An Education Bot for Radiation Safety in Radiotherapy
J Chow1*, L Sanders2, S Siddique3, L Xu4, S Ali3, R Nathanael3, K Li2, (1) Princess Margaret Cancer Centre, Toronto, ON, CA, (2) York University, Toronto, ON, CA, (3) Ryerson University, Toronto, ON, CA, (4) University of Toronto, Toronto, ON, CA
Presentations
PO-GePV-T-87 (Sunday, 7/10/2022) [Eastern Time (GMT-4)]
ePoster Forums
Purpose: An Education Bot was created to serve as a tool for radiation safety training in radiotherapy using AI and machine learning. The Bot was designed to help radiation staff working in the cancer center to learn or refresh knowledge on radiation safety requested in radiotherapy practice.
Methods: The Bot was created using IBM’s Watson Assistant, which was chosen as it provides a user-friendly interface, along with simple but powerful integration tools, that allow it to be integrated into different channels such as text messages, WebChat, or applications such as WhatsApp and Discord. A group of 15 questions regarding the general radiation safety practice in radiotherapy was selected to provide a comfortable length of communication process. Specific intent of the user’s input was detected by the Watson Assistant using the tool of Intents, based on the Natural Language Processing (NLP) using machine learning. The Bot was pre-tested by the radiation staff in the cancer center to collect comments for continuous upgrade and fine-tuning.
Results: The Education Bot was built and embedded to a website. The Bot has a frontend window started an introductory message to say hello to the user and ask for the user’s name to make the experience more personalized. The Bot then asked the user some radiation safety questions. When the user was not able to follow the communication, the Bot would provide sequential guidance to the user. The Bot received many positive comments from the users with many valuable feedbacks for future improvement. Human variations in communication were reduced by the NLP based on the users’ comments.
Conclusion: It is concluded that an AI-assisted Education Bot can help the radiation staff faster to gain radiation safety knowledge in radiotherapy. The Bot with character supported by machine learning can provide information regarding radiation safety to radiation staff.
Funding Support, Disclosures, and Conflict of Interest: This work was supported by a Canadian Institutes of Health Research Planning and Dissemination Grant: Institute Community Support under grant number CIHR PCS168296.
Keywords
Radiation Protection, Computer Software, Radiation Therapy
Taxonomy
Contact Email