Advancing Healthcare: Innovative ML Approaches for Improved Medical Imaging in Data-Constrained Environments

TL;DR

The proposed CAT-U-Net framework adds an extra concatenation layer with downsampling parts, thereby improving its ability to learn from limited data while maintaining patient privacy, and has the potential to make a big difference in medical image diagnostics in settings with limited data.

Abstract

Healthcare industries face challenges when experiencing rare diseases due to limited samples. Artificial Intelligence (AI) communities overcome this situation to create synthetic data which is an ethical and privacy issue in the medical domain. This research introduces the CAT-U-Net framework as a new approach to overcome these limitations, which enhances feature extraction from medical images without the need for large datasets. The proposed framework adds an extra concatenation layer with downsampling parts, thereby improving its ability to learn from limited data while maintaining patient privacy. To validate, the proposed framework's robustness, different medical conditioning datasets were utilized including COVID-19, brain tumors, and wrist fractures. The framework achieved nearly 98% reconstruction accuracy, with a Dice coefficient close to 0.946. The proposed CAT-U-Net has the potential to make a big difference in medical image diagnostics in settings with limited data.

Figures (7)

• Figure 1: The pipeline of proposed CAT-U-Net frameworks
• Figure 2: The proposed CAT-U-Net framework architecture with concatenation layers
• Figure 3: Performance Analysis of the Proposed Framework Across Different Datasets
• Figure 4: Dice-Cofficient value for Bone fracture dataset
• Figure 5: Proposed framework Confusion matrix