- 87 views
- 137 downloads
AI Assisted Mole Detection for Online Dermatology Triage in Telemedicine Settings
-
- Author / Creator
- Das, Debarpan
-
Skin moles are one of the most commonly occurring dermatological conditions prevalent nowadays. Early identification and diagnosis of moles are absolutely crucial since they often turn out to be precursors to serious conditions such as melanoma, a dangerous type of skin cancer. Therefore, to ensure an efficient treatment of cases based on their severity, they need to be assessed systematically. In this thesis, we present an artificial intelligence (AI )enabled triage tool to identify moles from images uploaded by patients to a teledermatology platform. The proposed approach employs NesT, one of the latest state-of-the-art transformer-based network for classification. Our system acts as a filter by sending a warning flag if a mole is detected. This can be used to help dermatologists set up consultation appointments in a physical setting by giving priority if the patient has a mole on their image.
A comparative study of the prediction performance of the different neural network models has been provided for different performance metrics of interest. The results presented in this thesis have been obtained from two sets of data, consisting of more than 26,000 clinical pictures with combined different dermatological conditions. Multiple experiments using different models yielded a macro-average recall value as high as 0.955, along with overall accuracy and macro-average precision values of 0.962 and 0.958, respectively.
-
- Subjects / Keywords
-
- Graduation date
- Fall 2023
-
- Type of Item
- Thesis
-
- Degree
- Master of Science
-
- License
- This thesis is made available by the University of Alberta Libraries with permission of the copyright owner solely for non-commercial purposes. This thesis, or any portion thereof, may not otherwise be copied or reproduced without the written consent of the copyright owner, except to the extent permitted by Canadian copyright law.