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

View PDF HTML (experimental) 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. Subjects: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI) Cite as: arXiv:2405.03420 [cs.CV] (or arXiv:2405.03420v2 [cs.CV] for this version) https://doi.org/10.48550/arXiv.2405.03420 arXiv-issued DOI via DataCite Submission history From: Marcin Denkowski [view email] [v1] Mon, 6 May 2024 12:40:15 UTC (2,732 KB) [v2] Thu, 6 Mar 2025 12:52:29 UTC (1,991 KB)