Image diagnosis models for the oral assessment of older people using convolutional neural networks: A retrospective observational study.

AIMS: The purpose of this study was to construct a model for oral assessment using deep learning image recognition technology and to verify its accuracy. BACKGROUND: The effects of oral care on older people are significant, and the Oral Assessment Guide has been used internationally as an effective oral assessment tool in clinical practice. However, additional training, education, development of user manuals and continuous support from a dental hygienist are needed to improve the inter-rater reliability of the Oral Assessment Guide. DESIGN: A retrospective observational study. METHODS: A total of 3,201 oral images of 114 older people aged >65 years were collected from five dental-related facilities. These images were divided into six categories (lips, tongue, saliva, mucosa, gingiva, and teeth or dentures) that were evaluated by images, out of the total eight items that comprise components of the Oral Assessment Guide. Each item was classified into a rating of 1, 2 or 3. A convolutional neural network, which is a deep learning method used for image recognition, was used to construct the image recognition model. The study methods comply with the STROBE checklist. RESULTS: We constructed models with a classification accuracy of 98.8% for lips, 94.3% for tongue, 92.8% for saliva, 78.6% for mucous membranes, 93.0% for gingiva and 93.6% for teeth or dentures. CONCLUSIONS: Highly accurate diagnostic imaging models using convolutional neural networks were constructed for six items of the Oral Assessment Guide and validated. In particular, for the five items of lips, tongue, saliva, gingiva, and teeth or dentures, models with a high accuracy of over 90% were obtained. RELEVANCE TO CLINICAL PRACTICE: The model built in this study has the potential to contribute to obtain reproducibility and reliability of the ratings, to shorten the time for assessment, to collaborate with dental professionals and to be used as an educational tool.

Cost-utility analysis of an advanced pressure ulcer management protocol followed by trained wound, ostomy, and continence nurses.

The high prevalence of severe pressure ulcers (PUs) is an important issue that requires to be highlighted in Japan. In a previous study, we devised an advanced PU management protocol to enable early detection of and intervention for deep tissue injury and critical colonization. This protocol was effective for preventing more severe PUs. The present study aimed to compare the cost-effectiveness of the care provided using an advanced PU management protocol, from a medical provider's perspective, implemented by trained wound, ostomy, and continence nurses (WOCNs), with that of conventional care provided by a control group of WOCNs. A Markov model was constructed for a 1-year time horizon to determine the incremental cost-effectiveness ratio of advanced PU management compared with conventional care. The number of quality-adjusted life-years gained, and the cost in Japanese yen (¥) ($US1 = ¥120; 2015) was used as the outcome. Model inputs for clinical probabilities and related costs were based on our previous clinical trial results. Univariate sensitivity analyses were performed. Furthermore, a Bayesian multivariate probability sensitivity analysis was performed using Monte Carlo simulations with advanced PU management. Two different models were created for initial cohort distribution. For both models, the expected effectiveness for the intervention group using advanced PU management techniques was high, with a low expected cost value. The sensitivity analyses suggested that the results were robust. Intervention by WOCNs using advanced PU management techniques was more effective and cost-effective than conventional care.

Clinical validity of the estimated energy requirement and the average protein requirement for nutritional status change and wound healing in older patients with pressure ulcers: A multicenter prospective cohort study.

AIM: Adequate nutritional intake is essential for pressure ulcer healing. Recently, the estimated energy requirement (30 kcal/kg) and the average protein requirement (0.95 g/kg) necessary to maintain metabolic balance have been reported. The purpose was to evaluate the clinical validity of these requirements in older hospitalized patients with pressure ulcers by assessing nutritional status and wound healing. METHODS: This multicenter prospective study carried out as a secondary analysis of a clinical trial included 194 patients with pressure ulcers aged ≥65 years from 29 institutions. Nutritional status including anthropometry and biochemical tests, and wound status by a structured severity tool, were evaluated over 3 weeks. Energy and protein intake were determined from medical records on a typical day and dichotomized by meeting the estimated average requirement. Longitudinal data were analyzed with a multivariate mixed-effects model. RESULTS: Meeting the energy requirement was associated with changes in weight (P < 0.001), arm muscle circumference (P = 0.003) and serum albumin level (P = 0.016). Meeting the protein requirement was associated with changes in weight (P < 0.001) and serum albumin level (P = 0.043). These markers decreased in patients who did not meet the requirement, but were stable or increased in those who did. Energy and protein intake were associated with wound healing for deep ulcers (P = 0.013 for both), improving exudates and necrotic tissue, but not for superficial ulcers. CONCLUSIONS: Estimated energy requirement and average protein requirement were clinically validated for prevention of nutritional decline and of impaired healing of deep pressure ulcers.

Undermining incision and healing of deep pressure ulcers: a prospective cohort study of pressure ulcers by the Japanese national hospital organization.

Undermining is one of the most challenging complications of deep pressure ulcers. Recommendations in most guidelines are based only on expert opinions. Here, we examined the relationship between surgical incision of the undermined space and pressure ulcer healing through a Japanese multicenter prospective cohort study. A total of 162 patients with undermining in 40 national hospitals in Japan were enrolled from July 2007 to June 2009. The incision group included 39 patients (24.1%) whose undermining was surgically incised during the observational period. Their 4-week follow-up data on pressure ulcer severity and areas of healthy granulation tissue were recorded as outcome variables using the DESIGN-R pressure ulcer assessment tool. The 4-week follow-up was restarted after the incision in the incision group. The outcome variables over time were compared between the two groups using a linear mixed model with or without adjustment for demographic and other variables. The incision group showed more rapid improvement in the total and granulation DESIGN-R scores compared with the nonincision group (p < 0.001 and p = 0.007, respectively, in the crude models). This study may provide the first considerable evidence to support that surgical incision of undermining may promote healing of deep pressure ulcers.

Reliability and validity of the Chinese version of DESIGN-R, an assessment instrument for pressure ulcers .

Lack of a valid and reliable assessment instrument may have contributed to a lack of wound healing research in China. The DESIGN-R tool scores the severity of pressure ulcers (PUs) and monitors healing using the observable state assessment; it was developed by the Scientific Education Committee of the Japanese Society of Pressure Ulcers (JSPU) and tested for reliability and validity. A Chinese version of DESIGN-R was developed based on Brislin's model of translationand tested for validity and reliability. Using a purposive sampling method, 44 practicing registered nurses (RNs) and 11 physicians (MDs) were recruited from 52 departments of Nanfang Hospital, Guangdong, China. Based on their experience, they were classified as general medical staff (gMS) or experienced medical staff (eMS). All used the Chinese version of DESIGN-R to assess eight photographs of PUs and descriptors. In addition, eight eMS also used the Bates- Jensen Wound Assessment Tool (BWAT) to assess the same wounds. Inter-rater reliability was high (total ICC score = 0.960). ICC inflammation/infection scores were 0.530 and 0.759 for gMS and eMS, respectively; granulation ICC scores were 0.532 and 0.794 in gMS and eMS, respectively. The correlation coefficients between the BWAT and DESIGN-R tool were >0.80 for all eight raters. The results suggest the Chinese version of DESIGN-R is valid and reliable and may be a useful scoring tool for RNs and MDs to monitor PU status in daily clinical practice. Additional research is warranted, and clinical instruments for inflammation/infection and granulation assessment must be developed for gMS.

Predictive validity of granulation tissue color measured by digital image analysis for deep pressure ulcer healing: a multicenter prospective cohort study.

This multicenter prospective cohort study examined the predictive validity of granulation tissue color evaluated by digital image analysis for deep pressure ulcer healing. Ninety-one patients with deep pressure ulcers were followed for 3 weeks. From a wound photograph taken at baseline, an image representing the granulation red index (GRI) was processed in which a redder color represented higher values. We calculated the average GRI over granulation tissue and the proportion of pixels exceeding the threshold intensity of 80 for the granulation tissue surface (%GRI80) and wound surface (%wound red index 80). In the receiver operating characteristics curve analysis, most GRI parameters had adequate discriminative values for both improvement of the DESIGN-R total score and wound closure. Ulcers were categorized by the obtained cutoff points of the average GRI (≤80, >80), %GRI80 (≤55, >55-80, >80%), and %wound red index 80 (≤25, >25-50, >50%). In the linear mixed model, higher classes for all GRI parameters showed significantly greater relative improvement in overall wound severity during the 3 weeks after adjustment for patient characteristics and wound locations. Assessment of granulation tissue color by digital image analysis will be useful as an objective monitoring tool for granulation tissue quality or surrogate outcomes of pressure ulcer healing.

Relationship between morphological characteristics and etiology of pressure ulcers in intensive care unit patients.

PURPOSE: Pressure ulcer (PU) prevention is crucial for critically ill patients in the intensive care unit, but etiologic factors leading to their development have not yet been completely elucidated. This study explores the relationships among etiologic factors, interventional nursing care, and morphological characteristics of PUs in intensive care unit patients. DESIGN: We used a qualitative exploratory method to link morphological characteristics of specific PUs to etiologic factors. METHODS: Details of individual PUs were described by sketching the PU photograph and categorized to characterize the morphology of PUs. After identification of characteristics, the development process was evaluated by in-depth review of medical records. RESULTS: The morphological characteristics of 30 PUs were organized into 4 categories. This process revealed a type of PU not previously described, which we labeled "leaf-type." These PUs were located on the lower sacrum, rhombic-oval in shape, and characterized by purpura and PU wrinkles. Possible etiologic factors for the specific PUs were divided into 4 categories: (1) the occurrence of PU risk episodes, (2) failure of the peripheral circulation, (3) periods of critical immobility, and (4) position change techniques inducing skin deformation. CONCLUSION: PU can be categorized into 4 morphological types, including a new category of leaf-shaped PU. We found that frequently repeated position changes such as lateral tilt and repeated head elevation caused deformation of the sacral skin that may play a role in PU development.

Concurrent validation and reliability of digital image analysis of granulation tissue color for clinical pressure ulcers.

Granulation tissue color is one indicator for pressure ulcer (PU) assessment. However, it entails a subjective evaluation only, and quantitative methods have not been established. We developed color indicators from digital image analysis and investigated their concurrent validity and reliability for clinical PUs. A cross-sectional study was conducted on 47 patients with 55 full-thickness PUs. After color calibration, a wound photograph was converted into three images representing red color: erythema index (EI), modified erythema index with additional color calibration (granulation red index [GRI]), and , which represents the artificially created red-green axis of L(*) a(*) b(*) color space. The mean intensity of the granulation tissue region and the percentage of pixels exceeding the optimal cutoff intensity (% intensity) were calculated. Mean GRI (ρ=0.39, p=0.007) and (ρ=0.55, p<0.001), as well as their % intensity indicators, showed positive correlations with a(*) measured by tristimulus colorimeter, but erythema index did not. They were correlated with hydroxyproline concentration in wound fluid, healthy granulation tissue area, and blood hemoglobin level. Intra- and interrater reliability of the indicator calculation using both GRI and had an intraclass correlation coefficient >0.9. GRI and from digital image analysis can quantitatively evaluate granulation tissue color of clinical PUs.

Risk factors related to the development of pressure ulcers in the critical care setting.

AIM: This study identified risk factors for pressure ulcer development early in the admission stage to determine what appropriate interventions might be conducted early in the admission stage to decrease the risk of pressure ulcer development. BACKGROUND: Among patients who develop pressure ulcers, 8-40% of them occur in a critical care setting. Therefore, the development of pressure ulcers is a common problem. DESIGN: Prospective cohort study. METHODS: The study facility was a 606-bed tertiary care hospital. Ninety eight patients admitted to the intensive care unit (ICU) or the high care unit (HCU) from 16 April 2003-15 July 2003 were evaluated in the final analysis. RESULTS: The mean patient age was 62.3 (SD 16.1) years, and the incidence of pressure ulcers in this study was 11.2% during the observation period. Multivariate analysis showed that 'emergency ICU/HCU patients' and 'infrequent turning' were related to pressure ulcer development. Patients with pressure ulcers experienced significantly fewer turns and repositionings (OR = 0.452, 95% CI: 0.212-0.966], p < 0.05. Fewer pressure ulcers developed in scheduled ICU/HCU patients than in emergency ICU/HCU patients (OR = 0.041 [95% CI: 0.004-0.470], p < 0.01). CONCLUSION: There was no relationship between pressure ulcer development and APACHE II score, or any medication that affected skin integrity. The frequency of turning and repositioning and patients with an emergency admission to the ICU/HCU can be the prognostic indicators for developing scoring system in critical care settings. RELEVANCE TO CLINICAL PRACTICE: These patients admitted directly to ICU or HCU were in a high risk group, further preventive strategies will be required.

Evaluating the effect of the new incentive system for high-risk pressure ulcer patients on wound healing and cost-effectiveness: a cohort study.

OBJECTIVE: To evaluate the effectiveness and cost-effectiveness of new incentive system for pressure ulcer management, which focused on skilled nurse staffing in terms of rate of healing and medical costs. DESIGN, SETTING AND PARTICIPANTS: A prospective cohort study included two types of groups: 39 institutions, which introduced the new incentive system, and 20 non-introduced groups (control). Sixty-seven patients suffering from severe pressure ulcers in the introduced group and 38 patients in the non-introduced group were included. Wound healing and medical costs were monitored weekly for three weeks by their skilled nurses in charge. MAIN OUTCOME MEASURES: Healing status and related medical costs. RESULTS: The introduced group showed significantly higher rate of healing compared with the control group at each weekly assessment. Multiple regression analysis revealed that the introduction of the new incentive system was independently associated with the faster healing rate (beta=3.44, P<.001). The budget impact analysis demonstrated that introducing this system could reduce cost of treating severe pressure ulcers by 1.776 billion yen per year. CONCLUSIONS: The new incentive system for the management of pressure ulcers, which focused on staffing with skilled nurses can improve healing rate with reduced medical cost.