An image dataset for surveillance of personal protective equipment adherence in healthcare

Abstract

Proper personal protective equipment (PPE) use is critical to prevent disease transmission to healthcare providers, especially those treating patients with a high infection risk. To address the challenge of monitoring PPE usage in healthcare, computer vision has been evaluated for tracking adherence. Existing datasets for this purpose, however, lack a diversity of PPE and nonadherence classes, represent single not multiple providers, and do not depict dynamic provider movement during patient care. We introduce the Resuscitation Room Personal Protective Equipment (R2PPE) dataset that bridges this gap by providing a realistic portrayal of diverse PPE use by multiple interacting individuals in a healthcare setting. This dataset contains 26 videos, 10,034 images and 123,751 bounding box annotations for 17 classes of PPE adherence and nonadherence for eyewear, masks, gowns, and gloves, and one additional head class. Evaluations using newly proposed metrics confirm R2PPE exhibits higher annotation density than three established general-purpose and medical PPE datasets. The R2PPE dataset provides a resource for developing computer vision algorithms for monitoring PPE use in healthcare.

Fig. 1

Sample images from R2PPE obtained from different medical setups. Bounding boxes are overlaid on items of interest to indicate the PPE adherence and nonadherence classes in different colors. (a) Scenario 1: a mannequin placed on a patient bed. (b) Scenario 2: a bed with no mannequin. (c) Scenario 3: individuals walking around.

Fig. 2

Distribution of PPE adherence and nonadherence classes represented in the dataset images.

Fig. 3

Percentage of images containing each class.

Fig. 4

Sample images: (a) and (b) from R2PPE-L, and (c) and (d) from R2PPE-H. Images in R2PPE-L contain fewer or less overlapped bounding boxes, while images in R2PPE-H have more densely packed and overlapped bounding boxes.

Fig. 5

Bounding box aspect and area ratio distributions in the dataset images. (a) Bounding box aspect ratio distribution. (b) Bounding box area ratio distribution.

Fig. 6

Comparison of R2PPE and three other relevant datasets using local and global density. The x-axis represents density, and the y-axis represents the proportion of data points proximate to the corresponding x-value relative to the total data count. (a) Local density D_l. (b) Global density D_g.