Advances in Skin & Wound Care最新文献

Objective: To evaluate the clinical utility of a point-of-care, artificial intelligence-enabled multispectral imaging device in guiding targeted debridement of chronic leg ulcers in patients with leprosy, using a within-person randomized controlled pilot design.

Methods: Five adult male patients with lepromatous leprosy and at least 2 chronic leg ulcers each were enrolled in a split-body design. One ulcer per patient was randomized to the experimental arm (EA), where weekly debridement was guided by multispectral imaging, and the other to the control arm (CA), which received standard care. The device used autofluorescence to identify areas of suspected bacterial colonization and provided Gram-type classification. Healing was assessed by changes in wound area and Pressure Ulcer Scale for Healing scores over 18 weeks. Microbial confirmation was performed using standardized swab cultures.

Results: At 18 weeks, the mean wound size reduction was greater in the EA (84.46%) than in the CA (73.28%). Pressure Ulcer Scale for Healing scores decreased more rapidly in the EA (from 11.4 to 4.75) compared with the CA (from 11.0 to 6.75). One ulcer in each arm achieved full epithelialization, but the EA ulcer healed faster (5 vs. 9 weeks). Autofluorescence imaging enabled targeted systemic antimicrobial use in several cases. No adverse events were reported.

Conclusions: This pilot, the first of its kind in leprosy ulcer care, demonstrates the potential of artificial intelligence-enabled multispectral imaging to enhance wound healing through guided debridement. The technology offers real-time, noninvasive infection assessment that may support more effective, individualized wound management. Larger, blinded studies are warranted to validate these findings.

Objective: This study explores neonatal intensive care unit (NICU) nurses' perceptions of artificial intelligence (AI)-assisted neonatal skin assessment, focusing on its benefits, challenges, and ethical implications. Optimizing AI integration requires understanding nurses' attitudes.

Methods: A qualitative phenomenological approach was employed. Semi-structured interviews were conducted with 23 NICU nurses from a public hospital in Gaziantep, Turkey, between January and March 2025. Data were analyzed using inductive content analysis to identify emerging themes related to AI's impact on clinical decision-making, workflow efficiency, and professional autonomy.

Results: Findings revealed that nurses acknowledged AI's potential to enhance diagnostic accuracy, standardize assessments, and reduce interobserver variability. However, concerns were raised regarding algorithm reliability, professional autonomy, and ethical considerations. Nurses recognized AI's potential but stressed the need for transparency, training, and safeguards against over-reliance. Participants emphasized human oversight to ensure patient-centered care.

Conclusions: Artificial intelligence may improve neonatal skin assessment, but integration must balance technology and ethics. Engaging NICU nurses in AI system development and implementation is essential to fostering trust and ensuring alignment with clinical needs. Future research should assess AI's long-term impact and support interdisciplinary tool development that complements nursing expertise.

Objective: Pressure injuries (PIs) have become a global issue due to the significant social costs associated with various factors. Although many factors have been shown to have an impact on PIs, what specifically contributes to the worsening of the disease remains unclear. The aim of this study was to analyze variables that are highly correlated with PI aggravation using machine learning.

Methods: This observational study examined 71 Stage 2 PI patients from May 2018 to June 2021. The authors classified patients into 2 groups according to wound progression: (1) group A, aggravated group, and (2) group B, healed group. All 24 factors were analyzed using a Random Forest with hyperensemble approach, one of the machine learning algorithms. Each Random Forest is composed of 50,000 decision trees, and results from 100 Random Forests were hyperensembled. The mean decrease accuracy was calculated to evaluate the importance of the factor, and overlapped partial dependence plots were obtained to interpret the risk factors.

Results: Group A had 14 patients, whereas group B had 57. In an analysis using machine learning, the following factors were found to be highly associated with the aggravation of PIs: serum-albumin, Braden Scale, hemoglobin, wound size, serum-blood urea nitrogen, body mass index, serum-protein, and serum-creatinine. But the following variables were less associated: end-stage renal disease, sex, and myocardial infarction.

Conclusions: The PIs prediction model has broad application as a PI prevention tool. In addition, these findings can aid in the development of strategies to minimize the risk of PI aggravation.

Background: Diabetic foot ulcer (DFU) care is a challenge in reconstructive surgery. Artificial intelligence (AI) tools represent a new resource for patients with DFUs to seek information.

Objective: To evaluate the efficacy of ChatGPT-3.5 in responding to frequently asked questions related to DFU care.

Methods: Researchers posed 11 DFU care questions to ChatGPT-3.5 in December 2023. Questions were divided into topic categories of wound care, concerning symptoms, and surgical management. Four plastic surgeons in the authors' wound care department evaluated responses on a 10-point Likert-type scale for accuracy, comprehensiveness, and danger, in addition to providing qualitative feedback. Readability was assessed using 10 readability indexes.

Results: ChatGPT-3.5 answered questions with a mean accuracy of 8.7±0.3, comprehensiveness of 8.0±0.7, and danger of 2.2±0.6. ChatGPT-3.5 answered at the mean grade level of 11.9±1.8. Physician reviewers complimented the simplicity of the responses (n=11/11) and the AI's ability to provide general information (n=4/11). Three responses presented incorrect information, and the majority of responses (n=10/11) left out key information, such as deep vein thrombosis symptoms and comorbid conditions impacting limb salvage.

Conclusions: The researchers observed that ChatGPT-3.5 provided misinformation, omitted crucial details, and responded at nearly 4 grade levels higher than the American average. However, ChatGPT-3.5 was sufficient in its ability to provide general information, which may enable patients with DFUs to make more informed decisions and better engage in their care. Physicians must proactively address the potential benefits and limitations of AI.

Objective: Prevalence of hospital-acquired pressure injury (PI), as a critical measurement of medical care quality, has shown an upward trend. The aim of this study was to determine the predictive indicators of potential PIs and ensure that the predictive indicators can automatically be mined from electronic medical record systems.

Methods: The methods include 2 parts. One is the modified Delphi for indicator development, including clinical health care provider interviews, literature review, research group meetings, and Delphi survey. The other is feature selection, including extracting indicators from the health care information system (HIS) by structured query language and selecting indicators using the Random Forest technique.

Results: A predictive indicator system (with feature extraction rules for each indicator) consisting of 3 categories and 14 indicators was constructed. The experts' consensus was reached on all indicators (mean=4.28±0.65 to 4.94±0.23; coefficient of variation=4.63% to 17.20%; agreement rate=83.30% to 100.00%). The agreement between manual extraction and the computer's automatic extraction was good, with a Cohen κ score of 0.64 to 1.00. The accuracy of the good parsimonious prediction model was 95.26%.

Conclusions: This predictive indicator system is prepared for automatic PI prediction in the HIS. Many revisions should be conducted in further studies and practices in a real-life medical environment.

Abstract: Artificial intelligence (AI) is revolutionizing the landscape of skin and wound care by improving diagnostic accuracy, treatment effectiveness, and patient outcomes. Artificial intelligence-driven tools, including machine learning models and large language models (LLMs), enhance the precision of wound assessments, facilitate early infection detection, and streamline clinical workflows. In addition, these tools may aid in patient symptom reporting, bridging the communication gap between patients and health care providers. Current AI applications include image recognition for wound classification, patient-facing symptom-checking chatbots, and personalized treatment recommendations. The integration of AI technologies not only supports better clinical decision-making but also empowers patients through improved access, engagement, and education. These tools are currently aimed at supporting clinical decision-making, not replacing clinicians. Moving forward, the expansion of AI capabilities in skin and wound care holds great promise, driving cost-effective, scalable, and equitable health care solutions.

General purpose: To describe how wound care clinicians can improve treatment outcomes by promoting resilience in people with chronic wounds.JOURNAL/aswca/04.03/00129334-202510000-00004/figure1/v/2025-09-15T111045Z/r/image-jpeg TARGET AUDIENCE: This continuing education activity is intended for physicians, physician assistants, nurse practitioners, and registered nurses with an interest in skin and wound care.

Learning objectives/outcomes: After participating in this educational activity, the participant will:Explain the impact of psychological stressors on wound healing.Identify interventions to assess and promote resilience in individuals with chronic wounds.Describe the impact of chronic wounds on psychological health.The prevalence of chronic wounds and their impact on physical, psychological, and social well-being continues to escalate. Optimal outcomes rely on the ability of clinicians to work collaboratively with patients and care partners to plan and implement holistic, individualized treatments. Physical health is heavily impacted by wound status and the patients' capacity to engage in usual daily activities. Functional limitations carry a host of consequences, including adverse effects on patients' sense of meaning and purpose in life, self-esteem, and body image. Negative emotions, fear, and loss of autonomy can create challenges to mental well-being, with higher rates of depression and anxiety reported in people with chronic wounds as compared with the general population. These challenges are exacerbated by stigma, patients' reluctance to disclose psychological symptoms, and clinicians' lack of preparation in assessing and addressing mental, as well as physical health. Psychological stress carries physiological consequences that can contribute to healing delays. These consequences can be offset by cognitive behavioral interventions, a strong therapeutic alliance, and peer support. Integrative, individualized plans of care are improved by shared decision-making and the application of social and behavioral theory to provide insight regarding patients' abilities and willingness to actively engage in collaborative wound management. Resilience, or the ability to productively cope with adversity, mediates the psychological burden associated with chronic wounds. The purpose of this targeted, narrative review of the literature is to assist clinicians in assessing the physiological consequences of delayed healing and promoting resilience in people with chronic wounds.

Background: Smartphone photography may play a role in various aspects of clinical practice in wound surgery. Its accuracy as a tool for wound assessment and clinical decision-making is yet to be proven. Moreover, data regarding the magnitude of its use in practice are lacking.

Methods: Eleven board-certified plastic surgeons performed 79 wound observations and completed a questionnaire regarding wound properties and decisions regarding management. Wounds were photographed using smartphones at the time of initial wound observation. At least 3 months later, photographs of the wounds were anonymously presented to the same surgeons who completed the questionnaires again. Statistical analysis was used to compare the results. In addition, an online survey was used to assess the magnitude and manner of smartphone photography use among plastic surgeons.

Results: Comparison of bedside and photographic wound evaluation found no statistically significant differences in nearly all descriptive wound parameters and aspects of clinical decision-making. Statistically significant differences were found for periwound subcutaneous space assessment (P=.02) and recommendation for operative wound closure (P=.035). Seventy-four plastic surgeons replied to the online survey, and 93% of them stated that they use smartphone photography in their daily practice, with the majority using it equally for patient follow-up, consulting other physicians, and communication with patients.

Conclusions: Smartphone photography seems to play a major role in present-day clinical practice. According to the findings of this study, assessment of wounds via smartphone photography can be safely used as an adjunct for clinical decision-making when used as a consultation aid between wound surgeons.

Objective: This study explores nurses' acceptability and readiness to integrate patient-centered artificial intelligence (AI) technologies for pressure injury (PI) prevention, aiming to inform the design of clinically applicable technologies.

Methods: This qualitative descriptive study gathered insights from 202 international nurses in 2 countries through focus group discussions and written responses. Thematic analysis was conducted using MAXQDA.

Results: Three main concepts were identified. Under the use of manual tools in risk assessment, the theme was clinical challenges of the Braden Scale, with subthemes of accuracy and reliability, limitations in specific patient populations, and patient nonmodifiable related risk stratification. Within integration of AI-based technologies, themes included expectations from AI-based systems, with subthemes of advanced risk stratification prediction and real-time data, and concerns about AI integration in the system, with subthemes of acceptability level, education and awareness, data accuracy and reliability, and ethical issues and patient safety. For patient-centered monitoring systems, themes included development of automated documentation with subthemes of reducing workload, time management, integration of early warning systems with subthemes of automated monitoring, early intervention, and AI-supported decision support systems with subthemes of personalized interventions and proactive intervention.

Conclusions: Current nurse-led risk assessment systems require improvement for specific patient groups, affecting safety and care quality. Artificial intelligence-based systems can provide more accurate risk predictions and personalized interventions, enhancing decision-making and clinical outcomes. Although nurses are ready for AI adoption, further education is needed for full integration to optimize patient care.