Recognition of PVD based on Walking Pattern Acquired from Smartphone and classification by CNN

PVD is a highly common blood vessel disease in modern society, which can be treated in a variety of ways. Treatment of the disease ranges from non-invasive treatments such as taking drugs to especially invasive treatments such as surgery under anesthesia. However, the problem we wish to solve relates to the accessibility of diagnosing the disease to the general population, with an emphasis on simple diagnosis.
Today, diagnosis of the disease requires expensive resources because the existing tests require hospital equipment and the interpretation of a specialist. In this project we are trying to implement a smart and accessible system for identifying a symptom of PVD. Prof. Aharon Hoffman suggested using a free smartphone application that measures the patient's acceleration during walking. The phone is placed in the user's pocket and by command is measuring his acceleration for a few minutes.
This project is a follow-up project. However, we needed to do all the work again due to significant failures in the previous results that led to the selection of a new solution. In Part I of the project, we were focused primarily on the initial processing phase, building the application, using a server for the data storage and the system used by the doctor, and processing the initial information in order to extract a number of features that we can use more easily to diagnose the disease. In Part II, we made improvements and adaptations to the system in order to meet the requirements of 'Clalit' Health Maintenance Organization (HMO) for extensive use by all the physicians at the HMO. In addition, we implemented a learning system that can classify the condition of the patient as sick with PVD or healthy.

Recognition of PVD based on Walking Pattern Acquired from Smartphone and classification by CNN
Recognition of PVD based on Walking Pattern Acquired from Smartphone and classification by CNN
Recognition of PVD based on Walking Pattern Acquired from Smartphone and classification by CNN
Collaboration:

Prof. Aharon Hoffman