Voice-assisted Image Labelling for Endoscopic Ultrasound Classification using Neural Networks

TL;DR

This paper introduces a multi-modal CNN that combines voice comments and ultrasound images to automatically label endoscopic ultrasound images, reducing manual labeling effort and improving classification accuracy.

Contribution

It presents a novel neural network architecture that integrates verbal comments with image data for ultrasound image labeling, enhancing accuracy and efficiency.

Findings

Achieved 76% accuracy in image classification

Voice comments improve labeling performance

Reduces manual dataset annotation effort

Abstract

Ultrasound imaging is a commonly used technology for visualising patient anatomy in real-time during diagnostic and therapeutic procedures. High operator dependency and low reproducibility make ultrasound imaging and interpretation challenging with a steep learning curve. Automatic image classification using deep learning has the potential to overcome some of these challenges by supporting ultrasound training in novices, as well as aiding ultrasound image interpretation in patient with complex pathology for more experienced practitioners. However, the use of deep learning methods requires a large amount of data in order to provide accurate results. Labelling large ultrasound datasets is a challenging task because labels are retrospectively assigned to 2D images without the 3D spatial context available in vivo or that would be inferred while visually tracking structures between frames during the procedure. In this work, we propose a multi-modal convolutional neural network (CNN) architecture that labels endoscopic ultrasound (EUS) images from raw verbal comments provided by a clinician during the procedure. We use a CNN composed of two branches, one for voice data and another for image data, which are joined to predict image labels from the spoken names of anatomical landmarks. The network was trained using recorded verbal comments from expert operators. Our results show a prediction accuracy of 76% at image level on a dataset with 5 different labels. We conclude that the addition of spoken commentaries can increase the performance of ultrasound image classification, and eliminate the burden of manually labelling large EUS datasets necessary for deep learning applications.