Development of an explainable artificial intelligence model for Asian vascular wound images

Abstract

Chronic wounds contribute to significant healthcare and economic burden worldwide. Wound assessment remains challenging given its complex and dynamic nature. The use of artificial intelligence (AI) and machine learning methods in wound analysis is promising. Explainable modelling can help its integration and acceptance in healthcare systems. We aim to develop an explainable AI model for analysing vascular wound images among an Asian population. Two thousand nine hundred and fifty-seven wound images from a vascular wound image registry from a tertiary institution in Singapore were utilized. The dataset was split into training, validation and test sets. Wound images were classified into four types (neuroischaemic ulcer [NIU], surgical site infections [SSI], venous leg ulcers [VLU], pressure ulcer [PU]), measured with automatic estimation of width, length and depth and segmented into 18 wound and peri-wound features. Data pre-processing was performed using oversampling and augmentation techniques. Convolutional and deep learning models were utilized for model development. The model was evaluated with accuracy, F1 score and receiver operating characteristic (ROC) curves. Explainability methods were used to interpret AI decision reasoning. A web browser application was developed to demonstrate results of the wound AI model with explainability. After development, the model was tested on additional 15 476 unlabelled images to evaluate effectiveness. After the development on the training and validation dataset, the model performance on unseen labelled images in the test set achieved an AUROC of 0.99 for wound classification with mean accuracy of 95.9%. For wound measurements, the model achieved AUROC of 0.97 with mean accuracy of 85.0% for depth classification, and AUROC of 0.92 with mean accuracy of 87.1% for width and length determination. For wound segmentation, an AUROC of 0.95 and mean accuracy of 87.8% was achieved. Testing on unlabelled images, the model confidence score for wound classification was 82.8% with an explainability score of 60.6%. Confidence score was 87.6% for depth classification with 68.0% explainability score, while width and length measurement obtained 93.0% accuracy score with 76.6% explainability. Confidence score for wound segmentation was 83.9%, while explainability was 72.1%. Using explainable AI models, we have developed an algorithm and application for analysis of vascular wound images from an Asian population with accuracy and explainability. With further development, it can be utilized as a clinical decision support system and integrated into existing healthcare electronic systems.

The CLoCk study: A retrospective exploration of loneliness in children and young people during the COVID-19 pandemic, in England

by Kelsey McOwat, Snehal M. Pinto Pereira, Manjula D. Nugawela, Shamez N. Ladhani, Fiona Newlands, Terence Stephenson, Ruth Simmons, Malcolm G. Semple, Terry Segal, Marta Buszewicz, Isobel Heyman, Trudie Chalder, Tamsin Ford, Emma Dalrymple, Consortium , Roz Shafran

During the COVID-19 pandemic children and young people (CYP) were socially restricted during a stage of life crucial to development, potentially putting an already vulnerable population at higher risk of loneliness, social isolation, and poorer wellbeing. The objectives of this study are to conduct an exploratory analysis into loneliness before and during the pandemic, and determine which self-reported factors are associated with loneliness.

Participants from The Children with Long COVID (CLoCk) national study were invited to take part via an online survey, with a total of 31,017 participants taking part, 31,016 of which reported on their experience of loneliness. Participants retrospectively answered questions on demographics, lifestyle, physical health and mental health and loneliness before the pandemic and at the time of answering the survey. Before the pandemic 6.5% (2,006/31,016) of participants reported experiencing loneliness “Often/Always” and at the time of survey completion 17.4% (5,395/31,016) reported feeling lonelier. There was an association between meeting the research definition of long COVID and loneliness [3.49 OR, 95%CI 3.28–3.72]. CYP who reported feeling lonelier at the time of the survey than before the pandemic were assigned female at birth, older CYP, those from Black/African/Caribbean/Black British or other ethnicity groups, those that had 3–4 siblings and lived in more deprived areas.

We demonstrate associations between multiple factors and experiences of loneliness during the pandemic. There is a need for a multi-faceted integrated approach when developing interventions targeted at loneliness. It is important to follow up the CYP involved at regular intervals to investigate the progression of their experience of loneliness over time.