Human Activity Recognition for Health Care using Deep Learning

Health is something which give cause to soul so one can live life without troubles. In simpler words one cannot compromise on health at any cost that's why he/she has to strive for its betterment. Digging more into it, Cardiovascular Diseases (CVD) are the leading source of deaths all arou

Project Title

Human Activity Recognition for Health Care using Deep Learning

Project Area of Specialization

Artificial Intelligence

Project Summary

Health is something which give cause to soul so one can live life without troubles. In simpler words one cannot compromise on health at any cost that's why he/she has to strive for its betterment. Digging more into it, Cardiovascular Diseases (CVD) are the leading source of deaths all around the world. Facing this reality, the European Union funded MyHeart Project, whose mission is empowering citizens to fight CVD by means of a preventive lifestyle and an early diagnosis. This initiative of European Union motivated us to do this kind of great cause for our people as in Pakistan all of us are aware that how contaminated our food is and this unhealthy food is leading source to grow more CVD patients in our country. Moreover, we know due to lake of technology and awareness how expensive is treatment of CVD in Pakistan. To save our beloved people from getting into that zone it important to diagnose CVD in its early stages. For that we want to create Heart Failure Management System (HFMS), in which we will be trying to help people suffering from CVD.

In HFMS different sensors will be used for measuring different parameters which are affecting heart and by those parameters we will diagnose how severe disease is and how severe it’s going to be then based on that severity level we will recommend patients different physical exercises and precautions which can stop disease from progressing further.

Project Objectives

• This project will help heart patients to diagnose their disease before severity level.

• It will allow patient self-management of their chronic condition.

• It will provide people health assistance who cannot afford expenses of CVDs treatment.

• It will reduce heart patients by daily updates about their health and recommending precaution to skip severity level.

• Somehow it will reduce effects of contaminated food which is major factor to spread deadlier diseases.

Project Implementation Method

Different deep learning algorithms will be used to assess values coming right from sensors. Moreover, we will try to create patient's virtual personality by his/her daily activities and routine and then we will recommend which activities he/she has to carry on and which precautions are currently needed to stop further enhancement of disease.

Deep Learning techniques to be used are Convolutional Neural Network (CNN), Recurrent Neural Network (RNN), Long Short Term Memory (LSTM), Hidden Markov Model (HMM).

These algorithms will be tested and based on greater accuracy one of them or fusion of 2 or 3 will be used accordingly.

Benefits of the Project

• Project will benefit different heart patients who are in early stages of their CVDs.

• It will also help different people other than CVD in their general diseases in daily health report along with suggestions and recommendations.

• It will surely decreases growing rate of diseases as mostly diseases are caused by heart malfunctioning.

• It will diagnose CVDs in most early stages as all of us are aware that how severe this disease can be after sometime.

• It will save loved ones of their families who are only to take care of their families.

Technical Details of Final Deliverable

Our final deliverable will be an Android Application (might be available on Google Play Store) and Web Application. An Android Application will get different values (accelerometer, gyroscope and magnetometer values) from its user based on user's different activities like sitting, standing, sleeping, smoking, running and other usual activities performed by user in his/her daily routine. All that values will be received by algorithm as input, will be processed and then will generate General Health Report (GHR) and Heart Functioning Report (HFR) of user on daily, weekly, or monthly basis (user can specify). Different exercises and precautions will also be prescribed by the help of various professional doctor's help to stop further progressing.

By Web Application we will be monitoring patient 24/7 without letting patient know that he/she is being monitored. Actually illusion will be created that he/she is not getting monitored so that he/she could be able to carry on normal routine without any hesitation. We will take advantage of illusion and will monitor patient’s routine to identify between healthy and unhealthy activities.

Final Deliverable of the Project

Hardware System

Core Industry

IT

Other Industries

Health

Core Technology

Artificial Intelligence(AI)

Other Technologies

Internet of Things (IoT)

Sustainable Development Goals

Good Health and Well-Being for People

Required Resources

If you need this project, please contact me on contact@adikhanofficial.com