Automated 3D patient model generation using ML technique for hyperthermia treatment planning

Targeted hyperthermia treatment administered as an adjuvant to radiotherapy requires heterogeneous 3D patient model for treatment planning. Treatment planning is crucial for targeted delivery of hyperthermia treatment to the tumor target while sparing the surrounding healthy organs. Computed tomography (CT) data of the patient has poor soft-tissue contrast. As a result, automating patient image segmentation for 3D model generation is a time-consuming semi-automated process. Automated patient model generation is an amalgamation of technology and medical expertise in radiology. In this work, we present our preliminary results on machine learning approach for automated generation of 3D patient models for hyperthermia treatment planning using patient CT data of cervical cancer patients.