Section: New Results

Application of deep learning on healthcare

Participants : Thanh Hung Nguyen, François Brémond.

Healthcare standards have changed dramatically over the past 100 years. People nowadays are aware of the importance of health in their lives. The rising income has enabled them to use private healthcare services. Since the demand is growing rapidly, the physicians need tools which can help them work effeciently in terms of time and cost. Parallely, deep learning has become very popular in the recent years because of its success in computer vision. We have proposed two applications ,VISIONUM and REMINARY, as demonstration of the impact of deep learning on the healthcare.

The first application, VISIONUM (see 9.2.2.1), can detect when the patients lose their focus on the therapeutic exercise and thereby reminds them to return back to the exercise. Since the patients can correct themselves on their own, while they are doing exercises, therapists can focus more on the performance of the exercise. It also helps the therapist to keep track of multiple patients at the same time and hence, save both time and money (see Figure 35).

The second application, REMINARY (see 9.2.2.3), aims to detect the movement of people. In this application, the movement of patients is tracked and analysed. The output gives therapists an overview of the movement of their patients. This information is used by the therapist to monitor the diseases which are related to the movement and decide if there is any improvement due to the treatment. For both applications, we can also analyse emotions like happiness of patient. This information is extremely important for the design of the exercise (see Figure 36 and Figure 37).