Early Risk Prediction of Fibromyalgia using Symptom Lifestyle Features and Machine Learning Models

Read full text Summarise with VIRAT
Authors: S. Balaji, S. Agilavani, B. Madhumithra, M. Kaviya
Cited by: 0
References: 22

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Fibromyalgia (FM) is a heterogeneous, multi-symptom disorder whose early recognition is hampered by symptom overlap with other conditions and reliance on subjective reports.This paper systematically reviews 25 studies (2015)(2016)(2017)(2018)(2019)(2020)(2021)(2022)(2023)(2024)(2025) and proposes an integrated ML framework for early FM risk stratification that fuses self-reported symptom scores, lifestyle indicators, and physiological signals.Methodology: structured literature search, comparative extraction of datasets/features/algorithms, and experimental comparison of Logistic Regression, Random Forest and XGBoost with class-imbalan ce handling.Key findings: ensemble models (XGBoost) performed best (AUC ≈ 0.85-0.90)and stress, sleep quality and fatigue were the most consistent predictors across studies.Main gap: lack of large-scale, multimodal longitudinal datasets and limited external validation.We conclude with a taxonomy of ML methods and targeted future directions (federated learning, wearables, XAI) to accelerate clinical translation.

Fields of Study & Subjects

Fibromyalgia and Chronic Fatigue Syndrome Research Artificial Intelligence in Healthcare Psychosomatic Disorders and Their Treatments

Publication details

Journal / source
Advances in intelligent systems research/Advances in Intelligent Systems Research
Pages
663–684
Publication date
1 April 2026
Language
EN
Document type
Book chapter