Predicting Chronic Wound Healing Time Using Machine Learning

Dynamic prediction of time to wound healing at routine wound care visits

Objective Having a wound decreases patients' quality of life and brings uncertainty, especially if the wound does not show a healing tendency. The objective of this study was to develop and validate a model to dynamically predict time to wound healing at subsequent routine wound care visits. Approach A dynamic prediction model was developed in a cohort of wounds treated by nurse practitioners between 2017-2022. Potential predictors were selected based on literature, expert opinion, and availability in the routine care setting. To assess performance for future wound care visits, the model was validated in a new cohort of wounds visited in early 2023. Reporting followed TRIPOD guidelines. Results We analyzed data from 92,098 visits, corresponding to 14,248 wounds and 7,221 patients. At external validation, discriminative performance of our developed model was comparable to internal validation (c-statistic = 0.70 [95% CI 0.69, 0.71]) and the model remained well-calibrated. Strong predictors were wound-level characteristics and indicators of the healing process so far (e.g., wound surface area). Innovation Going beyond previous prediction studies in the field, the developed model dynamically predicts the remaining time to wound healing for many wound types at subsequent wound care visits, in line with the dynamic nature of wound care. In addition, the model was externally validated and showed stable performance. Conclusion: The developed model can potentially contribute to patient satisfaction and reduce uncertainty around wound healing times when implemented in practice. When the predicted time of wound healing remains high, practitioners can consider adapting their wound management.

Improved Wound Healing by Direct Cold Atmospheric Plasma Once or Twice a Week: A Randomized Controlled Trial on Chronic Venous Leg Ulcers

Objective: This study compared the effect of two frequencies of direct cold atmospheric plasma (direct-CAP) treatment with standard of care (SOC) alone on healing of venous leg ulcers (VLUs). Approach: Open-label, randomized controlled trial (ClinicalTrials.gov NCT04922463) on chronic VLUs at two home care organizations in the Netherlands. All three groups received SOC for 12 weeks or until healing. In addition, treatment groups received direct-CAP once (1× direct-CAP) or twice (2× direct-CAP) a week, at specialized wound care facilities and the patients' residences. Primary outcome was percentage of wounds healed. Secondary outcomes included wound area reduction and adverse events. Results: In total, 46 patients were randomly allocated to receive SOC only (n = 15), SOC + direct-CAP once a week (n = 17), or SOC + direct-CAP twice a week (n = 14). A higher percentage of wounds healed within 12 weeks in the treatment groups 53.3% (1× direct-CAP, p = 0.16) and 61.5% (2× direct-CAP, p = 0.08) versus 25.0% (control). The largest wound area reduction was obtained with 2× direct-CAP (95.2%, p = 0.07), followed by 1× direct-CAP (63.9%, p = 0.58), versus control (52.8%). Absolute wound area reduced significantly compared with baseline in both treatment groups (p ≤ 0.001), not in control (p = 0.11). No device-related serious adverse events occurred. Innovation: Direct-CAP applied once or twice a week could substantially improve wound healing of VLUs in primary care. Conclusion: Together with other clinical safety and efficacy data, these results support the integration of direct-CAP as a valuable therapy for complex wounds.

Management of post-Mohs surgical wounds with a hypothermically stored amniotic membrane: a case series

OBJECTIVE

The aim of this case series is to present an alternative approach to managing post-Mohs Micrographic Surgery (Mohs) wounds with hypothermically stored amniotic membrane (HSAM).

METHOD

A case series of patients with post-Mohs wounds is presented, with four patients referred for hard-to-heal wounds following a Mohs procedure that was performed 1-3 months previously. All wounds underwent weekly assessment, debridement, and application of HSAM and secondary dressings. Treatment also included management of bioburden, proper skin care and compression therapy for lower extremity wounds.

RESULTS

This case series of seven wounds consisted of four females and three males with a mean age of 87.6 years. Mean wound size at first application of HSAM was 1.34±1.20cm2. All wounds closed, with an average time to wound closure of 43.7±27.1 days. Patients received an average of 4.6±2.5 HSAM applications. The four post-Mohs wounds with a history of being hard-to-heal had an average time to wound closure of 35.5±16.3 days, with an average duration of 86.5±32.4 days prior to the first HSAM application.

CONCLUSION

The results of this case series suggest that use of HSAM may provide an alternative approach to managing post-Mohs wounds. In addition, these findings suggest that HSAM may be of greatest benefit when applied early after Mohs surgery.

AI energized hydrogel design, optimization and application in biomedicine

Traditional hydrogel design and optimization methods usually rely on repeated experiments, which is time-consuming and expensive, resulting in a slow-moving of advanced hydrogel development. With the rapid development of artificial intelligence (AI) technology and increasing material data, AI-energized design and optimization of hydrogels for biomedical applications has emerged as a revolutionary breakthrough in materials science. This review begins by outlining the history of AI and the potential advantages of using AI in the design and optimization of hydrogels, such as prediction and optimization of properties, multi-attribute optimization, high-throughput screening, automated material discovery, optimizing experimental design, and etc. Then, we focus on the various applications of hydrogels supported by AI technology in biomedicine, including drug delivery, bio-inks for advanced manufacturing, tissue repair, and biosensors, so as to provide a clear and comprehensive understanding of researchers in this field. Finally, we discuss the future directions and prospects, and provide a new perspective for the research and development of novel hydrogel materials for biomedical applications.

Human mesenchymal stem cell secretomes: Factors affecting profiling and challenges in clinical application

The promising field of regenerative medicine is thrilling as it can repair and restore organs for various debilitating diseases. Mesenchymal stem cells are one of the main components in regenerative medicine that work through the release of secretomes. By adopting the use of the secretome in cell-free-based therapy, we may be able to address the challenges faced in cell-based therapy. As one of the components of cell-free-based therapy, secretome has the advantage of a better safety and efficacy profile than mesenchymal stem cells. However, secretome has its challenges that need to be addressed, such as its bioprocessing methods that may impact the secretome content and its mechanisms of action in clinical settings. Effective and standardization of bioprocessing protocols are important to ensure the supply and sustainability of secretomes for clinical applications. This may eventually impact its commercialization and marketability. In this review, the bioprocessing methods and their impacts on the secretome profile and treatment are discussed. This improves understanding of its fundamental aspects leading to potential clinical applications.

Lab-on-a-Fiber Wearable Multi-Sensor for Monitoring Wound Healing

Chronic wounds are regarded as a silent epidemic, affecting 1-2% of the population and representing 2-4% of healthcare expenses. The current methods used to assess the wound healing process are based on the visual evaluation of physical parameters. This work aims to design a wearable non-invasive device capable of evaluating three parameters simultaneously: the pH and the levels of glucose and matrix metalloproteinase (MMP) present in the wound exudate. The device is composed of three independent polymer optical fibers functionalized with fluorescent-based sensing chemistries specific to the targeted analytes. Each fiber is characterized in terms of detection sensitivity and selectivity confirming their suitability for monitoring the targeted parameters in ranges relevant to the wound environment. The selectivity and robustness of the multi-sensing device are confirmed with analyses using complex solutions with different pH levels (5, 6, and 7), different concentrations of glucose (1.25, 2.5, and 5 mm), and MMP (1.25, 2.5, and 5 µg mL-1 ). Given the simple set-up, the affordability of the materials used and the possibility of detecting additional parameters relevant to wound healing, such multi-sensing fiber-based devices could pave the way for novel non-invasive wearable tools enabling the assessment of wound healing from the molecular perspective.

Risk predictions of surgical wound complications based on a machine learning algorithm: A systematic review

Surgical wounds may arise due to harm inflicted upon soft tissue during surgical intervention, and many complications and injuries may accompany them. These complications can lead to prolonged hospitalization and poorer clinical outcomes. Also, Machine learning (ML) is a Section of artificial intelligence (AI) that has emerged in medical care and is increasingly used for diagnosis, complications, prognosis and recurrence prediction. This study aims to investigate surgical wound risk predictions and management using a ML algorithm by R programming language analysis. The systematic review, following PRISMA guidelines, spanned electronic databases using search terms like 'machine learning', 'surgical' and 'wound'. Inclusion criteria covered experimental studies from 1990 to the present on ML's application in surgical wound evaluation. Exclusion criteria included studies lacking full text, focusing on ML in all surgeries, neglecting wound assessment and duplications. Two authors rigorously assessed titles, abstracts and full texts, excluding reviews and guidelines. Ultimately, relevant articles were then analysed. The present study identified nine articles employing ML for surgical wound management. The analysis encompassed various surgical procedures, including Cardiothoracic, Caesarean total abdominal colectomy, Burn plastic surgery, facial plastic surgery, laparotomy, minimal invasive surgery, hernia repair and unspecified surgeries. ML was skillful in evaluating surgical site infections (SSI) in seven studies, while two extended its use to burn-grade diagnosis and wound classification. Support Vector Machine (SVM) and Convolutional Neural Network (CNN) were the most utilized algorithms. ANN achieved a 96% accuracy in facial plastic surgery wound management. CNN demonstrated commendable accuracies in various surgeries, and SVM exhibited high accuracy in multiple surgeries and burn plastic surgery. In sum, these findings underscore ML's potential for significant improvements in postoperative management and the development of enhanced care techniques, particularly in surgical wound management.

Integrating factors associated with complex wound healing into a mobile application: Findings from a cohort study

Complex, chronic or hard-to-heal wounds are a prevalent health problem worldwide, with significant physical, psychological and social consequences. This study aims to identify factors associated with the healing process of these wounds and develop a mobile application for wound care that incorporates these factors. A prospective multicentre cohort study was conducted in nine health units in Portugal, involving data collection through a mobile application by nurses from April to October 2022. The study followed 46 patients with 57 wounds for up to 5 weeks, conducting six evaluations. Healing time was the main outcome measure, analysed using the Mann-Whitney test and three Cox regression models to calculate risk ratios. The study sample comprised various wound types, with pressure ulcers being the most common (61.4%), followed by venous leg ulcers (17.5%) and diabetic foot ulcers (8.8%). Factors that were found to impair the wound healing process included chronic kidney disease (U = 13.50; p = 0.046), obesity (U = 18.0; p = 0.021), non-adherence to treatment (U = 1.0; p = 0.029) and interference of the wound with daily routines (U = 11.0; p = 0.028). Risk factors for delayed healing over time were identified as bone involvement (RR 3.91; p < 0.001), presence of odour (RR 3.36; p = 0.007), presence of neuropathy (RR 2.49; p = 0.002), use of anti-inflammatory drugs (RR 2.45; p = 0.011), stalled wound (RR 2.26; p = 0.022), greater width (RR 2.03; p = 0.002), greater depth (RR 1.72; p = 0.036) and a high score on the healing scale (RR 1.21; p = 0.001). Integrating the identified risk factors for delayed healing into the assessment of patients and incorporating them into a mobile application can enhance decision-making in wound care.

A Deep Learning Image-to-Image Translation Approach for a More Accessible Estimator of the Healing Time of Burns

OBJECTIVE

An accurate and timely diagnosis of burn severity is critical to ensure a positive outcome. Laser Doppler imaging (LDI) has become a very useful tool for this task. It measures the perfusion of the burn and estimates its potential healing time. LDIs generate a 6-color palette image, with each color representing a healing time. This technique has very high costs associated. In resource-limited areas, such as low- and middle-income countries or remote locations like space, where access to specialized burn care is inadequate, more affordable and portable tools are required. This study proposes a novel image-to-image translation approach to estimate burn healing times, using a digital image to approximate the LDI.

METHODS

This approach consists of a U-net architecture with a VGG-based encoder and applies the concept of ordinal classification. Paired digital and LDI images of burns were collected. The performance was evaluated with 10-fold cross-validation, mean absolute error (MAE), and color distribution differences between the ground truth and the estimated LDI.

RESULTS

Results showed a satisfactory performance in terms of low MAE ( 0.2370 ±0.0086). However, the unbalanced distribution of colors in the data affects this performance.

SIGNIFICANCE

This novel and unique approach serves as a basis for developing more accessible support tools in the burn care environment in resource-limited areas.

Clinical Efficacy of Hyaluronic Acid with Iodine in Hard-to-Heal Wounds

Hard-to-heal wounds do not heal spontaneously and need long-term care provided by specialists. That burdens the patients as well as the healthcare systems. Such wounds arise from several pathologies, which result in venous leg ulcers (VLU), diabetic foot ulcers (DFU), pressure ulcers (PU), or ulcers originating from post-surgical wounds (pSW). Given the complex nature of hard-to-heal wounds, novel treatments are sought to enable wound healing. We tested the clinical efficacy and applicability of fluid comprising hyaluronic acid and iodine complex (HA-I) in the treatment of hard-to-heal wounds. Patients (n = 56) with VLU, DFU, PU, or pSW hospitalised in multiple wound-care centres in the Czech Republic were treated with HA-I. Wound size, classically visible signs of infection, exudation, pain, and wound bed appearance were monitored for 12 weeks. The highest healing rate was in DFU (71.4%), followed by pSW (62.5%), VLU (55.6%), and PU (44.4%). Classical visible signs of infection were resolved within 8 weeks in all types of wounds. Wound bed appearance improved most noticeably in pSW and then in VLU. Exudation was lowered most significantly in DFU and pSW. The highest decrease in pain was in pSW and DFU. The treatment with HA-I successfully led to either complete closure or significant improvement in the wound's healing. Therefore, the complex of hyaluronic acid and iodine is suitable for the treatment of hard-to-heal wounds of various aetiologies.

Individualized Risk Prediction for Improved Chronic Wound Management

Significance: Chronic wounds are associated with significant morbidity, marked loss of quality of life, and considerable economic burden. Evidence-based risk prediction to guide improved wound prevention and treatment is limited by the complexity in their etiology, clinical underreporting, and a lack of studies using large high-quality datasets. Recent Advancements: The objective of this review is to summarize key components and challenges in the development of personalized risk prediction tools for both prevention and management of chronic wounds, while highlighting several innovations in the development of better risk stratification. Critical Issues: Regression-based risk prediction approaches remain important for assessment of prognosis and risk stratification in chronic wound management. Advances in statistical computing have boosted the development of several promising machine learning (ML) and other semiautomated classification tools. These methods may be better placed to handle large number of wound healing risk factors from large datasets, potentially resulting in better risk prediction when combined with conventional methods and clinical experience and expertise. Future Directions: Where the number of predictors is large and heterogenous, the correlations between various risk factors complex, and very large data sets are available, ML may prove a powerful adjuvant for risk stratifying patients predisposed to chronic wounds. Conventional regression-based approaches remain important, particularly where the number of predictors is relatively small. Translating estimated risk derived from ML algorithms into practical prediction tools for use in clinical practice remains challenging.

Diagnosis of Chest Pneumonia with X-ray Images Based on Graph Reasoning

Pneumonia is an acute respiratory infection that affects the lungs. It is the single largest infectious disease that kills children worldwide. According to a 2019 World Health Organization survey, pneumonia caused 740,180 deaths in children under 5 years of age, accounting for 14% of all deaths in children under 5 years of age but 22% of all deaths in children aged 1 to 5 years. This shows that early recognition of pneumonia in children is particularly important. In this study, we propose a pneumonia binary classification model for chest X-ray image recognition based on a deep learning approach. We extract features using a traditional convolutional network framework to obtain features containing rich semantic information. The adjacency matrix is also constructed to represent the degree of relevance of each region in the image. In the final part of the model, we use graph inference to complete the global modeling to help classify pneumonia disease. A total of 6189 children's X-ray films containing 3319 normal cases and 2870 pneumonia cases were used in the experiment. In total, 20% was selected as the test data set, and 11 common models were compared using 4 evaluation metrics, of which the accuracy rate reached 89.1% and the F1-score reached 90%, achieving the optimum.

Advances in non-invasive biosensing measures to monitor wound healing progression

Impaired wound healing is a significant financial and medical burden. The synthesis and deposition of extracellular matrix (ECM) in a new wound is a dynamic process that is constantly changing and adapting to the biochemical and biomechanical signaling from the extracellular microenvironments of the wound. This drives either a regenerative or fibrotic and scar-forming healing outcome. Disruptions in ECM deposition, structure, and composition lead to impaired healing in diseased states, such as in diabetes. Valid measures of the principal determinants of successful ECM deposition and wound healing include lack of bacterial contamination, good tissue perfusion, and reduced mechanical injury and strain. These measures are used by wound-care providers to intervene upon the healing wound to steer healing toward a more functional phenotype with improved structural integrity and healing outcomes and to prevent adverse wound developments. In this review, we discuss bioengineering advances in 1) non-invasive detection of biologic and physiologic factors of the healing wound, 2) visualizing and modeling the ECM, and 3) computational tools that efficiently evaluate the complex data acquired from the wounds based on basic science, preclinical, translational and clinical studies, that would allow us to prognosticate healing outcomes and intervene effectively. We focus on bioelectronics and biologic interfaces of the sensors and actuators for real time biosensing and actuation of the tissues. We also discuss high-resolution, advanced imaging techniques, which go beyond traditional confocal and fluorescence microscopy to visualize microscopic details of the composition of the wound matrix, linearity of collagen, and live tracking of components within the wound microenvironment. Computational modeling of the wound matrix, including partial differential equation datasets as well as machine learning models that can serve as powerful tools for physicians to guide their decision-making process are discussed.