Infrared Thermography and Radiomics: Early Detection of Metabolic Syndrome

Metabolic syndrome (MS) is a cluster of abnormalities—such as abdominal obesity, hypertension, insulin resistance, and dyslipidemia—that increase the risk of cardiovascular disease and type 2 diabetes. Early detection is key, but conventional methods are often invasive, slow, or dependent on costly equipment.

Metabolic syndrome is a set of physiological and metabolic alterations that include abdominal obesity, high blood pressure, insulin resistance, and dyslipidemia. The simultaneous presence of these factors significantly increases the risk of developing cardiovascular diseases and type 2 diabetes. It is a global public health problem, with a rising prevalence in recent decades due to changes in eating habits, sedentary lifestyles, and population aging. Early detection and continuous monitoring of these alterations are essential to prevent severe complications and optimize lifestyle interventions and medical treatments.

A study published in Scientific Reports by Guo et al. (2025) proposes an innovative approach: using infrared thermography (IRT) combined with radiomics to identify thermal patterns associated with MS in a non-invasive and rapid way. In this post, we present some of the most relevant details of this research.

Study with Thermal Imaging and Radiomics

The work included 200 adult men—100 diagnosed with MS and 100 healthy controls. Thermal images of the face and palm were captured, extracting a total of 1,656 radiomic features, including:

• First-order statistics (mean temperature, standard deviation)
• Textures (GLCM, GLRLM, GLSZM, among others)
• Multiscale filters to capture thermal pattern details

To select the most relevant variables, correlation analysis, t-tests, and LASSO regression were applied. The predictive model was trained using a Random Forest algorithm and validated on an independent dataset.

Main Findings

• The model based on radiomic features achieved an AUC of 0.91 in validation, far outperforming the use of mean temperature alone (facial AUC = 0.63; palmar = 0.54).
• The most relevant thermal patterns were located in specific regions of the face and palm, linked to microvascular dysfunction and chronic inflammation.
• Image acquisition and analysis were completed in less than 5 minutes per participant.
• The method requires no physical contact, radiation, or sample collection.

Interpretation and Practical Application

These results suggest that infrared thermography, combined with advanced image analysis techniques, can identify early systemic physiological changes associated with metabolic syndrome.

This offers enormous potential for population screening programs, risk monitoring, and intervention follow-up, especially in resource-limited settings.

Conclusion

In this study, the combination of infrared thermography and radiomics:

• Accurately distinguished subjects with and without metabolic syndrome.
• Outperformed conventional thermal measurements.
• Proved to be a rapid, non-invasive, and potentially scalable method.

Infrared thermography emerges as a highly promising tool for use in metabolic syndrome.

Relevance of Thermography in This Context

This work highlights the use of thermography to assess complex thermal patterns and link them to systemic pathophysiological states such as metabolic syndrome. Furthermore, in sports and performance contexts, this technology could also be used to detect early metabolic alterations that affect recovery, energy availability, and overall athlete health, thus contributing to long-term individualized optimization and prevention programs.

Reference

• Guo, J., Xue, H., Chen, Y., Li, X., Chen, Y., Zhang, X., ... & Xiao, Y. (2025). Infrared thermography-based radiomics for early detection of metabolic syndrome. Scientific Reports, 15(1), 13984.