Arsenicosis, a chronic health condition triggered by long-term exposure to unsafe levels of arsenic, is a significant concern, especially in places where people rely on drinking water that might be contaminated with arsenic. In areas dealing with arsenic contamination, like remote and underserved regions, the absence of dermatologists and proper healthcare makes it tough to spot skin problems early. We need a simple and accessible way to detect signs of arsenic intoxication on the skin, like keratosis, so we can act fast and prevent further harm.
- Develop and train a deep-learning model that can detect the manifestations caused by arsenic toxicity if present i.e. classify the image into arsenic infected or healthy.
- Predict the severity of the manifestation into low, moderate, and severe if infected.
- Deploy the model in a user-friendly web application, making it accessible to healthcare providers and patients for preliminary skin disease screening.
- A deep-learning model that can classify skin images into healthy or infected categories, along with the severity of the infected skin that is trained on a dataset that contains instances of diffuse melanosis, keratosis, etc on palms and soles.
- An entire solution containing a User Interface, trained model, and a backend connecting both.
- A platform to find nearest dermatologist easily.
