Evaluating Artificial Intelligence Tool Chatgpt Versus Multidisciplinary Team Decision-making in Prostate Cancer Care: a Comparative Analysis
Year: 2024
Type: article
Abstract: Chat Generative Pre- Trained Transformer (ChatGPT) and other Large Language Models (LLM's) are used as a patient education tool in Healthcare and urology, yet nothing has been published on their practical use within the cancer multi-disciplinary team (MDT) meeting. Here we test ChatGPT's treatment recommendations for patients with prostate cancer against those made in Real world multi-disciplinary team (MDT) meeting. 15 prostate cancer patient's details copied unedited from the MDT meeting containing their age, PSA, MRI Staging and prostate biopsy histopathology and articial intelligence (AI) was directed to recommend the best treatment option for the specic patient according to the guideline linked and then compared directly to the real world MDT meeting outcome. Urologist reviewed the answer and decided if concordance has been met or not. All 15 patients that were recommended treatment option by the AI was concordant with the real world decision made by the MDT meeting. 2 localized ca prostate patients recommended radical prostectomy (RP) out of which one patient underwent RP and another case RP deferred due to patient preference. 2 very low risk patient recommended active surveillance, 3 locally advanced ca prostate cases were recommended multimodality treatment with radiotheraphy and long term androgen deprivation therapy (ADT), 6 cases of metastatic ca prostate were recommended chemotherapy in addition to ADT. 2 case of metastatic castrate resistant ca prostate were recommended docetaxel chemotheraphy. ChatGPT has huge implications for streamlining and speeding up MDT meetings. We believed that Urologist must remain both vigilant and open to the capabilities of AI LLM's and to embrace this technology for the bettering of our Healthcare systems and patient care.
Authors: G Latha, Dhinesh A. M, Nikhil Tharkade
Institution: Madurai Medical College
Language: English
Subfield: Health Informatics
Field: Medicine
Domain: Health Sciences
Open Access status: diamond
OpenAlex ID: W4404152401