PEDITERM
Integration of Digital Infrared Thermography (DIT) with deep learning algorithms for automated Diabetic Foot (DF) screening.
Are you interested?Need
Diabetic Foot (DF) is a serious complication of diabetes mellitus (DM) affecting between 3% and 13% of patients with DM, and is responsible for 30% of associated hospital admissions. In Catalonia, diabetes affects 13.8% of the population (approximately 630,000 people), with a higher prevalence in men and patients with type 2 diabetes (DM2). The main risk factor for amputations in patients with DF is the combination of diabetic neuropathy (NP) and peripheral arterial disease (PAD). Current screening tools have limitations for the early detection of NP and PAD, being poorly sensitive and subjective. The unmet need is a more effective screening strategy that improves early detection and reduces complications.
Solution
The solution proposes integrating Digital Infrared Thermography (DIT) with deep learning algorithms for automated Diabetic Foot (DF) screening. Using thermographic cameras and convolutional neural networks, the system analyzes images to detect subclinical neuropathic and neuroischemic anomalies, improving diagnostic accuracy and reducing evaluation time. Preliminary results in healthy patients have demonstrated high accuracy in the detection of anomalies, reinforcing its potential in the clinical setting.
The system integrates easily into Primary Care (PC) thanks to its automation, reducing screening time and eliminating the need for prior training for professionals. Furthermore, it is a non-invasive tool that can detect complications in early stages, thereby reducing the risk of amputations and hospitalizations associated with DF. Regarding economic viability, this technology represents a low initial cost compared to the high costs of treating severe complications, such as amputations. In addition, integration with electronic health records (EHR) enables scalable adoption, improving early detection in different regions. Its application can not only reduce healthcare costs, but also improve the quality of life of patients, preventing serious complications and reducing the psychological impact associated with DF, such as depression or social isolation.
Goal
Improve DF screening, detecting subclinical thermal alterations, through the integration of Digital Infrared Thermography (DIT) with deep learning algorithms.