Page 62 - Read Online

Page 365 Ganesan et al. Art Int Surg 2024;4:364-75 https://dx.doi.org/10.20517/ais.2024.68

INTRODUCTION
Wound healing is a rapidly growing multidisciplinary field drawing clinicians from diverse backgrounds
including nursing, medicine, podiatry, plastic surgery, and physical therapy. The prevalence of chronic
wounds has increased in association with underlying conditions, such as aging, obesity, and diabetes, which
contribute to the nonhealing nature of many wounds. From 2014 to 2019, the number of Medicare
beneficiaries with a wound increased from 8.2 million to 10.5 million, with the largest increase in wound
[1]
prevalence in those less than 65 years of age . Economically, total Medicare spending estimates for all
[2]
wounds spanned $28.1 to $96.8 billion in 2014 . However, these numbers do not begin to capture the
impaired quality of life, lost wages, and lost productivity that is too often experienced by patients and their
family members. Providers face a new challenge: how to best care for an increasing number of wounds in an
age of strained resources. The rapid development of artificial intelligence (AI) may be an innovative method
to help reduce the burden that patients and providers face in the area of wound care.

In this review, we define AI as the ability of computers, machines, and other technology to use algorithms to
simulate human intelligence and problem-solving. AI’s power lies in its ability to process and interpret large
[3,4]
amounts of data quickly and improve upon itself without the need for manual input . AI can read
electronic medical records (EMR), process images, and predict clinical outcomes, all of which can be
[5]
applied to wound healing .

The following terminology is commonly described in AI-assisted medicine: machine learning, neural
networks, natural language processing, deep learning, and computer vision . Machine learning focuses on
[6,7]
using computers, data, and algorithms to imitate human learning and adaptation. Neural networks are sets
of interconnected algorithms that handle multiple inputs and outputs, identifying various data relationships
[7]
and filtering data as needed . Natural language processing comprehends language, translates texts, and
recognizes speech. Deep learning extracts progressively higher-level features from data through multiple
layers of processing to provide a single, high-level output. Computer vision enables computers to interpret
visual input. Morris et al. have previously reviewed these categories in the context of the general field of
[7]
surgery . These AI categories can additionally be applied to various stages of wound care, improving
wound diagnosis, classification, and measurement.

AI also aids wound management by assessing wounds for infection, necrosis, or healing. Additionally, AI
has contributed to more personalized care and better prognosis and preventative strategies. However, AI
brings challenges, including data privacy and equity of care. With appropriate safeguards and a cautiously
optimistic approach to AI in wound care, we can leverage AI to make significant improvements to the field.
In this review, we summarize AI advancements during stages of wound care, including diagnosis,
monitoring, therapy, and prognosis and prevention. We also discuss the challenges and future directions of
AI in wound care.

METHODS
A literature search was performed using freely accessible online databases, including PubMed, Scopus,
Cumulated Index in Nursing and Allied Health Literature, and Web of Science, from publication to July 20,
2024. Keywords included “wound healing”, “hard-to-heal wounds”, “wound care”, “artificial intelligence”,
“machine learning”, “deep learning”, “neural network”, and “arterial, venous, diabetic, pressure, or chronic
wounds and ulcers”. Articles were included for their specific discussions on the use of AI in common
chronic human wound diagnosis, management, prognosis, and prevention.