Predictive Models on Psychosocial Profile, Fatigue and Sleep Quality

Fibromyalgia

The primary goal of this research project is to develop different prediction models in fibromyalgia disease through the application of machine learning techniques and to assess the explainability of the results. As specific objective the research project intends to evaluate the influence of psychosocial variables, fatigue, and sleep quality on the prediction of disease severity in patients with fibromyalgia using an artificial intelligence-based model.

Fibromyalgia (FM) is a condition characterized by chronic musculoskeletal pain, the pathophysiology of which remains unclear. In addition, this disorder is frequently associated with sleep disturbances, pronounced fatigue, morning stiffness, poor quality of life, cognitive alterations (primarily memory-related problems), and psychological disturbances, including depression, anxiety, and stress. FM is significantly more prevalent in women, with an estimated female-to-male ratio of approximately 3:1. Its prevalence in the adult population is estimated at 2-3%, increasing with age and peaking between 45 and 65 years. The socioeconomic and healthcare burden of this condition is substantial, as FM is associated with high rates of medical consultations, unemployment, work disability, and the need for disability-related financial support. FM has been linked to higher levels of negative affect, defined as a general state of distress encompassing aversive emotions such as sadness, fear, anger, and guilt. Patients with FM commonly exhibit elevated levels of anxiety, depression, pain catastrophizing, and stress, which are associated with worsening of symptoms, including cognitive impairments. In recent years, machine learning, data mining, and artificial intelligence (AI) techniques have been successfully applied to the development of computer-aided diagnosis (CAD) systems for the identification of complex health conditions, achieving good levels of accuracy and efficiency by recognizing potentially meaningful patterns in health-related data. Accordingly, these technologies provide powerful tools for multivariable data analysis, enabling model-based predictions and offering a clear advantage in risk assessment across a wide range of diseases. In addition, these approaches not only support the development of clinical predictive models but also enhance clinicians' ability to interpret and apply their results in practice. Explainable machine learning models allow clinical experts to make data-driven decisions and to deliver personalized treatments while maintaining a high standard of care. These models fall within the field of explainable artificial intelligence (XAI), which aims to develop interpretable models that preserve high predictive accuracy while improving the transparency, understanding, and trustworthiness of model outputs. A previous publication by our research group demonstrated, using machine learning models, a strong association between mental health factors-particularly anxiety and depression-and fibromyalgia severity, with a greater influence than pain intensity itself. Other prevalent symptoms of fibromyalgia, such as fatigue and sleep quality disturbances, are also considered by patients to be major contributors to reduced quality of life, often perceived as more impactful than pain. Accordingly, the present study aims to develop predictive models of fibromyalgia severity using machine learning and explainable artificial intelligence techniques, incorporating fatigue and sleep quality as novel variables in the analysis. Understanding the individual and combined relevance of these psychosocial components in determining patients' clinical status may contribute to improved disease management and more personalized treatment strategies, ultimately enhancing the quality of care.

No related papers found yet.

None listed

Spain