SSHNN: Semi-Supervised Hybrid NAS Network for Echocardiographic Image Segmentation

TL;DR

This paper introduces SSHNN, a semi-supervised hybrid neural architecture search network that improves echocardiographic image segmentation by enhancing feature fusion, global context understanding, and leveraging semi-supervised learning.

Contribution

The paper proposes a novel semi-supervised hybrid NAS network with a convolution-based feature fusion, Transformer integration, and a U-shaped decoder for superior medical image segmentation.

Findings

SSHNN outperforms state-of-the-art methods on CAMUS dataset

The semi-supervised Mean-Teacher algorithm effectively handles limited labeled data

Convolution-based feature fusion enhances detail preservation in segmentation

Abstract

Accurate medical image segmentation especially for echocardiographic images with unmissable noise requires elaborate network design. Compared with manual design, Neural Architecture Search (NAS) realizes better segmentation results due to larger search space and automatic optimization, but most of the existing methods are weak in layer-wise feature aggregation and adopt a ``strong encoder, weak decoder" structure, insufficient to handle global relationships and local details. To resolve these issues, we propose a novel semi-supervised hybrid NAS network for accurate medical image segmentation termed SSHNN. In SSHNN, we creatively use convolution operation in layer-wise feature fusion instead of normalized scalars to avoid losing details, making NAS a stronger encoder. Moreover, Transformers are introduced for the compensation of global context and U-shaped decoder is designed to efficiently connect global context with local features. Specifically, we implement a semi-supervised algorithm Mean-Teacher to overcome the limited volume problem of labeled medical image dataset. Extensive experiments on CAMUS echocardiography dataset demonstrate that SSHNN outperforms state-of-the-art approaches and realizes accurate segmentation. Code will be made publicly available.