IEEE Access (Jan 2024)
The Impact of Artificial Intelligence and Deep Learning-Based Family-Centered Care Interventions on the Healing of Chronic Lower Limb Wounds in Children
Abstract
Currently, there are numerous challenges in the nursing and healing of children’s chronic lower limb wounds (CLLW), such as prolonged healing times, difficulties in pain management, high risk of infection, and insufficient parental caregiving knowledge. This study aims to investigate the impact of a family-centered care (FCC) system based on artificial intelligence (AI) and convolutional neural network (CNN) algorithms on the healing of CLLW. A FCC for Children’s CLLWs (CLLW FCC) system was designed, consisting primarily of four components: an interaction layer, a service layer, a data layer, and a hardware layer. The CNN algorithm was utilized to establish a segmentation method for chronic wound images, which was then applied to the wound management section of the CLLW FCC system, aiming to improve the nursing outcomes of children with CLLW, optimize wound management processes, and enhance the accuracy and efficiency of image processing and analysis. It focused on 92 pediatric patients with CLLW admitted to our hospital from January 2022 to June 2023. The patients were randomly assigned into a control (Con) group and an observation (Obs) group, with 46 cases in each. Patients in the Con group received routine care, while those in the Obs group received FCC system for CLLW. The wound infection rate (WIR), wound healing time (WHT), wound pain scores, parental knowledge scores, and satisfaction with care (SWC) were evaluated for both the Con and Obs groups. The results unveiled that the optimized CNN algorithm achieved mean intersection over union (mIOU) and Kappa values of 0.8965 and 0.8773, respectively, which were higher than those of other algorithms. Patients in the Obs group experienced a shorter WHT and lower wound pain scores in contrast to those in the Con group ( $P\lt 0.05$ ). The parental knowledge scores were higher in the Obs group, showing great differences with those in the Con group ( $P\lt 0.05$ ). The WIR in the Obs group was 2.17%, lower to 10.87% in the Con group ( $P\lt 0.05$ ). Additionally, the overall SWC was 95.65% in the Obs group, being higher to 80.32% in the Con group. This indicated that the FCC system based on AI and CNN algorithms in this work could sharply shorten the pediatric WHT, reduce pain scores and WIR during care, and improve the parental knowledge scores and SWC.
Keywords
