SHAP-Integrated Convolutional Diagnostic Networks for Feature-Selective Medical Analysis
This work addresses data scarcity in medical image analysis for healthcare applications, but it appears incremental as it builds on existing CNN and SHAP methods.
The study tackled the challenge of limited medical datasets due to privacy regulations by introducing an interpretable feature selection method, achieving over 97% accuracy on pneumonia and breast cancer classification tasks and outperforming four popular CNN models.
This study introduces the SHAP-integrated convolutional diagnostic network (SICDN), an interpretable feature selection method designed for limited datasets, to address the challenge posed by data privacy regulations that restrict access to medical datasets. The SICDN model was tested on classification tasks using pneumonia and breast cancer datasets, demonstrating over 97% accuracy and surpassing four popular CNN models. We also integrated a historical weighted moving average technique to enhance feature selection. The SICDN shows potential in medical image prediction, with the code available on https://github.com/AIPMLab/SICDN.