KEY TAKEAWAYS
- The study aimed to develop and validate a machine learning model for a web-based AI-CDSS application to improve BC recurrence prediction.
- The AI-CDSS tool, enhanced by ChatGPT, showed promise in predicting BC recurrence but needs broader validation.
Breast cancer (BC) is a major global health issue, highlighting the need for improved methods of recurrence prediction and management. AI-based clinical decision support systems (AI-CDSS) using tools like ChatGPT offer a promising solution to enhance prediction accuracy and user accessibility.
Tai-Han Lin and the team aimed to develop and validate an advanced machine learning model for a web-based AI-CDSS application. It leverages ChatGPT’s question-and-answer guidance capabilities to enhance data preprocessing and model development, ultimately improving predictions of BC recurrence.
The study used data from the Tri-Service General Hospital breast cancer registry, including 3577 patients from 2004 to 2016. This data, sourced from a tertiary medical centre and its referral branches, provided diversity in the study population. The model was trained using data from 2004 to 2012 and validated with data from 2013 to 2016.
ChatGPT played a key role in preprocessing data and developing the model, particularly in hormone receptor categorization, age binning, and one-hot encoding. The synthetic minority oversampling technique addressed data set imbalances. Several algorithms were used, including light gradient-boosting machine, gradient boosting, and extreme gradient boosting, with their performance assessed using metrics like area under the curve, accuracy, sensitivity, and F1-score.
Results showed the light gradient-boosting machine model performed best, achieving an area under the curve of 0.80, closely followed by the gradient boosting and extreme gradient boosting models. The AI-CDSS tool’s web interface proved effective in clinical decision-making scenarios, demonstrating its use in personalized treatment planning and patient involvement.
The AI-CDSS tool, enhanced by ChatGPT, significantly advances BC recurrence prediction, enabling a more individualized and accessible approach for clinicians and patients. However, further validation in diverse clinical settings is recommended to confirm its efficacy and broaden its application.
No funding information was given.
Source: https://pubmed.ncbi.nlm.nih.gov/39231422/
Lin TH, Chung HY, Jian MJ, et al. (2024). “An Advanced Machine Learning Model for a Web-Based Artificial Intelligence-Based Clinical Decision Support System Application: Model Development and Validation Study.” J Med Internet Res. 2024;26:e56022. Published 2024 Sep 4. doi:10.2196/56022
