high accuracy in diagnosing MS. Dataset 1 comprises 128 patients with 706 MRI images, 131 MS patients with 667 MRI images, and 150 healthy control subjects with 1373 MRI images. Dataset 2 includes an MS group with 650 MRI images and a healthy control group with 676 MRI images. The results of the study include 10-fold cross-validation results performed on different image sections (Axial, Sagittal, and Hybrid) for Dataset 1. Accuracy rates of 99.76% for Axial, 99.48% for Sagittal, and 98.02% for Hybrid sections were achieved. Furthermore, 100% accuracy was achieved on Dataset 2. In conclusion, this study demonstrates the effective use of powerful image processing methods such as the MobileNetV2 network and exemplar-based learning for the diagnosis of MS. These findings suggest that these methods can be further developed in future research and offer significant potential for clinical applications in the diagnosis and monitoring of MS.

Investigation and benchmarking of U-Nets on prostate segmentation tasks

In healthcare, a growing number of physicians and support staff are striving to facilitate personalized radiotherapy regimens for patients with prostate cancer. This is because individual patient biology is unique, and employing a single approach for all is inefficient. A crucial step for customizing radiotherapy planning and gaining fundamental information about the disease, is the identification and delineation of targeted structures. However, accurate biomedical image segmentation is time-consuming, requires considerable experience and is prone to observer variability. In the past decade, the use of deep learning models has significantly increased in the field of medical image segmentation. At present, a vast number of anatomical structures can be demarcated on a clinician's level with deep learning models. These models would not only unload work, but they can offer unbiased characterization of the disease. The main architectures used in segmentation are the U-Net and its variants, that exhibit outstanding performances. However, reproducing results or directly comparing methods is often limited by closed source of data and the large heterogeneity among medical images. With this in mind, our intention is to provide a reliable source for assessing deep learning models. As an example, we chose the challenging task of delineating the prostate gland in multi-modal images. First, this paper provides a comprehensive review of current state-of-the-art convolutional neural networks for 3D prostate segmentation. Second, utilizing public and in-house CT and MR datasets of varying properties, we created a framework for an objective comparison of automatic prostate segmentation algorithms. The framework was used for rigorous evaluations of the models, highlighting their strengths and weaknesses.