KIRAL aims to improve, flexibilise and individualise precision surgical care for patients with liver tumours. AI-supported algorithms are used to make patient-specific predictions about the expected effects of neoadjuvant therapy in order to identify patients who would benefit from primary resection. All image data after initial diagnosis, in combination with histological findings, serves as the basis for this. In the case of surgical treatment, preoperative resection planning is updated and interactively adjusted through AI-supported processing of intraoperative image information. Relevant anatomical structures, such as tumours and metastases, are recorded in terms of their size, number and intrahepatic location, as well as their proximity to the vessels, using IOUS and integrated into the existing 3D reconstruction or used to update the latter. This modified 3D reconstruction is in turn interactively visualised intraoperatively using augmented reality (AR), enabling the updated resection planning (updated resection areas, updated residual liver volume estimation) to be carried out. This involves the use of contactless interaction modalities (voice commands and hand gestures). In addition, a self-sufficient, remote training environment is created in which liver resections, including IOUS and resection planning, can be planned, simulated and trained using VR/AR applications. This environment is to be used for preclinical evaluation of the newly developed techniques.
