Implantable Adaptive Cells: A Novel Enhancement for Pre-Trained U-Nets in Medical Image Segmentation
This provides a cost-effective method for upgrading segmentation models in medical imaging, though it is incremental as it refines existing architectures.
The paper tackles the problem of improving medical image segmentation by enhancing pre-trained U-Nets without full retraining, using Implantable Adaptive Cells identified through gradient-based NAS, resulting in accuracy gains of about 5 percentage points across datasets, with up to 11% in best cases.
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.