 Development of computer database and application of machine learning algorithms to assist the real time differentiation between adenomatous and hyperplastic subcentrim THESIS TOPIC PROPOSAL Institute: University of Szeged theoretical medicine Theoretical Medicine Doctoral School Thesis supervisor: László Madácsy Location of studies (in Hungarian): Endokapszula Magánorvosi Centrum Székesfehérvár Abbreviation of location of studies: ENDOK Description of the research topic: The prediction of histology of polyps during colonoscopy can support the resect and discard strategy. The use of recent classification needs expertise and routine. A computer algorithm can be capable to predict the histology based on preprocessed high-resolution images or BASIC classification descriptors. The aim of the current study is to develop a cloud-based deep learning algorithm to predict the polyp histology on high-resolution images made with different virtual chromoendoscopic technique. Required language skills: english Number of students who can be accepted: 1 Deadline for application: 2024-12-31 |