ESTRO 2025 - Abstract Book

S2341

Interdisciplinary – Other

ESTRO 2025

138

Digital Poster Optimizing Radiotherapy Appointment Scheduling through Artificial Intelligence: A Lean Management Based Approach samia khalfi 1,2,3 , touria bouhafa 2 1 radiotherapy, allal ben abdellah university, Fès, Morocco. 2 radiotherapy, hassan II university hospital, Fès, Morocco. 3 research institute, IRC, Fès, Morocco Purpose/Objective Efficient scheduling in radiotherapy is crucial to ensure timely patient care and optimal resource utilization. However, manual scheduling processes often lead to inefficiencies, including delays in treatment initiation and suboptimal resource allocation. This study integrates Lean Management methodologies with artificial intelligence (AI) to address these challenges in a radiotherapy department. This study aims to develop and implement an AI-driven system that automates and optimizes appointment scheduling in radiotherapy, improving patient prioritization and resource management while reducing delays and manual workload. Material/Methods Lean Management techniques were employed to map and analyze current scheduling workflows, identifying key inefficiencies. An AI-based system, using machine learning algorithms such as Random Forest and KNeighbors Regressor, was developed to classify patient urgency levels and optimize scheduling dynamically. The system’s performance was evaluated based on its ability to accurately prioritize urgent cases and minimize scheduling conflicts. Results Algorithm Performance: The AI system demonstrated a high level of accuracy in predicting patient urgency levels, with the Random Forest algorithm outperforming others. The model achieved an accuracy rate of 92% in identifying high-priority cases, which directly impacted the scheduling process by ensuring that urgent patients were assigned the earliest available slots. Reduction in Waiting Times: Implementation of the AI-driven system led to a significant reduction in patient waiting times. The average delay for urgent cases was reduced from 5 days to less than 2 days, marking a 60% improvement. Additionally, the system reduced the overall variability in scheduling, leading to more predictable and manageable daily workloads. Resource Utilization: The optimized scheduling process resulted in better utilization of radiotherapy equipment and personnel. There was a 15% increase in the utilization rate of available treatment slots, with fewer instances of under- or overbooking. This improvement translated into higher throughput without compromising the quality of care. Clinician Efficiency: The automation of scheduling freed up approximately 20% of clinicians' time previously spent on manual scheduling tasks. This allowed clinicians to focus more on patient care and other critical activities, contributing to overall department efficiency. Conclusion Integrating AI with Lean Management principles in radiotherapy scheduling offers a robust solution to current inefficiencies, enhancing both patient care and operational efficiency. This approach not only improves scheduling accuracy and reduces delays but also optimizes resource utilization and frees up valuable clinical time, demonstrating potential for broader application in healthcare settings where scheduling complexity and resource management are critical.

Keywords: Radiotherapy, Artificial Intelligence,Management