Adaptive Smooth Activation for Improved Disease Diagnosis and Organ Segmentation from Radiology Scans

TL;DR

This paper introduces the Adaptive Smooth Activation Unit (ASAU), a new activation function that improves disease diagnosis and organ segmentation accuracy in medical imaging by enhancing gradient propagation.

Contribution

The study proposes ASAU, a novel activation function tailored for medical image analysis, demonstrating significant performance improvements over traditional activations in classification and segmentation tasks.

Findings

4.80% improvement in disease classification accuracy

1-3% increase in dice coefficient for liver segmentation

ASAU outperforms ReLU and other activations in experiments.

Abstract

In this study, we propose a new activation function, called Adaptive Smooth Activation Unit (ASAU), tailored for optimized gradient propagation, thereby enhancing the proficiency of convolutional networks in medical image analysis. We apply this new activation function to two important and commonly used general tasks in medical image analysis: automatic disease diagnosis and organ segmentation in CT and MRI. Our rigorous evaluation on the RadImageNet abdominal/pelvis (CT and MRI) dataset and Liver Tumor Segmentation Benchmark (LiTS) 2017 demonstrates that our ASAU-integrated frameworks not only achieve a substantial (4.80\%) improvement over ReLU in classification accuracy (disease detection) on abdominal CT and MRI but also achieves 1\%-3\% improvement in dice coefficient compared to widely used activations for `healthy liver tissue' segmentation. These improvements offer new baselines for developing a diagnostic tool, particularly for complex, challenging pathologies. The superior performance and adaptability of ASAU highlight its potential for integration into a wide range of image classification and segmentation tasks.