Development and evaluation of a 3D annotation software for interactive COVID-19 lesion segmentation in chest CT

TL;DR

This paper introduces an interactive 3D software tool for COVID-19 lesion segmentation in chest CT scans, combining automatic and manual steps to improve efficiency and consistency, and evaluates its performance across different user groups.

Contribution

The work presents a novel interactive segmentation software that integrates automatic level-set and shape modeling with manual correction, enhancing speed and reliability over manual methods.

Findings

The software achieved satisfactory agreement between radiologists and engineers.

Segmentation process was significantly faster than manual methods.

Inter-observer variability is influenced by subjective factors, highlighting uncertainty in results.

Abstract

Segmentation of COVID-19 lesions from chest CT scans is of great importance for better diagnosing the disease and investigating its extent. However, manual segmentation can be very time consuming and subjective, given the lesions' large variation in shape, size and position. On the other hand, we still lack large manually segmented datasets that could be used for training machine learning-based models for fully automatic segmentation. In this work, we propose a new interactive and user-friendly tool for COVID-19 lesion segmentation, which works by alternating automatic steps (based on level-set segmentation and statistical shape modeling) with manual correction steps. The present software was tested by two different expertise groups: one group of three radiologists and one of three users with an engineering background. Promising segmentation results were obtained by both groups, which achieved satisfactory agreement both between- and within-group. Moreover, our interactive tool was shown to significantly speed up the lesion segmentation process, when compared to fully manual segmentation. Finally, we investigated inter-observer variability and how it is strongly influenced by several subjective factors, showing the importance for AI researchers and clinical doctors to be aware of the uncertainty in lesion segmentation results.