The problem HomeCare.ai solves

• Whenever a patient visits a hospital it takes a significant amount of time before the updated health reports of the patient arrives making it difficult for the proper detection and hence decision making for the health official.
• Also, it has now become unsafe to go to the hospital every time we feel unwell, since there is a risk of getting affected by COVID 19. The pandemic has caused an influx in hospital cases, and the limitations on hospital beds have people wondering whether their symptoms are severe enough to warrant a doctor's appointment.
• Meanwhile, others experience ailments but are unable to afford a visit to the doctor due to a lack of proper health care. Further, if the patient recognizes his/her symptoms, and if somehow we can tell him what is the disease he is likely to be affected with then accordingly precautions can be taken at home only.
• Since, everyone should have easy access to great health care there is a need to connect patients virtually with doctors in an efficient manner.
• In a digital world affected by COVID 19, telemedicine is more necessary than ever to improve the quality and accessibility of medical care to distant users and also improve the decision making process of clinicians as well as help people know what disease they are likely to be infected with and what precautionary measures can be taken with the help of Artificial Intelligence.
• So we have designed a solution deployable in a healthcare setting that can help to improve the quality and accessibility of medical care facilities to distant users, help in remote diagnosis of patients by connecting them remotely to the doctors around the globe in an efficient manner, disease detection through analysis of symptoms followed by proper precautions and drug recommendations, provided mutli-disease computer aided diagnosis system, supports lab test appointment facility, patient health management and also aid in the decision making process of doctors.

Challenges we ran into

• Training of deep learning and machine learning models with high accuracy and low complexity. To tackle this, we first gathered dataset from kaggle open source platform and used various hyper parameter tuning techniques to improve our models such as augmentations, early stopping, random search, feature selection methods etc.
• Connecting patients with doctors over video call. To achieve this, we used dailyjs API