Application of multiple deep learning models for automatic burn wound assessment

Che, Chang; Ho, Chun Yee; Lai, Feipei; Christian, Mesakh; Huang, Shih-Chang; Chang, Dun Hao; Chen, Yo Shen

doi:10.1016/j.burns.2022.07.006

Cited by 12 publications

(7 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Though the combined number of public images was 11,173, the three largest proprietary datasets used contained 482,187 [ 48 ], 13,000 [ 49 ], and 8412 [ 50 ] images prior to applying any augmentation. Most of the public datasets had a simple image collection protocol (constant distance, conventional illumination, and perpendicular angle), and though samples within a single set were collected under constant conditions, using images from different datasets can be beneficial in improving dataset diversity.…”

Section: Resultsmentioning

confidence: 99%

“…Resizing was used to unify a dataset when different devices were used to capture images or when cropping resulted in multiple different image sizes. Three articles reported [ 32 , 50 , 51 ] the use of zero padding instead of resizing to avoid the distortion of image features. Non-uniform illumination was addressed by applying equalization [ 17 ], normalization [ 51 , 52 , 53 ], standardization [ 22 ], and other illumination adjustment techniques [ 16 ].…”

Section: Resultsmentioning

confidence: 99%

“…It was stated that this framework could also be used to synthesize other wound types. Three different DeeplabV3+ [ 97 ] models with the ResNet101 [ 77 ] backbone were used to estimate burn %TBSA while using the rule of palms [ 50 ]. Two models were used for wound and palm segmentation, and the third was used for deep burn segmentation.…”

Section: Resultsmentioning

confidence: 99%

See 2 more Smart Citations

Towards Home-Based Diabetic Foot Ulcer Monitoring: A Systematic Review

Kairys

Pauliukiene

Raudonis

et al. 2023

Sensors

View full text Add to dashboard Cite

It is considered that 1 in 10 adults worldwide have diabetes. Diabetic foot ulcers are some of the most common complications of diabetes, and they are associated with a high risk of lower-limb amputation and, as a result, reduced life expectancy. Timely detection and periodic ulcer monitoring can considerably decrease amputation rates. Recent research has demonstrated that computer vision can be used to identify foot ulcers and perform non-contact telemetry by using ulcer and tissue area segmentation. However, the applications are limited to controlled lighting conditions, and expert knowledge is required for dataset annotation. This paper reviews the latest publications on the use of artificial intelligence for ulcer area detection and segmentation. The PRISMA methodology was used to search for and select articles, and the selected articles were reviewed to collect quantitative and qualitative data. Qualitative data were used to describe the methodologies used in individual studies, while quantitative data were used for generalization in terms of dataset preparation and feature extraction. Publicly available datasets were accounted for, and methods for preprocessing, augmentation, and feature extraction were evaluated. It was concluded that public datasets can be used to form a bigger, more diverse datasets, and the prospects of wider image preprocessing and the adoption of augmentation require further research.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Towards Home-Based Diabetic Foot Ulcer Monitoring: A Systematic Review

Kairys

Pauliukiene

Raudonis

et al. 2023

Sensors

View full text Add to dashboard Cite

show abstract

“…Although some preliminary burn classification work using digital color images and deep learning technology had been reported prior to 2019, 27 the period from 2019 to 2023 saw a substantial increase in the use of deep learning approaches for burn wound classification. [28][29][30][31][32][33][34][35][36][37][38][39][40][41][42][43][44][45] Several studies in this time period used deep learning algorithms to segment images into burned and un-burned regions. 31,34,35,[38][39][40] A 2019 study 31 used 1,000 images to train a mask region with a convolutional neural network (Mask R-CNN) algorithm, comparing several different underlying network types and obtaining a maximum accuracy of 85% for identifying burn regions in images of different severities of burns.…”

Section: Studies From 2019 To 2023: Emergence Of Deep Learning Approa...mentioning

confidence: 99%

“…[28][29][30][31][32][33][34][35][36][37][38][39][40][41][42][43][44][45] Several studies in this time period used deep learning algorithms to segment images into burned and un-burned regions. 31,34,35,[38][39][40] A 2019 study 31 used 1,000 images to train a mask region with a convolutional neural network (Mask R-CNN) algorithm, comparing several different underlying network types and obtaining a maximum accuracy of 85% for identifying burn regions in images of different severities of burns. Another 2019 study 38 used deep learning with semantic segmentation to distinguish between burn, skin, and background portions of images.…”

Section: Studies From 2019 To 2023: Emergence Of Deep Learning Approa...mentioning

confidence: 99%

Review of machine learning for optical imaging of burn wound severity assessment

Wilson,

Rowland,

Kennedy

et al. 2024

J. Biomed. Opt.

View full text Add to dashboard Cite

Over the past decade, machine learning (ML) algorithms have rapidly become much more widespread for numerous biomedical applications, including the diagnosis and categorization of disease and injury.Aim: Here, we seek to characterize the recent growth of ML techniques that use imaging data to classify burn wound severity and report on the accuracies of different approaches.Approach: To this end, we present a comprehensive literature review of preclinical and clinical studies using ML techniques to classify the severity of burn wounds. Results:The majority of these reports used digital color photographs as input data to the classification algorithms, but recently there has been an increasing prevalence of the use of ML approaches using input data from more advanced optical imaging modalities (e.g., multispectral and hyperspectral imaging, optical coherence tomography), in addition to multimodal techniques. The classification accuracy of the different methods is reported; it typically ranges from ∼70% to 90% relative to the current gold standard of clinical judgment. Conclusions:The field would benefit from systematic analysis of the effects of different input data modalities, training/testing sets, and ML classifiers on the reported accuracy. Despite this current limitation, ML-based algorithms show significant promise for assisting in objectively classifying burn wound severity.

show abstract