Detection of Fall Risk in Multiple Sclerosis by Gait Analysis-An Innovative Approach Using Feature Selection Ensemble and Machine Learning Algorithms

AI-assisted assessment of fall risk in multiple sclerosis: A systematic literature review

BACKGROUND

Multiple sclerosis (MS) is an autoimmune disease that can increase the risk of falls in patients due to various factors. Traditional clinical assessments may not effectively identify those at risk of falling.

OBJECTIVE

This study aimed to use artificial intelligence and machine learning techniques to predict the likelihood of falls in patients with MS based on a review of previous research.

METHODS

A systematic review was conducted following PRISMA guidelines, searching electronic databases from 1990 to 2024. Data extraction and quality assessment were carried out.

RESULTS

Seven studies were analyzed, and it was determined that patient-reported outcomes (PROs) such as MSWS-12 and EMIQ performed better than other methods. Sensor-based systems such as GAITRite and Mobility Lab achieved high F1 scores. Random forest classifiers utilizing postural sway measures were effective in discriminating low-risk MS patients from healthy controls. Deep learning models, particularly BiLSTM architectures, outperformed traditional machine learning approaches in identifying recent fallers using wearable accelerometer data.

CONCLUSION

The findings highlight the potential of PROs, the promise of wearable sensors and deep learning, and the importance of optimizing data collection for effective fall risk assessment in the MS population.

Using machine learning algorithms to detect fear of falling in people with multiple sclerosis in standardized gait analysis

INTRODUCTION

Multiple sclerosis (MS) is the most common chronic inflammatory disease of the central nervous system. The progressive impairment of gait is one of the most important pathognomic symptoms which are associated with falls and fear of falling (FOF) in people with MS (pwMS). 60 % of pwMS show a FOF, which leads to restrictions in mobility as well as physical activity and reduces the quality of life in general. Therefore, early detection of FOF is crucial because it enables early implementation of rehabilitation strategies as well as clinical decision-making to reduce progression. Qualitative and quantitative evaluation of gait pattern is an essential aspect of disease assessment and can provide valuable insights for personalized treatment decisions in pwMS. Our objective was to identify the most appropriate clinical gait analysis methods to identify FOF in pwMS and to detect the optimal machine learning (ML) algorithms to predict FOF using the complex multidimensional data from gait analysis.

METHODS

Data of 1240 pwMS was recorded at the MS Centre of the University Hospital Dresden between November 2020 and September 2021. Patients performed a multidimensional gait analysis with pressure and motion sensors, as well as patient-reported outcomes (PROs), according to a standardized protocol. A feature selection ensemble (FS-Ensemble) was developed to improve the classification performance. The FS-Ensemble consisted of four filtering methods: Chi-square test, information gain, minimum redundancy maximum relevance and ReliefF. Gaussian Naive Bayes, Decision Tree, k-Nearest Neighbor, and Support Vector Machines (SVM) were used to identify FOF.

RESULTS

The descriptive analysis showed that 37 % of the 1240 pwMS had a FOF (n = 458; age: 51 ± 16 years, 76 % women, median EDSS: 4.0). The FS-Ensemble improved classification performance in most cases. The SVM showed the best performance of the four classification models in detecting FOF. The PROs showed the best F1 scores (Early Mobility Impairment Questionnaire F1 = 0.81 ± 0.00 and 12-item Multiple Sclerosis Scale F1 = 0.80 ± 0.00).

CONCLUSION

FOF is an important psychological risk factor associated with an increased risk of falls. To integrate a functional early warning system for fall detection into MS management and progression monitoring, it is necessary to detect the relevant gait parameters as well as assessment methods. In this context, ML strategies allow the integration of gait parameters from clinical routine to support the initiation of early rehabilitation measures and adaptation of course-modifying therapeutics. The results of this study confirm that patients' self-assessments play an important role in disease management.

A machine learning approach to determine the risk factors for fall in multiple sclerosis

BACKGROUND

Falls in multiple sclerosis can result in numerous problems, including injuries and functional loss. Therefore, determining the factors contributing to falls in people with Multiple Sclerosis (PwMS) is crucial. This study aims to investigate the contributing factors to falls in multiple sclerosis using a machine learning approach.

METHODS

This cross-sectional study was conducted with 253 PwMS admitted to the outpatient clinic of a university hospital between February and August 2023. A sociodemographic data collection form, Fall Efficacy Scale (FES-I), Berg Balance Scale (BBS), Fatigue Severity Scale (FSS), Expanded Disability Status Scale (EDSS), Multiple Sclerosis Impact Scale (MSIS-29), and Timed 25 Foot Walk Test (T25-FW) were used for data collection. Gradient-boosting algorithms were employed to predict the important variables for falls in PwMS. The XGBoost algorithm emerged as the best performed model in this study.

RESULTS

Most of the participants (70.0%) were female, with a mean age of 40.44 ± 10.88 years. Among the participants, 40.7% reported a fall history in the last year. The area under the curve value of the model was 0.713. Risk factors of falls in PwMS included MSIS-29 (0.424), EDSS (0.406), marital status (0.297), education level (0.240), disease duration (0.185), age (0.130), family type (0.119), smoking (0.031), income level (0.031), and regular exercise habit (0.026).

CONCLUSIONS

In this study, smoking and regular exercise were the modifiable factors contributing to falls in PwMS. We recommend that clinicians facilitate the modification of these factors in PwMS. Age and disease duration were non-modifiable factors. These should be considered as risk increasing factors and used to identify PwMS at risk. Interventions aimed at reducing MSIS-29 and EDSS scores will help to prevent falls in PwMS. Education of individuals to increase knowledge and awareness is recommended. Financial support policies for those with low income will help to reduce the risk of falls.

The role of miRNAs in multiple sclerosis pathogenesis, diagnosis, and therapeutic resistance

In recent years, microRNAs (miRNAs) have gained increased attention from researchers around the globe. Although it is twenty nucleotides long, it can modulate several gene targets simultaneously. Their mal expression is a signature of various pathologies, and they provide the foundation to elucidate the molecular mechanisms of each pathology. Among the debilitating central nervous system (CNS) disorders with a growing prevalence globally is the multiple sclerosis (MS). Moreover, the diagnosis of MS is challenging due to the lack of disease-specific biomarkers, and the diagnosis mainly depends on ruling out other disabilities. MS could adversely affect patients' lives through its progression, and only symptomatic treatments are available as therapeutic options, but an exact cure is yet unavailable. Consequently, this review hopes to further the study of the biological features of miRNAs in MS and explore their potential as a therapeutic target.