Automatic segmentation method of pelvic floor levator hiatus in ultrasound using a self-normalising neural network

TL;DR

This paper introduces a fully automatic CNN-based method utilizing SELU activation for segmenting the pelvic floor levator hiatus in ultrasound images, achieving high accuracy and overcoming training memory constraints.

Contribution

The study applies the self-normalising SELU activation in CNNs for medical image segmentation, demonstrating its advantages over traditional batch-normalised networks in pelvic ultrasound analysis.

Findings

Median Dice coefficient of 0.90 indicating high segmentation accuracy

SELU-based CNN outperforms U-Net without batch normalization

Method achieves equivalent performance to manual operators

Abstract

Segmentation of the levator hiatus in ultrasound allows to extract biometrics which are of importance for pelvic floor disorder assessment. In this work, we present a fully automatic method using a convolutional neural network (CNN) to outline the levator hiatus in a 2D image extracted from a 3D ultrasound volume. In particular, our method uses a recently developed scaled exponential linear unit (SELU) as a nonlinear self-normalising activation function, which for the first time has been applied in medical imaging with CNN. SELU has important advantages such as being parameter-free and mini-batch independent, which may help to overcome memory constraints during training. A dataset with 91 images from 35 patients during Valsalva, contraction and rest, all labelled by three operators, is used for training and evaluation in a leave-one-patient-out cross-validation. Results show a median Dice similarity coefficient of 0.90 with an interquartile range of 0.08, with equivalent performance to the three operators (with a Williams' index of 1.03), and outperforming a U-Net architecture without the need for batch normalisation. We conclude that the proposed fully automatic method achieved equivalent accuracy in segmenting the pelvic floor levator hiatus compared to a previous semi-automatic approach.