Title: Semantics-driven annotation of 3D medical data to support information retrieval

Type: Ph.D (in progress)

Abstract: Research objective of the Ph.D. is to develop effective approaches to support semantics-driven analysis and management of patient specific 3D data by exploiting bio-medical domain ontologies. This will be achieved by semantic annotation method which allows creating correspondence between patient specific 3D anatomical model or model parts, and conceptual tags present in the domain ontology; once the content is analyzed and its relevant constituents annotated with formalized knowledge, it can be exploited further for searching and browsing of patient-specific 3D data from distributed medical repositories. In this context, my research is expected to define and develop a framework that supports part-based semantic annotation of patient-specific 3D anatomical model. According to the current state-of-the-art, it has been observed that still significant semantic gap exists between available 3D digital patient data and formalized domain knowledge. Because of that creation of patient-specific 3D atlas is truly a challenging research area. The main expected contribution of my Ph.D. thesis will be the creation of patient-specific 3D atlas of Carpal bones to capture and preserve accurate quantitative description of patient-specific 3D geometrical aspects. Further, we aim to exploit patient-specific whole and part-based 3D annotation together with the formalized knowledge in a web-base environment to facilitate 3D medical data retrieval, and to inspect the temporal change of a specific anatomical part for the clinical follow-up.

Author: Imon Banerjee

Advisors: Dr. Michela Spagnuolo, Dr. Giuseppe Patanè

University: DIBRIS, University of Genova

Defence Date (tentative): April, 2016