A Weakly Supervised Consistency-based Learning Method for COVID-19 Segmentation in CT Images

Abstract

Coronavirus Disease 2019 has spread aggressively across the world causing an existential health crisis. Thus, having a system that automatically detects COVID-19 in tomography (CT) images can assist in quantifying the severity of the illness. Unfortunately, labelling chest CT scans requires significant domain expertise, time, and effort. We address these labelling challenges by only requiring point annotations, a single pixel for each infected region on a CT image. This labeling scheme allows annotators to label a pixel in a likely infected region, only taking 1-3 seconds, as opposed to 10-15 seconds to segment a region. Conventionally, segmentation models train on point-level annotations using the crossentropy loss function on these labels. However, these models often suffer from low precision. Thus, we propose a consistency-based (CB) loss function that encourages the output predictions to be consistent with spatial transformations of the input images. The experiments on 3 open-source COVID-19 datasets show that this loss function yields significant improvement over conventional pointlevel loss functions and almost matches the performance of models trained with full supervision with much less human effort.

A multi‐class COVID‐19 segmentation network with pyramid attention and edge loss in CT images

“…Laradji et al. [46] proposed a weakly supervised learning algorithm, which performed point‐level annotations on CT images and used a consistency‐based loss function for training. The dice coefficient obtained is 0.73.…”

Learning Action Completeness from Points for Weakly-supervised Temporal Action Localization

“…Bearman et al [2] introduce the first weakly-supervised semantic segmentation framework that takes as supervision a single annotated pixel for each object. Since that work, a great amount of efforts [15,18,19,49,74] have been endeavored to utilize point-level supervision to solve various segmentation tasks in images or videos, thanks to its affordable annotation cost. Meanwhile, there are also attempts to employ point-level supervision to train object detectors [38,46,47].…”

Counting Cows: Tracking Illegal Cattle Ranching From High-Resolution Satellite Imagery

“…This might also include localization where the goal is to identify the locations of the objects in the image. This task has important real-life applications such as ecological surveys [1,24,16] and cell counting [7,12,11]. The datasets used for counting tasks [9,1] are often labeled with point-level annotations where a single pixel is labeled for each object.…”