Convolutional Neural Networks in Chronic Wound Segmentation and Tissue Classification Using Real‐World Images.
Chronic wounds cause a significant burden to affected patients and to society. Effective and objective diagnostic and monitoring methods are needed in wound care, and artificial intelligence offers one promising alternative. In this study, real‐world wound images were used to train a convolutional n...
| Publicado en: | International Wound Journal Vol. 23; no. 4; pp. 1 - 9 |
|---|---|
| Autores principales: | , , , , , , |
| Formato: | pictorial research tables/charts Journal Article |
| Publicado: |
Wiley-Blackwell
Apr2026
|
| Acceso en línea: | Ver este registro en EBSCOhost |
| fields | @attributes: recordID: 1 pdfLink: plink: https://search.ebscohost.com/login.aspx?direct=true&db=ccm&AN=193225082&site=ehost-live header: @attributes: shortDbName: ccm uiTerm: 193225082 longDbName: CINAHL Complete uiTag: AN controlInfo: bkinfo: dissinfo: jinfo: jid: 17424801 1725 jtl: International Wound Journal issn: 17424801 maglogo: Y pubinfo: dt: Apr2026 vid: 23 iid: 4 pid: 480 pub: Wiley-Blackwell place: Malden, Massachusetts artinfo: ui: 193225082 193225082 193225082 10.1111/iwj.70912 193225082 ppf: 1 ppct: 8 formats: tig: atl: Convolutional Neural Networks in Chronic Wound Segmentation and Tissue Classification Using Real‐World Images. aug: au: Huttunen, Ellen Kimpimäki, Teija Salenius, Jenni E. Pölönen, Ilkka Yambasu, Thomas Huttunen, Maria Salmi, Teea affil: Faculty of Medicine and Health Technology, Tampere University, Tampere, Finland sug: subj: Convolutional Neural Networks Utilization Wounds, Chronic Classification Image Processing, Computer Assisted Leg Ulcer Diagnosis Leg Ulcer Therapy Wound Healing Wound Care Human Funding Source Finland Female Male Adult Middle Age Aged Aged, 80 and Over Retrospective Design Record Review Electronic Health Records Descriptive Statistics Data Analysis Software Comparative Studies Confidence Intervals Vasculitis Pyoderma Gangrenosum Artificial Intelligence Machine Learning Adult: 19-44 years Middle Aged: 45-64 years Aged: 65+ years Aged, 80 & over Female Male ab: Chronic wounds cause a significant burden to affected patients and to society. Effective and objective diagnostic and monitoring methods are needed in wound care, and artificial intelligence offers one promising alternative. In this study, real‐world wound images were used to train a convolutional neural network to automatically segment wound area and wound tissues on an image. The study included altogether 362 images of venous, arterial, vasculitis and pyoderma gangrenosum wounds. The model was based on a convolutional neural network architecture U‐Net, and fully supervised learning was utilised during the training phase. Wound area reached a Dice Similarity Coefficient (DSC) of 0.927 and Intersection over Union (IoU) of 0.868 using an augmented dataset with pretraining. Fibrinous exudate and granulation performed fairly well with DSC 0.750 and 0.696, and with IoU 0.659 and 0.601, respectively. Necrosis present in only 56 images achieved lower performance with DSC 0.503 and IoU 0.502. In conclusion, this study suggested that it is possible to train a neural network to perform well with images taken for purely clinical purposes. Besides wound area, several wound structures can be identified, but wound structure identification performance is dependent on the number of images featuring the structure. Summary: Clinical images taken in real‐world conditions can be utilised with CNN to provide wound area segmentation and tissue classification.The combination of augmented data, optimised input size and appropriate epoch count achieved the best segmentation performance among the models tested.Wound area and the most common structures in chronic wounds, like fibrinous exudation and granulation, can be detected with CNN quite reliably.Traditional augmentation methods did not improve the CNN performance in wound segmentation as expected; the best way to boost performance was by adding more images. pubtype: Academic Journal doctype: pictorial research tables/charts Journal Article ougenre: Article language: English refInfo: holdings: @attributes: islocal: N |
|---|