Developing a Deep Learning Classifier for Gout Management

Gout is a form of arthritis which results from hyperuricemia, the process where excess uric acid accumulates in the bloodstream. According to the NHANES survey, there are about 3.9% of total adults in the United States who do not suffer from gout, with 9.2 million individuals affected. In order for those suffering from gout to stay healthy, eating right and avoiding flare-ups will mean controlling one's diet and avoiding foods rich in purines. Unfortunately, dieting is quite troublesome. While eating meals, the majority of the individuals having gout have problems in looking at the plate of food and determining whether it contains dangerous amounts of purines.


As opposed to developing the usual ML model capable of classifying food into different categories, this work aims to develop a clinically relevant algorithm for categorizing food according to danger levels.