Pelvic floor muscle contraction automatic evaluation algorithm for pelvic floor muscle training biofeedback using self-performed ultrasound.

INTRODUCTION: Non-invasive biofeedback of pelvic floor muscle training (PFMT) is required for continuous training in home care. Therefore, we considered self-performed ultrasound (US) in adult women with a handheld US device applied to the bladder. However, US images are difficult to read and require assistance when using US at home. In this study, we aimed to develop an algorithm for the automatic evaluation of pelvic floor muscle (PFM) contraction using self-performed bladder US videos to verify whether it is possible to automatically determine PFM contraction from US videos. METHODS: Women aged ≥ 20 years were recruited from the outpatient Urology and Gynecology departments of a general hospital or through snowball sampling. The researcher supported the participants in their self-performed bladder US and videos were obtained several times during PFMT. The US videos obtained were used to develop an automatic evaluation algorithm. Supervised machine learning was then performed using expert PFM contraction classifications as ground truth data. Time-series features were generated from the x- and y-coordinate values of the bladder area including the bladder base. The final model was evaluated for accuracy, area under the curve (AUC), recall, precision, and F1. The contribution of each feature variable to the classification ability of the model was estimated. RESULTS: The 1144 videos obtained from 56 participants were analyzed. We split the data into training and test sets with 7894 time series features. A light gradient boosting machine model (Light GBM) was selected, and the final model resulted in an accuracy of 0.73, AUC = 0.91, recall = 0.66, precision = 0.73, and F1 = 0.73. Movement of the y-coordinate of the bladder base was shown as the most important. CONCLUSION: This study showed that automated classification of PFM contraction from self-performed US videos is possible with high accuracy.

Relationship between gene expression associated with cellular senescence in cells from discarded wound dressings and wound healing: A retrospective cohort study.

AIM: Senescent cells, inducing a senescence-associated secretory phenotype (SASP), lead to chronic inflammation in hard-to-heal wound tissue. However, eliminating senescent cells may impede normal wound healing due to their important role in the wound healing mechanism. Accordingly, we focused on wound exudates in hard-to-heal wounds, which contain many inflammation biomarkers consistent with SASP. Therefore, we hypothesized that senescent cells might be present in the exudates and induce chronic inflammation. This study investigated the relationship between gene expression associated with cellular senescence in exudates from pressure injuries and wound healing status. METHODS: This retrospective cohort study involved patients treated by a pressure injury team. We collected viable cells from wound dressings and analyzed gene expression. Pearson's correlation coefficient was calculated between cellular senescence and SASP expression. The relationship between the gene expression of cellular senescence and the wound area reduction rate by the following week was examined using a mixed-effects model. RESULTS: CDKN1A-related to cellular senescence-was expressed in 96.3 % of 54 samples, and CDKN1A expression and SASPs positively correlated (PLAU: r = 0.68 and TNF: r = 0.34). Low CDKN1A expression was statistically associated with a large wound area reduction rate (ß = 0.83, p < 0.01). CONCLUSIONS: Gene expression of both cellular senescence and SASP factor in wound dressings suggests the presence of cellular senescence. Senescent cells in wound dressings could be associated with delayed wound healing in the following week.

A method for harvesting viable cells from wound dressings.

Wound fluid has been well studied for exploring protein biomarkers contained in it. However, cells in wound fluid have not received much attention due to the difficulty in their collection. Our study aimed to establish a method for collecting viable cells from discarded wound dressings. A protocol was designed to wash out nonadherent cells and detach adherent cells from silicone-faced foam wound dressings using trypsin-EDTA. The optimal concentration and incubation time of trypsin-EDTA for collecting equivalent proportions of different cell types to the original cell population were determined in vitro. Cell composition and gene expression changes in monocytes, lymphocytes, neutrophils, fibroblasts and keratinocytes were confirmed using immunocytochemistry and RNA-sequencing ex vivo. Full-thickness wounds were created on 9-week-old male C57BL/6J mice. Wound fluid was collected, and half of it was applied to the wound dressings. The original cell population in the wound fluid and the cell population collected from wound dressings were compared. In the in vitro study, 0.25% trypsin-EDTA and 2.5-min incubation time were considered optimal for collecting adherent cells from wound dressings. In the ex vivo study, among all cell types, only CD3+ lymphocytes showed a significantly higher cell proportion in the collected group. The relative gene expression of the five selected cells showed no significant changes (p-value >0.05, |log2 fold change| < 1.5, differential gene expression analysis). Viable nonadherent and adherent cells were collected from wound dressings without altering gene expression and could be used in future studies for cellular analysis of wound fluid.

Efficacy of wearable vibration dressings on full-thickness wound healing in a hyperglycemic rat model.

Local low-frequency vibration promotes blood flow and wound healing in hard-to-heal diabetic foot ulcers (DFUs). However, vibration treatment is challenging in patients with DFUs due to wound management difficulties and low adherence. Consequently, developing wearable self-care devices becomes imperative for effective wound healing. This study introduces a wearable vibration dressing and assesses its impact on wound healing in hyperglycemic rats. Low-frequency vibration at 52 Hz was applied to the wound for 40 min/day in awake rats. Relative wound areas on post-wounding days (PWDs) 4-7 were significantly smaller and the wound closure rate was significantly higher in the vibration group than in the control group (p < 0.05, respectively). The total haemoglobin at baseline and after vibration on post-wounding day 7 was significantly larger in the vibration group than in the control group (p < 0.05). On PWD 7, the thickness of the granulation tissue was significantly higher in the vibration group than in the control group (p < 0.05). Moreover, the number of blood vessels at the wound site and vascular endothelial growth factor A protein expression were significantly higher in the vibration group than in the control group (p < 0.05, respectively). The ratio of (CD68+ /iNOS+ )/(CD163+ ) macrophages in the vibration group was significantly lower than that in the control group (p < 0.05). These results indicate the potential of wearable vibration dressings as new self-care devices that can promote angiogenesis and blood flow, improve inflammation, and enhance wound healing in DFUs.

Introduction of augmented reality to the remote wound care nursing consultation system.

OBJECTIVE: We implemented augmented reality (AR) in remote consultations for enhanced pressure injury (PI) care in homecare nursing and improved the efficiency of on-site technical education for homecare nurses. The study aimed to depict expert techniques using AR technology to improve PI healing time in a male patient. METHOD: We developed and implemented a new system that combines a transparent hand with an image and gives an output as a video image in the existing remote consultation software. The system was used to support remote care of PIs by nurses. RESULTS: We succeeded in superimposing the expert nurse's nonverbal hand gestures onto the patient's wound in real time. The visiting nurse's understanding of the system had improved and there was an improvement in the patient's PI healing. CONCLUSION: These results suggest that remote consultation using AR is effective to observe precise wound care demonstrations of the steps of the PI treatment and provide effective treatment.

Relationship between healing status and microbial dissimilarity in wound and peri-wound skin in pressure injuries.

AIM: Wound infection is the most serious cause of delayed healing for patients with pressure injuries. The wound microbiota, which plays a crucial role in delayed healing, forms by bacterial dissemination from the peri-wound skin. To manage the bioburden, wound and peri-wound skin care has been implemented; however, how the microbiota at these sites contribute to delayed healing is unclear. Therefore, we investigated the relationship between healing status and microbial dissimilarity in wound and peri-wound skin. METHODS: A prospective cohort study was conducted at a long-term care hospital. The outcome was healing status assessed using the DESIGN-R® tool, a wound assessment tool to monitor the wound healing process. Bacterial DNA was extracted from the wound and peri-wound swabs, and microbiota composition was analyzed using 16S rRNA gene analysis. To evaluate microbial similarity, the weighted UniFrac dissimilarity index between wound and peri-wound microbiota was calculated. RESULTS: Twenty-two pressure injuries (7 deep and 15 superficial wounds) were included in the study. For deep wounds, the predominant bacteria in wound and peri-wound skin were the same in the healing wounds, whereas they were different in all cases of hard-to-heal wounds. Analysis based on the weighted UniFrac dissimilarity index, there was no significant difference for healing wounds (p = 0.639), while a significant difference was found for hard-to-heal wounds (p = 0.047). CONCLUSIONS: Delayed healing is possibly associated with formation of wound microbiota that is different in composition from that of the skin commensal microbiota. This study provides a new perspective for assessing wound bioburden.

An in vivo critically colonised wound model with dysbiotic wound microbiota.

In critically colonised wounds, many of the signs of infection are often absent, and delayed healing may be the only clinical sign. The prevention of critical colonisation is important, but its pathophysiology has not yet been elucidated. We have previously reported that dysbiotic microbiota dissimilar to the peri-wound skin microbiota may develop in critically colonised wounds. To investigate the role of dysbiotic microbiota, this study aimed to develop a critically colonised wound model by transplantation of dysbiotic microbiota. To transplant microbiota, a bacterial solution (dysbiosis group) or with Luria-Bertani medium (commensal group) was inoculated to full-thickness wounds of rats. The bacterial solution was prepared by anaerobically culturing bacteria from donor rats on an artificial dermis in Luria-Bertani medium for 72 hours. As a result, the degree of the change in the microbial similarity between pre- and post-transplantation of microbiota was significantly higher in the dysbiosis group (P < .001). No signs of infection were observed in any rat in either group. The wound area in the dysbiosis group was significantly larger (P < .001), and there was a significant infiltration of neutrophils (P < .001). All rats of the dysbiosis group represented the clinical features of critically colonised wounds. Furthermore, there were significantly fewer regulatory T cells in the wounds of the dysbiosis group. This is the first study to develop a novel animal model that represents the clinical features of critically colonised wounds and will be useful in investigating the pathogenesis of critical colonisation via regulatory T cells.

Expression levels of NPPB, ITGB6, CPNE4, EML5, and ITSN1 in fresh exudates swabbed from critically colonised and infected full-thickness wounds in rats.

Pressure injury management requires reliable identification of critical colonisation due to lack of infection signs. Our research group previously proposed the mRNAs natriuretic peptide B (Nppb), integrin subunit beta 6 (Itgb6), copine 4 (Cpne4), echinoderm microtubule-associated protein like 5, and intersectin 1 as candidate markers in pooled exudates of critically colonised wounds. However, it is unclear whether mRNAs or proteins of the candidate genes would be suitable as biomarkers in fresh exudate. Therefore, this study aimed to evaluate the validity of the mRNAs and proteins as fresh exudate markers for critical colonisation. Three wound models of normal healing, critical colonisation, and infection were created in rats. Fresh swab-collected exudates were collected, and mRNA and protein expression levels were measured. In the fresh wound exudates, the detection frequency of Itgb6 tended to decrease in the critically colonised and infected wounds (P = .067), and those of Cpne4 and Nppb tended to be lower in the infected wounds than in the normal healing and critically colonised wounds (P = .006 and .067, respectively). In contrast, there was no difference in protein expression in the exudates. This study suggests that Itgb6 mRNA in fresh exudates is a promising biomarker for critical colonisation.

Bacterial invasion into the epidermis of rats with sodium lauryl sulphate-irritated skin increases damage and induces incontinence-associated dermatitis.

Incontinence-associated dermatitis (IAD) is caused by prolonged exposure to urine/liquid stool. It is a common and often painful skin condition in older incontinent adults because of poor prevention. Patients with urinary infections are at risk of developing IAD, and to guide the development of novel prevention strategies, we aimed to develop an animal model of IAD by urine and bacteria. First, contralateral sites on the dorsal skin of Sprague-Dawley rats were compromised by sodium lauryl sulphate (SLS), simulating frequent cleansing with soap/water. Filter discs were then placed inside ring-shaped chambers on foam dressings, inoculated with or without Pseudomonas aeruginosa, covered with agarose gels immersed in cultured filtrated urine, and secured in place with an occlusive dressing for 3 days. Untreated and SLS-compromised sites served as controls. The IAD was developed at bacteria-inoculated sites, characterised by severe IAD-like redness that persisted for up to 3 days post-exposure and higher disruption of the skin barrier function compared with non-inoculated sites. Pathological changes included epidermal thickening, partial skin loss, inflammatory cell infiltration, accumulation of red blood cells, and invasion of bacteria into the epidermis. This novel, clinically relevant IAD rat model can serve for future prevention developments.

Incidence of and risk factors for self-load-related and medical device-related pressure injuries in critically ill patients: A prospective observational cohort study.

There are two types of pressure injuries: self-load-related pressure injuries (PIs) and medical device-related pressure injuries (MDRPIs), but the differences in risk factors between PIs and MDRPIs have not yet been clarified. If risk factors for PIs and MDRPIs differ, preventive interventions should take this into account. This is a prospective cohort study aimed to determine the cumulative incidence of PIs and MDRPIs in critically ill patients and to identify corresponding risk factors. The study included 1418 patients who were admitted to the critical care medical center of a single university hospital in Tokyo, Japan, between 1 December 2019 and 31 August 2020. The Kaplan-Meier method was used to calculate the cumulative incidence of PIs and MDRPIs. Furthermore, the Cox proportional hazards model was used to analyse the predictors in both the PI and MDRPI incidence and non-incidence groups. Predictors were based on data from the 0th disease day. The cumulative incidence of PIs and MDRPIs was 4.6/1000 person-days and 3.6/1000 person-days, respectively. Multivariate analysis using the Cox proportional hazards model showed that common risk factors for the occurrence of PIs and MDRPIs were possession of PI on admission, higher blood lactate, blood purification therapy, and positional restriction. The risk factors for PIs only were diabetes mellitus, septic shock, and a lower serum albumin level, while the risk factors for MDRPIs only were the D-dimer level and extracorporeal membranous oxygenation treatment. Therefore, prophylactic interventions need to consider these different risk factors.

Ultrasonographic features of diabetic foot osteomyelitis: a case series.

OBJECTIVE: Osteomyelitis is a limb-threatening complication of diabetic foot ulcers. Early identification of the disease is key to ensuring successful prognosis. In this study, we describe ultrasonographic features for the identification of osteomyelitis. METHOD: Patients were screened through clinical, ultrasonographic and probe-to-bone tests. RESULTS: Ultrasonographic features in three patients that could be used to identify diabetic foot osteomyelitis included periosteal reaction, periosteal elevation, cortical erosions and presence of sequestrum, all of which were confirmed by a plain X-ray. CONCLUSION: An ultrasonographic examination could be used for the early detection of osteomyelitis, which could help clinicians devise prompt treatment strategies.

Nationwide time-series surveys of pressure ulcer prevalence in Japan.

OBJECTIVE: The Japanese Society of Pressure Ulcers (JSPU) has two purposes: first, to improve knowledge and skills among health professionals related to preventing and managing pressure ulcers (PUs); and second, to represent those in the field managing PUs, including with government and health authorities. Since 2006, JSPU has conducted fact-finding surveys about every four years to identify PU prevalence in Japan (2006, 2010, 2013 and 2016). Based on the prevalence identified by these surveys, an attempt was made to validate the achievements of JSPU's activities. METHOD: Information from one-day surveys of hospitals, long-term care health facilities, long-term care welfare facilities, and home visit nursing care stations was analysed. We used generalised estimating equations to estimate the proportions of PUs and their 95% confidence intervals (CIs) for each survey. RESULTS: A total of 662,419 patients in 2631 facilities participated in the surveys. The estimated proportions for all facilities (95% CI) in chronological order, from the first to the fourth survey, were: 2.67% (2.52-2.83); 2.61% (2.43-2.80); 1.99% (1.83-2.17); and 1.79% (1.65-1.94), respectively. In all facility types, the proportion of PUs was lower in the fourth survey than the first survey. CONCLUSION: The proportion of PUs showed a decreasing trend and was low according to global standards, demonstrating the efficacy of JSPU's activities.

Identification of microRNAs responsive to shear loading in rat skin.

Pressure injuries (PIs) are localised skin injuries that result from pressure with or without shear force. Shear force is more destructive than pressure in clinical settings. Therefore, determining the critical external forces is important for selecting the appropriate care to prevent PIs. To quantitatively distinguish pressure and shear loading with high specificity, we focused on microRNAs (miRs). This study aimed to identify the miRs that are distinguishable between pressure with and without shear loading in rat skin. Microarray analysis identified six candidate miRs from the comparisons among the pressure, shear, and unloaded groups. We analysed the expression levels of the candidate miRs in the process of PI development using real-time reverse transcriptase polymerase chain reaction. In the pressure and shear groups, miR-92b expressions at 6 hours after loading were 2.3 ± 1.3 and 2.9 ± 1.0, respectively, which were significantly higher than those in the control group (P = .014 and .004, respectively). miR-877 expression at 6 hours after loading was significantly increased only in the shear group (2.8 ± 0.9) compared with the control group (P = .016). These results indicate that miR-92b and miR-877 are promising biomarkers to determine for which external force healthcare professionals should intervene.

Polymorphism analysis of candidate risk genes for pressure injuries in older Japanese patients: A cross-sectional study at a long-term care hospital.

Advances in patient care for pressure injuries (PIs) have reduced the prevalence of PIs in Japan, although not in recent years. Several single-nucleotide polymorphisms (SNPs) have been identified in genes potentially associated with PIs. However, individual variance among PI risks require targeted investigations that may lead to the identification of PI susceptibilities or preventive care options that directly influence PI development pathways. This cross-sectional study examined the association between PIs and SNPs in genes related to tissue tolerance in patients in a long-term care hospital in Japan. A total of 178 participants (130 control, 20 with superficial PI history, and 28 with deep PI history) were enrolled in this study of eight SNPs in hypoxia inducible factor 1 subunit alpha (HIF1A), vascular endothelial growth factor C (VEGFC), heat shock protein 90 alpha family class A member 1 (HSP90AA1), myostatin (MSTN), and vitamin D receptor (VDR). The primary outcome was a history of superficial and deep PIs in the last 6 months. SNPs were examined by real-time polymerase chain reaction, followed by multivariate logistic regression analyses of the associations between the SNPs and PI history. The results showed a significant association between VEGFC rs1485766 and the history of superficial PIs (odds ratio = 2.95; 95% confidence interval = 1.07-8.11; p = 0.04). Stratified analysis using the Braden Scale (≤14) indicated a significant association between HIF1A rs11549465 and deep PIs (p = 0.04). Our study demonstrated that VEGFC rs1485766 and HIF1A rs11549465 were associated with superficial and deep PI susceptibilities, respectively.

Promoting effect of acylated homoserine lactone on the healing of tissue damage in model rats with incontinence-associated dermatitis.

OBJECTIVE: One of the most common complications in patients with incontinence is incontinence-associated dermatitis (IAD). This study was conducted to determine the pathophysiology of the healing process of IAD and to develop an effective therapeutic approach according to its pathophysiology. METHOD: IAD was reproduced on a dorsal rat skin by applying agarose gel containing water and enzymes, and inoculating it with bacteria. Examination of the IAD healing process suggested that the promotion of keratinocyte migration and improvement of basement membrane enhance keratinocyte layer elongations, which contribute to IAD healing. A therapeutic approach using N-(3-oxotetradecanoyl)-L-homoserine lactone, which is one of the acylated homoserine lactones (AHLs) and can promote keratinocyte migration in vitro, was applied on the IAD area in rats. RESULTS: AHL treatment after IAD development resulted in an earlier tipping point for recovery than the vehicle treatment. Histological and immunohistological analyses revealed that the tissue surface was already covered by the epidermis, indicating the results of elongation of the keratinocyte layer from hair follicles. The characteristics of the alignment of basal keratinocytes, the existence of stratum corneum, and the membrane-like distribution of the components of basement membrane were similar to those of a normal epidermis. CONCLUSION: These results suggested that AHL application possibly contributed to earlier IAD healing before progressing to a severe state. Although elongation of the keratinocyte layer was observed in both the AHL and vehicle groups, the possibility that AHL application promotes IAD healing was suggested. The new concept of the enhancement of keratinocyte migration as a therapeutic approach for IAD would change the skin care strategy for IAD in the healthcare setting.

Concurrent validity of biofilm detection by wound blotting on hard-to-heal wounds.

OBJECTIVE: Wound biofilms delay healing of hard-to-heal wounds. Convenient biofilm identification tools for clinical settings are currently not available, hindering biofilm-based wound management. Wound blotting with biofilm staining is a potential tool for biofilm detection, owing to its convenience. Although predictive validity of wound blotting has been established, it is necessary to confirm its concurrent validity. Furthermore, current staining systems employing ruthenium red have some disadvantages for clinical use. This study aimed to evaluate the usability of alcian blue as a substitute for ruthenium red. METHOD: Both in vitro and in vivo clinical samples were used to investigate validity and usability. RESULTS: The in vitro study showed that proteins and extracellular DNA in biofilms did not affect staining ability of ruthenium red and alcian blue in the detection of biofilms. In the in vivo study, using a wound biofilm model with Pseudomonas aeruginosa, the staining sensitivity of ruthenium red was 88.9% and 100% for alcian blue, with correlation coefficients of signal intensities with native polyacrylamide gel electrophoresis (PAGE) of r=0.67 (p=0.035) and r=0.67 (p=0.036) for ruthenium red and alcian blue, respectively. Results from clinical samples were r=0.75 (p=0.001) for ruthenium red and r=0.77 (p<0.001) for alcian blue. The sensitivities of wound blotting staining by ruthenium red and alcian blue were very high and had a good correlation with native PAGE analysis. CONCLUSION: Because the alcian blue procedure is more convenient than the ruthenium red procedure, wound blotting with alcian blue staining would be a promising tool to guide clinicians in delivering biofilm-based wound management.

Assessing absorbent products' effectiveness for the prevention and management of incontinence-associated dermatitis caused by urinary, faecal or double adult incontinence: A systematic review.

OBJECTIVE: Adults who suffer from incontinence are at substantial risk of developing incontinence-associated dermatitis (IAD). In healthcare settings, several interventions have been implemented to prevent or manage IAD, and several absorbent products have been developed for incontinent patients; however, there is no systematic review that has reported on which absorbent products are effective for the prevention or management of incontinence-associated dermatitis. We conducted a systematic review to investigate the effectiveness of absorbent products in the prevention and management of IAD. METHODS: MEDLINE (1946-August 31, 2020), CINAHL (1982-August 31, 2020), and Cochrane Library (August 31, 2020) were searched for relevant articles. RESULTS: Eight studies met the eligibility criteria and were included in this review, including two randomized controlled trials that were designed to evaluate the efficacy of absorbent products on the prevention or management of incontinence-associated dermatitis. Quality of evidence was assessed as low or very low. The findings revealed that some outcomes related to IAD prevention or improvement of IAD can be positively affected by the introduction of a new absorbent product or a difference in the frequency of pad changing, which can control the overhydration of the skin. CONCLUSIONS: The studies included in this review indicated that the problem of control of overhydration of the skin associated with urine and/or faeces can be controlled by absorbent products and these products may be effective for the prevention or management of incontinence-associated dermatitis. Future research with high-quality studies is required.

Does the presence of bacterial urinary infection contribute to the development of incontinence-associated dermatitis? A scoping review.

OBJECTIVE: Incontinence-associated dermatitis (IAD) is an inflammatory skin condition caused by the repeated exposure to urine and faeces. It is not common for urinary incontinence only to cause IAD, however patients with urinary tract infections (UTIs) are also at increased risk for IAD. This scoping review aimed to provide a summary of the relationship between bacterial urinary infections and IAD. METHODS: We conducted a scoping review following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews. PubMed, CINAHL, Medline, and Web of Science were searched for relevant articles from January 2007 through February 2020. RESULTS: Based on eligibility criteria, 13 research studies and review articles were included. Despite the acknowledged role of bacterial infections can play in IAD and the importance of eradicating infections for the prevention of skin breakdown, there have been limited studies that have investigated how uropathogenic bacteria, in combination with urine, lead to skin damage and IAD. The use of urinary catheters also predisposes to UTIs; however, prevalence/incidence rates of IAD in these patients are not clear, as they were considered as continent of urine in the included studies. CONCLUSION: Further research is needed to elucidate the mechanisms of how bacteria, in combination with urine, lead to IAD.

Ultrasound assessment of deep tissue on the wound bed and periwound skin: A classification system using ultrasound images.

AIMS: Given the utility of ultrasonography in assessing pressure injury, some ultrasonographic findings have already been used as indicators of deep tissue pressure injury. Despite reports showing that a cloud-like ultrasonographic pattern reflected the presence of deep tissue necrosis, identifying cloud-like patterns was difficult given the presence of similar findings, such as a cobblestone-like pattern. This case series reports patients with pressure injuries who presented with a cloud-like (five cases) and cobblestone-like (four cases) pattern during ultrasonography. METHODS: This study was conducted at a Japanese university hospital. Participants included patients who underwent routine examination by an interdisciplinary pressure injury team. Pressure injury severity was assessed using the DESIGN-R® scoring system and the wound size were measured using ImageJ software based on the wound photograph. RESULTS: Among the five cases showing a cloud-like pattern upon ultrasonography, all exhibited an increase in the total DESIGN-R® score, while three exhibited an increase in wound size. On the other hand, all four cases showing a cobblestone-like pattern displayed no increase in the total DESIGN-R® score and a decrease in wound size. CONCLUSION: This study suggested that distinguishing between cloud-like and cobblestone-like ultrasonography patterns is necessary for determining the presence or absence of deep tissue pressure injury. In order to comprehensively assess pressure injuries with ultrasonography, future studies should be conducted in a large number of participants.

Skin tear classification using machine learning from digital RGB image.

AIM: Skin tears are traumatic wounds characterised by separation of the skin layers. Severity evaluation is important in the management of skin tears. To support the assessment and management of skin tears, this study aimed to develop an algorithm to estimate a category of the Skin Tear Audit Research classification system (STAR classification) using digital images via machine learning. This was achieved by introducing shape features representing complicated shape of the skin tears. METHODS: A skin tear image was separated into small segments, and features of each segment were estimated. The segments were then classified into different classes by machine learning algorithms, namely support vector machine and random forest. Their performance in classifying wound segments and STAR categories was evaluated with 31 images using the leave-one-out cross validation. RESULTS: Support vector machine showed an accuracy of 74% and 69% in classifying wound segments and STAR categories, respectively. The corresponding accuracy using random forest were 71% and 63%. CONCLUSION: Machine learning algorithms revealed capable of classifying categories of skin tears. This could offer the potential to aid nurses in their management of skin tears, even if they are not specialised in wound care.