Implantable Adaptive Cells: A Novel Enhancement for Pre-Trained U-Nets in Medical Image Segmentation

TL;DR

This paper proposes Implantable Adaptive Cells (IACs), small modules inserted into pre-trained U-Nets via gradient-based Neural Architecture Search, to improve medical image segmentation accuracy without full retraining.

Contribution

Introducing IACs, a novel method for enhancing pre-trained U-Nets using NAS, achieving significant accuracy gains with minimal retraining effort.

Findings

Segmentation accuracy improved by approximately 5 percentage points.

Up to 11% improvement in the best cases across datasets.

Method is cost-effective and adaptable to other architectures.

Abstract

This paper introduces a novel approach to enhance the performance of pre-trained neural networks in medical image segmentation using gradient-based Neural Architecture Search (NAS) methods. We present the concept of Implantable Adaptive Cell (IAC), small modules identified through Partially-Connected DARTS based approach, designed to be injected into the skip connections of an existing and already trained U-shaped model. Unlike traditional NAS methods, our approach refines existing architectures without full retraining. Experiments on four medical datasets with MRI and CT images show consistent accuracy improvements on various U-Net configurations, with segmentation accuracy gain by approximately 5 percentage points across all validation datasets, with improvements reaching up to 11\%pt in the best-performing cases. The findings of this study not only offer a cost-effective alternative to the complete overhaul of complex models for performance upgrades but also indicate the potential applicability of our method to other architectures and problem domains.