The problem GUI for Diseases Classification solves

It is always said that “happiness is the highest form of health” and one should always take care of his/her health in every way possible. But sometimes knowingly or unknowingly we come in contact with many diseases which can be fatal sometimes. Sometimes even doctors facing difficulty to identify diseases and this leads to delay in treatment. As looking toward COVID pandemic, doctors and other medical proficiencies are too busy. So even after vaccination of COVID, there might some other diseases or virus approach in future. As Doctors are already busy with their current work, so it will be a problematic situation to control the diseases. This problem especially persists in rural and large populated areas. So, to solve this problem we have an amazing solution which will help the co-medical workers to identify some fatal disease and also helps the patients who living in remote areas as well.

Our Solution will help the medical staff ’s in such a way that in any of the critical condition, if doctors are not available to tackle few patient as due to emergency, so other medical supervisors can too use it as for primary detection and then consult to higher medical staff.
We are solving the following diseases:-

1. Diabetic Retinopathy (70.45%)
2. Breast Cancer Detection (98.25%)
3. Chronic Kidney Diseases Detection (99.74%)
4. COVID phenomenon detection (95.45%)
5. Parkinson Spiral through detection (50% & working for improving accuracy)
6. Heart Diseases Prediction (80.32%)
7. Brain Tumour Detection (working for improving accuracy)

As modern problem need modern solution, so for this problem we are providing a AI based diseases diagnostic system (DDS) which is a software that will detect the diseases. Our Solution will help the medical staff ’s in such a way that in any of the critical condition, if doctors are not available to tackle few patient as due to emergency, so other medical supervisors can too use it as for primary detection and then consult to higher medical

Challenges we ran into

The main challenge AI faces, especially in the healthcare domain, is the amount of data available for training and testing. With privacy and legal issues this poses a complex gap for data sharing. Moreover, annotation of data is a laborious task which often requires a specialist. The natural imbalance of the data available is another issue. In some cases, most of the data is of healthy subjects and does not include rare conditions that need to be detected or monitored. However, in other cases the situation is the complete reverse whereby the data mostly contains medically ill subjects.
Using transfer learning is the common approach used today to deal with the data limitation issue in which the trained models’ previous knowledge can be exploited. The training essentially fine tunes or adjusts the previous trained model using only a small amount of data to the medical task in hand. It has been proven that the previously trained models are able to hold some low-level information that is required in the current medical imaging task.
We have used different Machine learning and Deep Learning algorithms. We have experimented with different activation functions like, RELU, Sigmoid, Softmax. We have experimented with different matrices like accuracy, loss, F-1 score, recall, precision, etc to draw the conclusion. We have experimented with different layers in deep learning to improve our accuracy. We experimented creating a desktop application by making use of HTML, CSS as front-end but, we were not successful and we are working over it. The complete GUI is made in Python including Back-end and front-end. For Deep Learning model training purpose we made use of Kaggle and colab notebook which provide free GPU. The GUI created is in python and the speed of processing is little slow, So we are finding solution to it. We cannot do experiment from the live industry data and updated data. We have to get the data from kaggle to proceed for our further work.