A paid machine learning project on multi-label chest x-ray classification. This project is part of funded research. The NIH Chest X-ray Dataset was used and I managed to obtain a classification accuracy of ~75% on this dataset using the top 6 occurring classes. Achieving this accuracy was difficult since the dataset was highly imbalanced and the x-rays them self are difficult to interpret. The accuracy obtained fulfilled the requirements of the clients and thus the project was considered successful.