Colour-based wound assessment application for android smartphone / Nicholas Aaron Nigel

This study focuses on the subject of chronic wounds and the associated wound assessment. The high cost of medical expenses coupled with the overall difficulties in properly assessing a wound has proven to be a major problem associated with chronic wounds. The aim of this study is to create a functional application that would help healthcare workers to capture wound images using their smartphones and assess wound stages based on wound colour from the acquired image. Data on real wound images obtained over the internet were processed using the proposed application to gauge its accuracy in assessing the wound stage. A usability test was also conducted to obtain data about the application’s performance. GrabCut is chosen as the algorithm for image segmentation because of its easy-to-use algorithm and HSV colour model is chosen as the main colour model in assessing wound colour. The application overall performed to its minimum expectation, whereby it can detect wound colours and assign stages albeit with rather low accuracy. Based on the result, it can be concluded that the application can perform its function but it will need more calibration and tweaking in order for it to accurately assign the appropriate wound stage.