Snehana is a simple machine learning and artificial intelligence (ML/AI) based method for detecting skin texture (oily or dry) and prescribing the appropriate treatment.
Patients have a variety of skin disorders. The biggest difficulty is that they have different types of skin. Various skin illnesses can develop depending on the type of skin. When the skin becomes excessively dry, it begins to peel, which can lead to skin illnesses such as psoriasis and eczema. Similarly, oily skin can develop to pimples, which can contribute to scarring. As a result, determining your skin type is critical.
It takes a snapshot of the user from a web-portal and displays real-time data about the texture of their skin and any diseases they may have.
For the model and deployment, we used Tensorflow, Keras, Numpy, Pandas, Opencv, and Matplotlib, as well as HTML, CSS, and Tensorflow. js
The first issue we encountered was model accuracy, which was initially about 5% to 20% but later increased to 90.31 percent. Second, we had challenges with Realtime Detection and Integration with Deployment. We have had troubles with Recommender's system as well.
The model has been successfully put on a web-app and can distinguish between oily and dry skin textures. It can also detect skin illness if any is present. The model now has an accuracy of 90.31%, which is a fantastic achievement for newcomers!!
For the model and deployment, we used Tensorflow, Keras, Numpy, Pandas, Opencv, and Matplotlib, as well as HTML, CSS, and Tensorflow. js
The model needs to be improved further to achieve greater accuracy, because when it comes to healthcare, the level of certainty should be as high as possible. Furthermore, the model can be trained so that the user receives an initial prescription as well as hospital recommendations for contacting a dermatologist.
