Analysis of Facial Expressions for Pain Recognition in Fibromyalgia: Using Artificial Intelligence and Biomarkers

Analysis of Facial Expressions for Pain Recognition: Using Artificial Intelligence and Biomarkers as a Pain Diagnostic Tool in Fibromyalgia, a Pilot Study

Fibromyalgia (FM) is a chronic musculoskeletal pain syndrome with characteristics of generalized body pain, low pain threshold, tenderness and stiffness in muscles, tendons and joints. The assessment of pain in this condition is a challenge due to its subjective nature. A promising approach to assessing pain intensity is facial expression analysis, which can serve as an objective indicator. In addition, research seeks to identify molecular molecular markers to quantify pain. However, the lack of a standardized system has made it difficult to identify reliable markers. In summary, the search for objective methods of assessing pain in fibromyalgia is essential in order to develop more effective more effective treatments. Facial expression analysis and the investigation of molecular markers are promising ways of quantifying pain intensity more accurately and intensity of pain more accurately and reliably in fibromyalgia.

Introduction:

Fibromyalgia (FM) is a chronic syndrome characterized by diffuse musculoskeletal pain, fatigue and sleep disturbances, with a major impact on quality of life. Due to the subjectivity of pain assessment, the development of objective methods is essential. This study explores the use of artificial intelligence (AI) in the analysis of facial expressions, combined with the investigation of molecular markers, as an innovative and quantitative approach to pain assessment in patients with FM.

Objective:

To validate the application of an AI tool combined with facial expression analysis and molecular biomarker research to measure pain intensity in FM patients.

Methodology:

An observational cohort study was carried out with 122 participants, divided into two groups: patients with FM (n=61) and without FM (n=61). Data collection included:

1. Facial expression recording: A convolutional neural network algorithm was used to analyze facial patterns associated with pain.
2. Biological samples: 1mL of saliva will be collected from each participant using the salivette method and processed to extract DNA, RNA and plasma proteins. The proteins will be quantified by ELISA and the genes associated with FM will be analyzed by RT-qPCR.
3. Clinical Questionnaires: Psychometric instruments such as the Visual Analogue Scale (VAS) and the Generalized Pain Index (GDI) were used to validate the results.
4. Statistical analysis: The data was analyzed using Kappa and Bland-Altman correlations to assess the agreement between the AI methods and the questionnaires, with a significance level of p<0.05.

The AI algorithm will use consistent facial patterns correlating them to the reported pain intensity, in agreement (Kappa=0.82) with the results of the clinical scales.The molecular markers analyzed are expected to show significant differences between the groups, with increased expression of inflammatory proteins in FM patients (p<0.05). The integration of facial and molecular analysis aims to amplify the accuracy of pain intensity classification.

This approach represents a promising advance in the diagnosis and management of the syndrome, contributing to personalized therapies and improving patients' quality of life.

Inclusion Criteria:

• The inclusion criteria are patients with no diagnosed cognitive deficit and who are willing to take part in the study.

Exclusion Criteria:

• Exclusion criteria are patients who use medication that can affect anxiety or depression or inability to understand the instructions.