In recent years, infrared thermography has established itself as an indispensable tool in medicine and sports science. Whether to prevent injuries, monitor fatigue, or individualize recovery processes in elite athletes, the ability to "see" and measure physiology through real-time thermal metrics is adding increasingly significant value to the daily work of sports institutions and clinics worldwide.
However, until recently, a major invisible obstacle persisted in the workflow of any thermography professional: image analysis and processing.
Traditionally, to extract reliable quantitative data, an evaluator had to spend hours in front of a screen manually segmenting the so-called regions of interest (ROIs) on each image. This manual process is not only time-consuming but also introduces human bias shaped by the evaluator's experience, which can affect data reproducibility.
The solution? Automated analysis. But is automated software as accurate and reliable as the eye of a human expert?
To answer this question definitively, an international team of researchers led by Dr. Jose Ignacio Priego-Quesada (GIBD, University of Valencia) has published a crucial study in the prestigious scientific journal Journal of Thermal Biology. In it, they evaluate for the first time in lower limb (leg) protocols the level of agreement and time efficiency of the ThermoHuman® automatic segmentation software compared to traditional manual methods before and after exercise. This work expands upon previous scientific evidence from Requena (2020), which had already validated the algorithm specifically for the soles of the feet.
Below, we analyze the details of this new study and what it contributes to previously published data on the scientific validation of our technology.
To test the software's accuracy in a real and metabolically demanding scenario, the researchers involved 19 recreational runners (10 men and 9 women). The participants completed a 10-kilometer outdoor run under standardized environmental conditions.
A total of 76 thermographic images (anterior and posterior views of the lower limbs) were captured at two key moments: before and immediately after running. Image processing was split between an expert evaluator (with 5 years of experience in thermography research) and ThermoHuman’s automated algorithm (web version, July 2025).
To assess agreement, three different methodologies were compared across 22 regions of interest (ROIs) on the legs:
The study's findings left no room for doubt, confirming with robust data the two main hypotheses: ThermoHuman is scientifically equivalent to manual analysis in terms of accuracy, and overwhelmingly superior in terms of time efficiency.
Statistical analysis using linear mixed-effects models and concordance correlation coefficients (CCC) demonstrated that temperature differences between ThermoHuman and manual methods are virtually non-existent.
In simple terms: the automated software provides the exact same skin temperature data as an expert human eye manually analyzing each area.
While accuracy is identical, the time required to obtain the data tells a radically different story. To process the 76 images in the study (which involved analyzing a total of 1,672 regions of interest):
This represents an 85.6% time saving compared to the conventional manual method, and a 93% saving compared to the exact replication of boundaries. These figures not only support but improve upon the framework analyzed in Requena's (2020) scientific paper—which reported an 86% reduction—explained by the implementation of a second-generation technology (ThermalApp 2.0) in recent years. Imagine what this means for a professional football club or a clinic where dozens of patients or athletes are evaluated daily: hours of tedious work can now be dedicated directly to decision-making, data interpretation, and athlete care.
One of the major added advantages of ThermoHuman in its latest software versions is the ability to make quick manual adjustments if the automatic segmentation makes a minor error.
In this study, out of the 1,672 regions automatically analyzed by the ThermoHuman algorithm, only 18 regions (1.08%) required minor manual correction by the evaluator. These occasional errors were concentrated in distal areas or those more prone to shape variations, such as the anterior foot or the adductor region. This minimal percentage demonstrates the immense maturity the algorithm has reached through the ongoing use of Machine Learning techniques.
The study also reflects the typical limitations faced by any professional in daily fieldwork. For instance, 229 ROIs (13.70%) were excluded from the final analysis because the participants' clothing partially covered the skin (mainly in the gluteal and upper lateral thigh areas).
This detail, far from being a flaw, highlights the methodological honesty of the research and serves as a clinical reminder: to obtain reliable thermographic data of the lower body, it is essential to strictly adhere to clothing protocols, fully exposing the skin of the regions to be analyzed.
For sports and health professionals, this study provides three key certainties:
The study published by Januário et al. (2026) scientifically validates ThermoHuman®'s software once again, demonstrating a qualitative leap in its performance with superior ICC (Intraclass Correlation Coefficient) values and an increase in time savings, which rises from the previous 86% to an extraordinary 93%. These results confirm ThermoHuman® as a fully reliable, validated, and fast tool, ideal for both rigorous scientific research and daily professional work in sports environments. Consequently, it solidifies its position as the solution offering the highest guarantees of accuracy and efficiency on the market.
Automation and artificial intelligence are no longer the future of thermography; they are the present, backed by science.
ThermoHuman has had the support of the Funds of the European Union and the Community of Madrid through the Operational Programme on Youth Employment. Likewise, ThermoHuman within the framework of the Export Initiation Program of ICEX NEXT, had the support of ICEX and the co-financing of the European Regional Development Fund (ERDF).
Thermohuman has received funding from the Centre for the Development of Industrial Technology (CDTI), in participation with the European Regional Development Fund (ERDF), for the R+D activities involved in creating a new tool, based on thermography, for the prediction and prevention of rheumatoid arthritis. See project detail.
