Convolutional neural networks for wound detection: the role of artificial intelligence in wound care

Abstract

Objective: Telemedicine is an essential support system for clinical settings outside the hospital. Recently, the importance of the model for assessment of telemedicine (MAST) has been emphasised. The development of an eHealth-supported wound assessment system using artificial intelligence is awaited. This study explored whether or not wound segmentation of a diabetic foot ulcer (DFU) and a venous leg ulcer (VLU) by a convolutional neural network (CNN) was possible after being educated using sacral pressure ulcer (PU) data sets, and which CNN architecture was superior at segmentation. Methods: CNNs with different algorithms and architectures were prepared. The four architectures were SegNet, LinkNet, U-Net and U-Net with the VGG16 Encoder Pre-Trained on ImageNet (Unet_VGG16). Each CNN learned the supervised data of sacral pressure ulcers (PUs). Results: Among the four architectures, the best results were obtained with U-Net. U-Net demonstrated the second-highest accuracy in terms of the area under the curve (0.997) and a high specificity (0.943) and sensitivity (0.993), with the highest values obtained with Unet_VGG16. U-Net was also considered to be the most practical architecture and superior to the others in that the segmentation speed was faster than that of Unet_VGG16. Conclusion: The U-Net CNN constructed using appropriately supervised data was capable of segmentation with high accuracy. These findings suggest that eHealth wound assessment using CNNs will be of practical use in the future.

Wound assessment, imaging and monitoring systems in diabetic foot ulcers: A systematic review

“…Thirty-five full text articles were assessed for eligibility for the study and 17 articles were included in the final analysis. [11][12][13][14][15][16][17][18][19][20][21][22][23][24][25][26][27] The entire process of systematic review is reported in accordance to the PRISMA guidelines and flowchart ( Figure 1).…”

“…There are five main categories of wound assessment or monitoring techniques or methods: (a) computer applications or hand-held devices (n = 5), 18,19,22,26,27 (b) mobile applications (n = 2), 15,17 (c) optical imaging (n = 2), 13,21 (d) spectroscopy or hyperspectral imaging (n = 4), 20,[23][24][25] and (e) artificial intelligence (n = 4). 11,12,14,16 Uses of wound assessment or monitoring methods were classified into: (a) wound assessment or monitoring and (b) data capturing and storage of information. Sixteen studies reported on wound assessment or monitoring [11][12][13][14][15][16][17][18][19][20][21][22][23][24][25]27 ; one study reported on data capturing and storage of information.…”

“…11,12,14,16 Uses of wound assessment or monitoring methods were classified into: (a) wound assessment or monitoring and (b) data capturing and storage of information. Sixteen studies reported on wound assessment or monitoring [11][12][13][14][15][16][17][18][19][20][21][22][23][24][25]27 ; one study reported on data capturing and storage of information. 26 Of the 16 studies (94.1%) which reported on wound assessment or monitoring, four assessed likelihood of wound healing using spectroscopy or hyperspectral imaging.…”

Competitive neural layer-based method to identify people with high risk for diabetic foot

“…A variety of methods based on AI techniques have recently been applied to diabetic foot screening. The most common ones include ANNs (20,28,29,30,31) , fuzzy logic (32,33) , and support vector machines (SVMs) (34,35) . Most systems are trained using a supervised learning approach, where an input is mapped to an output based on example input-output pairs.…”

“…Of the AI-based works, Wang et al (2017) (34) , Goyal et al. (2018) (28) , Goyal et al (2019) (30) , Li et al (2019) (33) , Wijesinghe et al (2019) (20) , and Ohura et al (2019) (31) used images to analyze diabetic foot, while Gomes et al (2015) (32) , Wang et al. (2017) (34) , Goyal et al (2018) (28) , Goyal et al (2019) (30) , Wijesinghe et al (2019) (20) , and Ohura et al (2019) (31) looked for alternatives to monitor already existing DFUs.…”