Automated framework for accurate segmentation of pressure ulcer images

Multi-Class Wound Classification via High and Low-Frequency Guidance Network

Wound image classification is a crucial preprocessing step to many intelligent medical systems, e.g., online diagnosis and smart medical. Recently, Convolutional Neural Network (CNN) has been widely applied to the classification of wound images and obtained promising performance to some extent. Unfortunately, it is still challenging to classify multiple wound types due to the complexity and variety of wound images. Existing CNNs usually extract high- and low-frequency features at the same convolutional layer, which inevitably causes information loss and further affects the accuracy of classification. To this end, we propose a novel High and Low-frequency Guidance Network (HLG-Net) for multi-class wound classification. To be specific, HLG-Net contains two branches: High-Frequency Network (HF-Net) and Low-Frequency Network (LF-Net). We employ pre-trained models ResNet and Res2Net as the feature backbone of the HF-Net, which makes the network capture the high-frequency details and texture information of wound images. To extract much low-frequency information, we utilize a Multi-Stream Dilation Convolution Residual Block (MSDCRB) as the backbone of the LF-Net. Moreover, a fusion module is proposed to fully explore informative features at the end of these two separate feature extraction branches, and obtain the final classification result. Extensive experiments demonstrate that HLG-Net can achieve maximum accuracy of 98.00%, 92.11%, and 82.61% in two-class, three-class, and four-class wound image classifications, respectively, which outperforms the previous state-of-the-art methods.

Machine Learning Techniques, Applications, and Potential Future Opportunities in Pressure Injuries (Bedsores) Management: A Systematic Review

Pressure Injuries (PI) are one of the most common health conditions in the United States. Most acute or long-term care patients are at risk of developing PI. Machine Learning (ML) has been utilized to manage patients with PI, in which one systematic review describes how ML is used in PI management in 32 studies. This research, different from the previous systematic review, summarizes the previous contributions of ML in PI from January 2007 to July 2022, categorizes the studies according to medical specialties, analyzes gaps, and identifies opportunities for future research directions. PRISMA guidelines were adopted using the four most common databases (PubMed, Web of Science, Scopus, and Science Direct) and other resources, which result in 90 eligible studies. The reviewed articles are divided into three categories based on PI time of occurrence: before occurrence (48%); at time of occurrence (16%); and after occurrence (36%). Each category is further broken down into sub-fields based on medical specialties, which result in sixteen specialties. Each specialty is analyzed in terms of methods, inputs, and outputs. The most relevant and potentially useful applications and methods in PI management are outlined and discussed. This includes deep learning techniques and hybrid models, integration of existing risk assessment tools with ML that leads to a partnership between provider assessment and patients' Electronic Health Records (EHR).

Application of an infrared thermography-based model to detect pressure injuries: a prospective cohort study

BACKGROUND

It is challenging to detect pressure injuries at an early stage of their development.

OBJECTIVES

To assess the ability of an infrared thermography (IRT)-based model, constructed using a convolution neural network, to reliably detect pressure injuries.

METHODS

A prospective cohort study compared validity in patients with pressure injury (n = 58) and without pressure injury (n = 205) using different methods. Each patient was followed up for 10 days.

RESULTS

The optimal cut-off values of the IRT-based model were 0·53 for identifying tissue damage 1 day before visual detection of pressure injury and 0·88 for pressure injury detection on the day visual detection is possible. Kaplan-Meier curves and Cox proportional hazard regression model analysis showed that the risk of pressure injury increased 13-fold 1 day before visual detection with a cut-off value higher than 0·53 [hazard ratio (HR) 13·04, 95% confidence interval (CI) 6·32-26·91; P < 0·001]. The ability of the IRT-based model to detect pressure injuries [area under the receiver operating characteristic curve (AUC)_{lag 0 days} , 0·98, 95% CI 0·95-1·00] was better than that of other methods.

CONCLUSIONS

The IRT-based model is a useful and reliable method for clinical dermatologists and nurses to detect pressure injuries. It can objectively and accurately detect pressure injuries 1 day before visual detection and is therefore able to guide prevention earlier than would otherwise be possible. What is already known about this topic? Detection of pressure injuries at an early stage is challenging. Infrared thermography can be used for the physiological and anatomical evaluation of subcutaneous tissue abnormalities. A convolutional neural network is increasingly used in medical imaging analysis. What does this study add? The optimal cut-off values of the IRT-based model were 0·53 for identifying tissue damage 1 day before visual detection of pressure injury and 0·88 for pressure injury detection on the day visual detection is possible. Infrared thermography-based models can be used by clinical dermatologists and nurses to detect pressure injuries at an early stage objectively and accurately.

Experimental Study on Wound Area Measurement with Mobile Devices

Healthcare treatments might benefit from advances in artificial intelligence and technological equipment such as smartphones and smartwatches. The presence of cameras in these devices with increasingly robust and precise pattern recognition techniques can facilitate the estimation of the wound area and other telemedicine measurements. Currently, telemedicine is vital to the maintenance of the quality of the treatments remotely. This study proposes a method for measuring the wound area with mobile devices. The proposed approach relies on a multi-step process consisting of image capture, conversion to grayscale, blurring, application of a threshold with segmentation, identification of the wound part, dilation and erosion of the detected wound section, identification of accurate data related to the image, and measurement of the wound area. The proposed method was implemented with the OpenCV framework. Thus, it is a solution for healthcare systems by which to investigate and treat people with skin-related diseases. The proof-of-concept was performed with a static dataset of camera images on a desktop computer. After we validated the approach’s feasibility, we implemented the method in a mobile application that allows for communication between patients, caregivers, and healthcare professionals.

A Systematic Investigation of Models for Color Image Processing in Wound Size Estimation

In recent years, research in tracking and assessing wound severity using computerized image processing has increased. With the emergence of mobile devices, powerful functionalities and processing capabilities have provided multiple non-invasive wound evaluation opportunities in both clinical and non-clinical settings. With current imaging technologies, objective and reliable techniques provide qualitative information that can be further processed to provide quantitative information on the size, structure, and color characteristics of wounds. These efficient image analysis algorithms help determine the injury features and the progress of healing in a short time. This paper presents a systematic investigation of articles that specifically address the measurement of wounds’ sizes with image processing techniques, promoting the connection between computer science and health. Of the 208 studies identified by searching electronic databases, 20 were included in the review. From the perspective of image processing color models, the most dominant model was the hue, saturation, and value (HSV) color space. We proposed that a method for measuring the wound area must implement different stages, including conversion to grayscale for further implementation of the threshold and a segmentation method to measure the wound area as the number of pixels for further conversion to metric units. Regarding devices, mobile technology is shown to have reached the level of reliable accuracy.

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.

A Systematic Overview of Recent Methods for Non-Contact Chronic Wound Analysis

Chronic wounds or wounds that are not healing properly are a worldwide health problem that affect the global economy and population. Alongside with aging of the population, increasing obesity and diabetes patients, we can assume that costs of chronic wound healing will be even higher. Wound assessment should be fast and accurate in order to reduce the possible complications, and therefore shorten the wound healing process. Contact methods often used by medical experts have drawbacks that are easily overcome by non-contact methods like image analysis, where wound analysis is fully or partially automated. Two major tasks in wound analysis on images are segmentation of the wound from the healthy skin and background, and classification of the most important wound tissues like granulation, fibrin, and necrosis. These tasks are necessary for further assessment like wound measurement or healing evaluation based on tissue representation. Researchers use various methods and algorithms for image wound analysis with the aim to outperform accuracy rates and show the robustness of the proposed methods. Recently, neural networks and deep learning algorithms have driven considerable performance improvement across various fields, which has a led to a significant rise of research papers in the field of wound analysis as well. The aim of this paper is to provide an overview of recent methods for non-contact wound analysis which could be used for developing an end-to-end solution for a fully automated wound analysis system which would incorporate all stages from data acquisition, to segmentation and classification, ending with measurement and healing evaluation.

Phomopsidione-Loaded Chitosan Polyethylene Glycol (PEG) Nanocomposite Dressing for Pressure Ulcers

Pressure ulcers are commonly associated with microbial infections on the wounds which require an effective wound dressing for treatment. Thus far, the available silver dressing has shown tremendous result, however, it may cause argyria and complicate the internal organ function. Hence, our study aims to develop and characterize phomopsidione-loaded chitosan-polyethylene glycol nanocomposite hydrogel (C/PEG/Ph) as an antimicrobial dressing. Physically, the C/PEG/Ph hydrogel demonstrated a uniform light blue color, soft, flexible, and elastic, with no aggregation form. The evaluation via Fourier Transform Infrared (FTIR) exposed the C/PEG/Ph hydrogel has a notable shift towards lower frequency at 1600 and 1554 cm^-1. For drug release test, the phomopsidione attained plateau at 24 h, with a total release of 67.9 ± 6.4% from the C/PEG/Ph hydrogel. There was a null burst release effect discovered throughout the experimental period. The C/PEG/Ph hydrogel showed significant results against all 4 Gram-negative bacteria and 1 yeast, with 99.99-100% reduction of microbial growth. The findings revealed that the C/PEG/Ph hydrogel can potentially act as an antimicrobial dressing for pressure ulcers.

Automated framework for accurate segmentation of leaf images for plant health assessment

Leaf segmentation is significantly important in assisting ecologists to automatically detect symptoms of disease and other stressors affecting trees. This paper employs state-of-the-art techniques in image processing to introduce an accurate framework for segmenting leaves and diseased leaf spots from images. The proposed framework integrates an appearance model that visually represents the current input image with the color prior information generated from RGB color images that were formerly saved in our database. Our framework consists of four main steps: (1) Enhancing the accuracy of the segmentation at minimum time by making use of contrast changes to automatically identify the region of interest (ROI) of the entire leaf, where the pixel-wise intensity relations are described by an electric field energy model. (2) Modeling the visual appearance of the input image using a linear combination of discrete Gaussians (LCDG) to predict the marginal probability distributions of the grayscale ROI main three classes. (3) Calculating the pixel-wise probabilities of these three classes for the color ROI based on the color prior information of database images that are segmented manually, where the current and prior pixel-wise probabilities are used to find the initial labels. (4) Refining the labels with the generalized Gauss-Markov random field model (GGMRF), which maintains the continuity. The proposed segmentation approach was applied to the leaves of mangrove trees in Abu Dhabi in the United Arab Emirates. Experimental validation showed high accuracy, with a Dice similarity coefficient 90% for distinguishing leaf spot from healthy leaf area.

Burn image segmentation based on Mask Regions with Convolutional Neural Network deep learning framework: more accurate and more convenient

BACKGROUND

Burns are life-threatening with high morbidity and mortality. Reliable diagnosis supported by accurate burn area and depth assessment is critical to the success of the treatment decision and, in some cases, can save the patient's life. Current techniques such as straight-ruler method, aseptic film trimming method, and digital camera photography method are not repeatable and comparable, which lead to a great difference in the judgment of burn wounds and impede the establishment of the same evaluation criteria. Hence, in order to semi-automate the burn diagnosis process, reduce the impact of human error, and improve the accuracy of burn diagnosis, we include the deep learning technology into the diagnosis of burns.

METHOD

This article proposes a novel method employing a state-of-the-art deep learning technique to segment the burn wounds in the images. We designed this deep learning segmentation framework based on the Mask Regions with Convolutional Neural Network (Mask R-CNN). For training our framework, we labeled 1150 pictures with the format of the Common Objects in Context (COCO) data set and trained our model on 1000 pictures. In the evaluation, we compared the different backbone networks in our framework. These backbone networks are Residual Network-101 with Atrous Convolution in Feature Pyramid Network (R101FA), Residual Network-101 with Atrous Convolution (R101A), and InceptionV2-Residual Network with Atrous Convolution (IV2RA). Finally, we used the Dice coefficient (DC) value to assess the model accuracy.

RESULT

The R101FA backbone network gains the highest accuracy 84.51% in 150 pictures. Moreover, we chose different burn depth pictures to evaluate these three backbone networks. The R101FA backbone network gains the best segmentation effect in superficial, superficial thickness, and deep partial thickness. The R101A backbone network gains the best segmentation effect in full-thickness burn.

CONCLUSION

This deep learning framework shows excellent segmentation in burn wound and extremely robust in different burn wound depths. Moreover, this framework just needs a suitable burn wound image when analyzing the burn wound. It is more convenient and more suitable when using in clinics compared with the traditional methods. And it also contributes more to the calculation of total body surface area (TBSA) burned.