One of the most surprising facts we encounter is that the professionals and experts who work their best and to the fullest is none but the doctors who face tremendous obstacles in carrying out their service. They are often considered to be the avatars of God, it is said that the doctors are God's representatives walking on the Earth to save our lives. But despite their tireless efforts it is keenly observed that due to several factors there are thousands and thousands of people who are unable to get the righteous treatment, solution or guidance in their gravest scenarios of health. It is also observed that either the reachability is low, the doctor-patient ratio is highly unbalanced or the cost is high (far beyond affordability). This way it is hard to say whether the reason of suffering and of course the mortality is the system, or the doctors, or the healthcare sectors, or the population explosion, or the less spread reach of facilities. Honestly none can be blamed alone, as blaming anyone would be highly unfair and discriminatory. Hence the primary solution that we can find for the betterment of the process, which indeed is our main concern, is automating some portions of the healthcare structure and with the breakthrough of Artificial Intelligence the process of automation can be taken to a different level. The system shall indeed act as a true assistant to the doctors and the healthcare professionals as well. From the seeds of this idea we were able to decide on constructing a CVD prediction model which shall not just allow a faster and relatively cheaper model for tests but also reduce the uncanny burden of the doctors allowing them to meticulously focus more on the severe cases. We believe, this way the whole process of automation and integration of AI in CVD prediction will help in increasing time-efficiency, less burden of the healthcare professionals and higher accuracy of systems. And that was the driving motivation for our work...
What it does
We have developed an efficient cardiovascular disease prediction using ML and DL models for predicting the risk level of cardiovascular disease . The model uses 11 medical parameters such as age, gender, blood pressure, cholesterol, for prediction. The model predicts the likelihood of patients getting heart disease. It derives significant patterns and relationships between the input features and learns the features . After the model is trained it is then used for predicting for the testing data . The obtained results have illustrated that the model can effectively predict the risk level of cardiovascular diseases.
How we built it
The FlowChart of the project was like first we have collected the dataset then the data was analyzed, after the analyzation process Data Preprocessing was done. Once we have preprocessed the data then we went for modelling and modelling optimization and at the last we deployed our model
Challenges we ran into
There were few challenges we ran into like getting the perfect dataset to make the prediction and get the highest accuracy.
Accomplishments that we're proud of
It was really a great learning experience working on this project . Me and my team member was working on Deep learning for the first time so it was hard for us but we were successful to learn about Deep learning and working on it in a project based manner . We were also able to learn about its possibilities in future . Coming from the same university , we knew each other’s strength and were moving forward together as a team . Team work was our key to successfully completing our project
What's next for CardioVascular Disease Prediction Using Machine Learning
Our current project is implemented on a static dataset . It can be enhanced with dynamic dataset with real time inputs . Better datasets , more training will lead to much better results and eliminate all possible limitations.