Inter-body metabolic processes determine body surface temperature. Due to the different activity of various organs, the temperature is not the same for all parts of the body or in all body situations (e.g. cold or warm weather, jugging?). However, body temperature follows some rules widely used in medical imaging to diagnose different abnormalities like cancerous tumours, musculoskeletal injuries and etc. However, infrared thermography systems are stationary, bulky and expensive.
Wearable pervasive healthcare systems are proposed in everyday life to sense body signals and estimate the user healthy situation. The wearable property causes some limitations and opportunities for such systems. The wearable systems are very limited in flexibility and accuracy. Besides, they suffer from extra noises due to the user movements and environmental variations. On the other hand, wearable systems sense the bio-signals for long periods. More information is usually helpful to improve the estimation quality.
The idea of this project is to design and implementation of a wearable system to sense the body temperature to estimate some abnormalities based on the temperature variations. To gain better results, a primary study on the thermography basics is also suggested. Moreover, due to their limited accuracy, thermometer sensors should be calibrated.
Second step of the project is signal processing. In the step, proper feature extraction and classification algorithms should be applied on the aggregated signals to estimate abnormalities.