In aesthetic medicine clinics, cellulite remains one of the most frequent concerns. It affects between 85% and 98% of post-pubertal women and, although it does not represent a serious health problem, it does have a significant impact on self-esteem and quality of life. For years, its diagnosis and monitoring have depended almost exclusively on clinical observation and the professional’s experience. However, a tool born in the field of biomedical engineering is changing the landscape: infrared thermography.
A recent study, published in The EPMA Journal, has shown that combining non-invasive thermography with artificial intelligence algorithms not only allows identification of cellulite in advanced stages, but also detects it in very early phases, with an accuracy exceeding 90% at the initial stage.
Infrared thermography, widely used in medicine to detect thermal alterations associated with tumors, muscle injuries, or vascular problems, works by measuring the infrared radiation emitted by the skin. In the case of cellulite, irregularities in microcirculation and adipose tissue structure generate characteristic thermal patterns, very similar to the hot spots produced by exercise, but in this case caused by subcutaneous fat.
The great advantage over traditional methods is twofold: it requires no physical contact and provides objective measurement. This means that two different professionals, looking at the same image, will reach the same conclusion, eliminating much of the subjectivity that accompanies simple visual inspection.
In the study by Bauer et al. (2020), thermal images were processed using an algorithm capable of automatically isolating the region of interest (in this case, the posterior thighs) and extracting morphological features of the thermal distribution. This is where artificial intelligence comes into play.
The researchers combined an analysis method called Histogram of Oriented Gradients (HOG) with an artificial neural network (ANN). The first translates the images into numerical data describing the shape and direction of thermal changes; the second “learns” to associate those data with a specific stage of cellulite according to the Nürnberger-Müller scale.
The result: a system capable of automatically classifying the four stages (0 to 3) with an average accuracy of 80.95%, reaching 90% in initial cases. The most revealing metric of the study, the area under the ROC curve (AUC), showed values close to 0.8, reflecting a high ability to differentiate between healthy skin and different degrees of involvement.
The authors of the study foresee that, in the near future, AI-powered thermography will not only classify the stage but also differentiate between types of cellulite (hard, soft, edematous, or mixed) and relate thermal findings to variables such as lifestyle, physical activity, or hormonal conditions.
For the clinician, this means having a “digital assistant” that processes and classifies images in seconds, without fatigue or human bias.
In aesthetic medicine, evolution is everything. It is not enough to diagnose: it is necessary to demonstrate that a treatment works. Thermography makes it possible to capture images at different stages of treatment and compare them with millimetric precision, generating quantitative reports that support the patient’s progress.
The use of specialized software takes this process to another level. Integration between thermal camera and algorithm allows automating analysis, storing images and associated data, and generating standardized reports for the medical record or to show the patient clearly and objectively. The result: less consultation time spent on measurements and more time dedicated to therapeutic planning.
Cellulite will not disappear from the list of aesthetic concerns, but its evaluation may cease to be a subjective field. Incorporating thermography with artificial intelligence into daily practice means offering reproducible diagnoses, precise monitoring, and personalized treatments.
And, above all, it means speaking the same language as evidence-based medicine: objective, measurable, and comparable data.
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.
