Trial Assesses Automated Wound Measurement Application vs Manual Measurements

Artificial intelligence already features in some areas of wound care, such as diagnosis based on patient-provided images, evaluating rates of wound healing, and predicting the rate—or lack thereof—of future healing. However, because machine learning algorithms are based on the data and reference images provided, concerns exist regarding bias or inaccuracy when it comes to imaging wounds for patients with darker skin. A research team conducted a study to assess the accuracy of an automated wound measurement application (AWMA) in providing wound measurements for patients with dark skin pigmentation compared to patients with lighter skin pigmentation and presented their findings at SAWC Spring 2023 in National Harbor, Maryland. The team included Alisha Oropallo, MD; Amit Rao, MD; Christina Del Pin, MD; Marisa Ranire-Maguire, MD; and Kane Genser, MD.

The researchers enrolled a total of 33 patients with dark skin pigmentation who presented with chronic wounds. All wounds were measured both manually, using acetate tracing, and by supplying a photo for AWMA evaluation, and the surface area measurements for both procedures were compared.

According to the results, the AWMA measurements demonstrated a 90.1% (±14.5%) agreement with those yielded by manual acetate-based measurements. The poster noted that these findings were not significantly different from those of previous studies evaluating this specific application.¹ Based on their findings, the authors concluded that no obvious algorithm bias exists for the studied AWMA.

-Kelsey Kaustinen, Associate Editor

Poster Reference: Oropallo A, Rao A, Del Pin C, Ranire-Maguire M, Genser K. Clinical evaluation of an automated wound measurement mobile application for patients with dark skin pigmentation. Poster presented at Symposium on Advanced Wound Care Spring; April 26-30, 2023; National Harbor, MD.

References

1. Budman J, Keenahan K, Acharya S, Brat GA. Design of a smartphone application for automated wound measurements for home care. Iproceedings. 2015;1(1):e16. doi:10.2196/iproc.4703