Introducing the automated software for skin image processing and diagnosing melanoma
A. Meisamy1, M.A. Ansari1*
1- Optical Bio-Optical Imaging Lab, Laser and Plasma Research Institute, Shahid Beheshti University, Tehran, Iran
Abstract
Melanoma is the most dangerous skin cancer that quickly grows and spreads to other organs. Monitoring skin lesions and their changes can help diagnose and treat them early and decrease mortality. Dermatoscopy and processing result images are valuable and essential tools to allow physicians in melanoma detection. We propose software as a graphical user interface for dermatoscope, developed in the Bio-Optical Imaging Laboratory at Shahid Beheshti university (Parto Ava Atlas, Iran), which can process skin images. In addition, using the ABCD algorithm, users can automatically measure critical indexes such as asymmetry, border irregularity, color, and diameter of lesions. Based on the score of each index, the software automatically analyzes and classifies whether the lesion is benign or malignant. This software is written in python, uses the OpenCV package, a helpful library in computer vision and image processing, and performs well on actual data. In the near future, we are going to use machine learning algorithms in our proposed software for the classification and segmentation of skin images.
This work is based upon research funded by Iran national Science Foundation (INSF) under project number 98029460.
Speaker
Mohammad Ali Ansari
Laser and Plasma Research Institute, Shahid Beheshti University
Iran
Discussion
Ask question
