New framework offers interpretable AI for medical data classification

Researchers have developed a new framework for interpretable classification of medical data using a statistically grounded approach. This method employs a Bernoulli Naïve Bayes model with $\chi^2$-guided statistical binarization to transform continuous variables into interpretable thresholds. The framework was tested on three benchmark datasets for diabetes, breast cancer, and heart failure prediction, achieving high AUC scores and improved probabilistic reliability through calibration analysis. This approach aims to enhance trust and generalizability of AI in healthcare by providing clinically meaningful decision rules and reproducible inference. AI

IMPACT This framework could enhance trust and adoption of AI in healthcare by providing transparent and reproducible classification models.

RANK_REASON The item is a research paper published on arXiv detailing a new framework for medical data classification. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• alphaXiv
• arXiv
• Bernoulli Naïve Bayes
• CatalyzeX
• DagsHub
• Gotit.pub
• Heart failure prediction based on random forest algorithm using genetic algorithm for feature selection
• Hugging Face
• IArxiv
• Influence Flower
• Pima Indians diabetes mellitus classification based on machine learning (ML) algorithms
• ScienceCast
• Wisconsin Breast Cancer Coalition

• paper
• model release
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →