Abstract:
"The purpose of this research project is to improve skin disease detection in Sri Lanka through
the creation of a mobile application by using ML. The software offers patients and healthcare
providers a user-friendly, precise, and non-invasive tool by utilizing Convolutional Neural
Networks (CNN) in conjunction with multimodal imaging methods. In light of the limited
resources available for healthcare, it tackles the common problem of accurately and promptly
identifying skin diseases. A three-tier system design is used in the architecture of the
application to securely handle and store sensitive medical data. A user-friendly presentation
tier is integrated with data tiers and robust logic.
The performance measurements show that the app has good diagnosis accuracy due to
significant picture data augmentation and machine learning model training. By providing a
scalable and economical method that has the potential to be widely adopted in settings with
limited resources, this work makes a valuable contribution to the area of medical informatics."
