Using Artificial Intelligence to detect early signs of Diabetic Foot Disease

It has been estimated, on a global scale, that a limb is lost every 20 seconds due to diabetes. Early detection and treatment have been shown to reduce the risk of amputations and foot ulcers.

Many solutions have been presented for identification of early signs of diabetic foot disease (DFD). In particular the use of thermal imaging has shown promise within the scientific and clinical literature.

However, there is still much research work to be done to create a commercially viable solution for capturing and processing thermal images for the purpose of identifying diabetic foot disease.

Self-management and the provision of effective foot care in the clinic and at home, is viewed as instrumental in providing appropriate care for people with diabetes. This in turn can reduce the costs associated with treating and managing diabetes and its associated complications.

In particular early detection and treatment are vital for managing complications and improving the long-term prognosis of those with diabetes. For prevention and care specific to diabetic foot disease (DFD), it is recommended that patients check their feet daily.

Signs of ulceration may include; redness, pain, loss of sensation, callus or thickened skin around the ulcer. For people with diabetes, an increase in the skin temperature on the plantar surface of the foot can signify inflammation or injury.

A temperature differential of 2.2°C or more between the same region on the contralateral feet may indicate inflammation and a potential ulceration. It is advised that patients should regularly monitor the skin temperature of the plantar foot so as to detect early signs of inflammation.

New products and services that facilitate easier and more effective measurement will have a major impact on the quality of life of patients in addition to reducing the socioeconomic costs of DFD.

This project was focused on using thermal imaging to detect areas of the foot sole that have an increased temperature and are likely to become ulcerated. The system uses an android application and a FLIR ONE thermal camera to capture images and temperature data.

The data is transmitted to a web service where supervised machine learning, registration processes, and temperature comparison are used to evaluate the data and determine whether a temperature hotspot is present. Deep learning approaches are used to detect and segment each distinct foot within an RGB image.

Images captured during trial

There are 3.9 million people in the UK have been diagnosed with diabetes. If not properly managed and treated diabetes can lead to serious complications such as heart disease, stroke, blindness, kidney failure and amputation.

In the UK, Diabetes leads to around 135 amputations every week. In Northern Ireland Diabetes causes approximately 150 amputations every year, of which about 80% could be prevented.

Approximately £1 in every £150 spent in the NHS was for diabetic foot ulceration or amputation. Technologies that facilitate self-management and self-care is viewed as instrumental in providing appropriate care for people with diabetes and reducing the costs associated with treating and managing diabetes and its associated complications.

App screenshot