A BERT-Based Chatbot to Support Cancer Treatment Follow-Up

The aftermath of primary cancer treatment presents a multitude of challenges for patients, necessitating prolonged recovery periods that can span months or even years. Survivors contend with a range of debilitating side effects, including fatigue, constant pain, lymphedema, weight fluctuations, swallowing difficulties, and menopause symptoms. These adversities often demand consistent counseling, compelling patients to visit hospitals regularly. However, the advent of advanced Artificial Intelligence (AI) offers a transformative solution. AI-powered chatbots, also known as conversational agents, emerge as potent allies capable of providing mental support, answering queries, diagnosing issues based on texted symptoms, and more. This paper explores the potential of AI-driven conversational agents to address the unique needs of cancer patients during their recovery phase. Drawing inspiration from successful applications in other chronic diseases, the proposed model capitalizes on the capabilities of the Bert-large-cased-whole-word-masking model. Demonstrating an impressive F1 score of 93.46%, this model signifies its efficacy in serving as an intelligent ally for cancer survivors. By facilitating mental support and information dissemination, the proposed conversational agent system aligns with the evolving landscape of AI-driven healthcare, forging a path toward enhanced patient care and improved quality of life during the recovery journey.