부산대학교 도서관

Pressure ulcers are a challenge for patients and healthcare professionals. In the UK, pressure ulcers affect 700,000 people each year. Treating them costs the National Health Service €3.8 million every day. Their etiology is complex and multifactorial. However, evidence has shown a strong link between old age, disease-related sedentary lifestyles, and unhealthy eating habits. Direct skin contact with a bed or chair without frequent position changes can cause pressure ulcers. Urinary and faecal incontinence, diabetes, and injuries that restrict body position and nutrition are also known risk factors. Guidelines and treatments exist but their implementation and success vary across different healthcare settings. This is primarily because healthcare practitioners have a) minimal experience in dealing with pressure ulcers, and b) a general lack of understanding of pressure ulcer treatments. Poorly managed, pressure ulcers can lead to severe pain, a poor quality of life, and significant healthcare costs. In this paper, we report the findings of a clinical trial conducted by Mersey Care NHS Foundation Trust that evaluated the performance of a faster region-based convolutional neural network and mobile platform that categorised and documented pressure ulcers automatically. The neural network classifies category I, II, III, and IV pressure ulcers, deep tissue injuries, and pressure ulcers that are unstageable. District nurses used their mobile phones to take pictures of pressure ulcers and transmit them over 4/5G communications to an inferencing server for classification. The approach uses existing deep learning technologies to provide a novel end-to-end pipeline for pressure ulcer categorisation that works in ad hoc domiciliary settings. The strength of the approach resides within MLOPS, model deployment at scale, and the platforms in-situ operation. While solutions exist in the NHS for analysing pressure ulcers none of them automatically classify and report pressure ulcers from a service users’ residential home automatically. We acknowledge that there is a great deal of work to do, but the approach offers a convincing solution to standardise pressure ulcer categorisation and reporting. The results from the study are encouraging and show that using 216 images, collected over an eight-month trial, it was possible to generate a mean average Precision=0.6796, Recall=0.6997, F1-Score=0.6786 with 45 false positives using an @.75 confidence score threshold.