Sheila C. Wang

Background: Diabetes-related foot ulceration (DFU) is a common complication of diabetes, with a significant impact on survival, health care costs, and health-related quality of life. The prognosis of DFU varies widely among individuals. The International Working Group on the Diabetic Foot recently updated their guidelines on how to classify ulcers using “classical” classification and scoring systems. No system was recommended for individual prognostication, and the group considered that more detail in ulcer characterization was needed and that machine learning (ML)–based models may be the solution. Despite advances in the field, no assessment of available evidence was done. Objective: This study aimed to identify and collect available evidence assessing the ability of ML-based models to predict clinical outcomes in people with DFU. Methods: We searched the MEDLINE database (PubMed), Scopus, Web of Science, and IEEE Xplore for papers published up to July 2023. Studies were eligible if they were anterograde analytical studies that examined the prognostic abilities of ML models in predicting clinical outcomes in a population that included at least 80% of adults with DFU. The literature was screened independently by 2 investigators (MMS and DAR or EH in the first phase, and MMS and MAS in the second phase) for eligibility criteria and data extracted. The risk of bias was evaluated using the Quality In Prognosis Studies tool and the Prediction model Risk Of Bias Assessment Tool by 2 investigators (MMS and MAS) independently. A narrative synthesis was conducted. Results: We retrieved a total of 2412 references after removing duplicates, of which 167 were subjected to full-text screening. Two references were added from searching relevant studies’ lists of references. A total of 11 studies, comprising 13 papers, were included focusing on 3 outcomes: wound healing, lower extremity amputation, and mortality. Overall, 55 predictive models were created using mostly clinical characteristics, random forest as the developing method, and area under the receiver operating characteristic curve (AUROC) as a discrimination accuracy measure. AUROC varied from 0.56 to 0.94, with the majority of the models reporting an AUROC equal or superior to 0.8 but lacking 95% CIs. All studies were found to have a high risk of bias, mainly due to a lack of uniform variable definitions, outcome definitions and follow-up periods, insufficient sample sizes, and inadequate handling of missing data. Conclusions: We identified several ML-based models predicting clinical outcomes with good discriminatory ability in people with DFU. Due to the focus on development and internal validation of the models, the proposal of several models in each study without selecting the “best one,” and the use of nonexplainable techniques, the use of this type of model is clearly impaired. Future studies externally validating explainable models are needed so that ML models can become a reality in DFU care. Trial Registration: PROSPERO CRD42022308248; https://www.crd.york.ac.uk/PROSPERO/view/CRD42022308248

Background: Chronic wounds affect 1%-2% of the global population, and pose significant health and quality-of-life challenges for patients and caregivers. Advances in artificial intelligence (AI) and computer vision (CV) technologies present new opportunities for enhancing wound care, particularly through remote monitoring and patient engagement. A digital wound care solution (DWCS) that facilitates wound tracking using AI was redesigned as a patient-facing mobile app to empower patients and caregivers to actively participate in wound monitoring and management. Objective: This study aims to evaluate the feasibility, usability, and preliminary clinical outcomes of the Patient Connect app (Swift Medical Inc) in enabling patients and caregivers to remotely capture and share wound data with health care providers. Methods: A feasibility study was conducted at 2 outpatient clinics in Canada between May 2020 and February 2021. A total of 28 patients with chronic wounds were recruited and trained to use the Patient Connect app for wound imaging and secure data sharing with their care teams. Wound images and data were analyzed using AI models integrated into the app. Clinicians reviewed the data to inform treatment decisions during follow-up visits or remotely. Key metrics included app usage frequency, patient engagement, and wound closure rates. Results: Participants captured a median of 13 wound images per wound, with images submitted every 8 days on average. The study cohort included patients with diabetic ulcers, venous ulcers, pressure injuries, and postsurgical wounds. A median wound closure surface area closure of 80% (range 15-100) was achieved across all patients, demonstrating the app’s clinical potential. Feedback from patients and clinicians highlighted during the feasibility testing support insight into the app’s usability, data security features, and ability to enhance remote monitoring that need to be explored in further qualitative research. Conclusions: The Patient Connect app effectively engaged patients and caregivers in chronic wound care, demonstrating feasibility and promising clinical outcomes. By enabling secure, remote wound monitoring through AI technology, the app has the potential to improve patient adherence, enhance care accessibility, and optimize clinical workflows. Future studies should focus on evaluating its scalability, cost-effectiveness, and broader applicability in diverse health care settings.

BACKGROUND Chronic wounds affect 1-2% of the global population, and pose significant health and quality-of-life challenges for patients and caregivers. Advances in artificial intelligence (AI) and computer vision (CV) technologies present new opportunities for enhancing wound care, particularly through remote monitoring and patient engagement. A Digital Wound Care Solution (DWCS) that objective wound tracking using AI/CV was redesigned as a patient-facing mobile application to empower patients and caregivers to actively participate in wound monitoring and management. OBJECTIVE This study aimed to evaluate the feasibility, usability, and preliminary clinical outcomes of the Patient Connect application in enabling patients and caregivers to remotely capture and share wound data with healthcare providers. METHODS A feasibility study was conducted at two outpatient clinics in Canada between May 2020 and February 2021. Twenty-eight patients with chronic wounds were recruited and trained to use the Patient Connect app for wound imaging and secure data sharing with their care teams. Wound images and data were analyzed using AI/CV models integrated into the app. Clinicians reviewed the data to inform treatment decisions during follow-up visits or remotely. Key metrics included app usage frequency, patient engagement, and wound closure rates. RESULTS Participants captured a median of 13 wound images per wound, with images submitted every 8 days on average. The study cohort included patients with diabetic ulcers, venous ulcers, pressure injuries, and post-surgical wounds. A median wound closure rate of 80% was achieved across all patients, demonstrating the app’s clinical potential. Feedback from patients and clinicians highlighted the app’s usability, data security features, and ability to enhance remote monitoring. CONCLUSIONS The Patient Connect application effectively engaged patients and caregivers in chronic wound care, demonstrating feasibility and promising clinical outcomes. By enabling secure, remote wound monitoring through AI/CV technology, the app has the potential to improve patient adherence, enhance care accessibility, and optimize clinical workflows. Future studies should focus on evaluating its scalability, cost-effectiveness, and broader applicability in diverse healthcare settings.

BACKGROUND Diabetes-related foot ulceration (DFU) is a common complication of diabetes, with a significant impact on survival, health care costs, and health-related quality of life. The prognosis of DFU varies widely among individuals. The International Working Group on the Diabetic Foot recently updated their guidelines on how to classify ulcers using “classical” classification and scoring systems. No system was recommended for individual prognostication, and the group considered that more detail in ulcer characterization was needed and that machine learning (ML)–based models may be the solution. Despite advances in the field, no assessment of available evidence was done. OBJECTIVE This study aimed to identify and collect available evidence assessing the ability of ML-based models to predict clinical outcomes in people with DFU. METHODS We searched the MEDLINE database (PubMed), Scopus, Web of Science, and IEEE Xplore for papers published up to July 2023. Studies were eligible if they were anterograde analytical studies that examined the prognostic abilities of ML models in predicting clinical outcomes in a population that included at least 80% of adults with DFU. The literature was screened independently by 2 investigators (MMS and DAR or EH in the first phase, and MMS and MAS in the second phase) for eligibility criteria and data extracted. The risk of bias was evaluated using the Quality In Prognosis Studies tool and the Prediction model Risk Of Bias Assessment Tool by 2 investigators (MMS and MAS) independently. A narrative synthesis was conducted. RESULTS We retrieved a total of 2412 references after removing duplicates, of which 167 were subjected to full-text screening. Two references were added from searching relevant studies’ lists of references. A total of 11 studies, comprising 13 papers, were included focusing on 3 outcomes: wound healing, lower extremity amputation, and mortality. Overall, 55 predictive models were created using mostly clinical characteristics, random forest as the developing method, and area under the receiver operating characteristic curve (AUROC) as a discrimination accuracy measure. AUROC varied from 0.56 to 0.94, with the majority of the models reporting an AUROC equal or superior to 0.8 but lacking 95% CIs. All studies were found to have a high risk of bias, mainly due to a lack of uniform variable definitions, outcome definitions and follow-up periods, insufficient sample sizes, and inadequate handling of missing data. CONCLUSIONS We identified several ML-based models predicting clinical outcomes with good discriminatory ability in people with DFU. Due to the focus on development and internal validation of the models, the proposal of several models in each study without selecting the “best one,” and the use of nonexplainable techniques, the use of this type of model is clearly impaired. Future studies externally validating explainable models are needed so that ML models can become a reality in DFU care. CLINICALTRIAL PROSPERO CRD42022308248; https://www.crd.york.ac.uk/PROSPERO/view/CRD42022308248

A 57-Year-Old Man with Type 1 Diabetes Mellitus and a Chronic Foot Ulcer Successfully Managed with a Remote Patient-Facing Wound Care Smartphone Application - Article abstract #933879

Introduction. Pressure injuries (PIs) are a serious, avoidable condition that affect many patients during hospital stays. Yet, to date, there is no comprehensive assessment of the financial and clinical risks of PIs. Objective. This study evaluates the cost of treatment, impact of reimbursement policies, and clinical consequences of PIs for US hospitals. Methods. A financial and clinical calculator was created to estimate the impact of PI prevention using a traditional literature review to drive assumptions. Results. Two drivers of hospital revenue loss resulting from PIs were identified: nonpayment for PI treatment by health insurance providers and personal injury litigation. Increased hospital length of stay (LOS) and patient mortality associated with PIs further contributed to negative consequences. For an average 160-bed hospital, the authors estimated an annual total financial risk of $5.97 million, 911 days added to LOS, and 16.4 deaths related to avoidable PIs. Conclusions. Results of this analysis will be useful for health care organizations implementing quality improvement initiatives and new technologies, such as digital wound care management systems, to reduce the prevalence of PIs, thereby protecting patients and mitigating financial and clinical risks.

Pressure ulcers (PUs) are a serious health care problem for nursing home residents and a key quality metric for regulators. Three initiatives were introduced at a 128‐bed facility to improve PU prevention. First, a Quality Assurance and Performance Improvement project and a Root Cause Analysis were conducted to improve the facility's wound care programme. Second, a digital wound care management solution was adopted to track wound management. Third, the role of skin integrity coordinator was created as a central point of accountability for wound care‐related activities and related performance metrics. Improvements in PU prevention were tracked using Centers of Medicare and Medicaid data, specifically (a) the percentage of long‐stay high‐risk residents with PUs and (b) the percentage of short‐stay residents with PUs that are new or have worsened. PU prevalence for long‐stay high‐risk residents was 12.99% (Q4 2016), and upon implementation of these initiatives, the facility saw continued reductions in PU prevalence to 2.9% (Q4 2017), while PUs for short‐stay residents were maintained at zero throughout this period. This study highlights the power of effective management combined with real‐time data analytics, as enabled by digital wound care management, to make significant improvements in health care delivery.