ARTIfiCIAL INTELLIGENCE-DRIVEN SYNOPTIC REPORTING FOR NEURO-ONCOLOGY TREATMENT PLANNING: A PROTOCOL FOR MULTI-STAGE EVALUATION

Abstract

Abstract AIMS Artificial intelligence (AI) is increasingly used in neuro-oncology for tumour detection and characterisation. However, its integration into patient workflows remains underdeveloped. This study aims to design an AI- driven synoptic reporting tool to assist multidisciplinary team (MDT) members in treatment decision-making for meningiomas, pituitary adenomas, and gliomas. The tool will synthesise key clinical and radiological data into a standardised structured report, optimising preoperative planning, surveillance imaging and surgical strategy formulation. METHODS Firstly, a systematic literature review will examine current AI applications in synoptic reporting to ensure nov- elty and feasibility. Next, we propose a two-stage cross-sectional survey of a large tertiary neurosciences centre prior to dissem- ination to national and international neuro-oncology societies. Members of neuro-oncology, skull-base and stereotactic radiosurgery MDTs —including neurosurgeons, neuroradiologists, oncologists, and pathologists— will identify critical parameters for AI-driven synoptic reporting. Using a mixed-methods approach, partici- pants will assess essential reporting elements, desired functionalities, and the potential impact on clinical work- flow. Finally, process mapping will outline the current preoperative workflow, identifying key decision points and opportunities for AI integration. RESULTS The survey is expected to reveal consensus on key elements required for AI-driven synoptic reporting, including tumour and patient specific radiological features. Process mapping highlighted workflow inefficiencies and socio-ethical concerns regarding AI and large-scale data usage. These insights will guide the development of an optimised reporting template, ensuring clinical relevance and practical applicability. CONCLUSION This study will establish an AI-assisted synoptic reporting model designed to enhance treatment planning in neuro-oncology. By aligning with MDT priorities, this tool aims to improve workflow efficiency, surgical pre- paredness and surveillance imaging. Our findings inform the development of a gold-standard structured re- porting framework tailored to pre-operative treatment and surveillance imaging needs.

Ähnliche Arbeiten