Supporting Dataset for thesis - "From Conventional to Computational: Artificial Intelligence-Driven Advances in the Evaluation and Management of Oral Potentially Malignant Disorders"

Supporting Data for Thesis<b>Thesis Title:</b> "From Conventional to Computational: Artificial Intelligence-Driven Advances in the Evaluation and Management of Oral Potentially Malignant Disorders"This data package contains supporting resources for the doctoral thesis titled above. The research presented in the thesis comprises three original studies derived from the two primary datasets included in these files. Ethical approval for the collection and use of all data was obtained from the relevant institutional review boards.<b>Dataset 1: Oral Potentially Malignant Disorders (OPMD) Dataset</b> This dataset was retrieved from the Hospital Authority Clinical Management System (HA-CMS). It contains longitudinal clinical data for patients diagnosed with oral leukoplakia and oral lichenoid mucositis. The primary objective of this dataset is to facilitate the development of predictive models for malignant transformation in OPMD cases.<b>Dataset 2: OED Clinical Image Dataset</b> This dataset consists of high-resolution clinical photographs of Oral Epithelial Dysplasia (OED) and oral leukoplakia. These images were curated to support the evaluation and clinical management of leukoplakia, specifically focusing on the integration of artificial intelligence for the visual assessment of dysplastic features.For datasets, comprehensive documentation explaining individual variables, coding systems, and metadata is provided within the respective spreadsheets. The study protocol was approved by the Institutional Review Board (IRB) of the University of Hong Kong/Hospital Authority Hong Kong West Cluster (Reference: UW23-094). To safeguard patient privacy and maintain strict confidentiality, all clinical data were meticulously anonymized and stripped of personal identifiers by the research team prior to the commencement of data analysis.