TRIPOD – Text-based RIsk PriOritization of Dermatological clinical notes


Skin cancers can be cured, as long as they’re diagnosed in time, which is often not the case. This work,part of DermAI project, will use retrospective data of NHS from CTH (“Consulta a Tempo e Horas”) to build the AI models to be incorporated in RSE-SIGA (a recent Integrated Access Management System), improving the risk assessment time. Besides image analysis, there is relevant information to be mined from the clinical notes that usually accompany a referral process, as demonstrated for other types of cancer.



Telemedicine solutions are especially relevant for the dermatological area given the mismatch between skin cancer numbers in Portugal and the scarcity of available resources and its uneven distribution on a region-level. Important developments have used skin lesions image analysis for diagnosis support or risk assessment. On the other hand, clinical notes, often consisting of free-text, comprise significant potential that can be explored. This thesis then aims at developing a set of methods for the analysis of unstructured clinical text for feature extraction, and their use for diagnostic classification or risk assessment.


Author: Catarina Dias

Type: MSc thesis

Partner: FEUP – Faculdade de Engenharia da Universidade do Porto

Year: 2021