An Image Based Object Recognition System for Wound Detection and Classification of Diabetic Foot and Venous Leg Ulcers

Hüsers, Jens; Moelleken, Maurice; Richter, Mats L.; Przysucha, Mareike; Malihi, Leila; Busch, Dorothee; Götz, Nina-Alexandra; Heggemann, Jan; Hafer, Guido; Wiemeyer, Stefan; Babitsch, Birgit; Heidemann, Gunther; Dissemond, Joachim; Erfurt‐Berge, Cornelia; Hübner, Ursula

doi:10.3233/shti220397

Cited by 3 publications

(2 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore, we do not make the comparison here. In comparison to the study conducted by Jens et al on the classification of D and V [52], we employ a smaller dataset for D and V classification. Our method achieves 1.3% lower precision than previous work, but a recall of 100%.…”

Section: Comparison With Previous Workmentioning

confidence: 99%

Multi-Class Wound Classification via High and Low-Frequency Guidance Network

Guo,

Yi,

Dong

et al. 2023

Bioengineering

View full text Add to dashboard Cite

Wound image classification is a crucial preprocessing step to many intelligent medical systems, e.g., online diagnosis and smart medical. Recently, Convolutional Neural Network (CNN) has been widely applied to the classification of wound images and obtained promising performance to some extent. Unfortunately, it is still challenging to classify multiple wound types due to the complexity and variety of wound images. Existing CNNs usually extract high- and low-frequency features at the same convolutional layer, which inevitably causes information loss and further affects the accuracy of classification. To this end, we propose a novel High and Low-frequency Guidance Network (HLG-Net) for multi-class wound classification. To be specific, HLG-Net contains two branches: High-Frequency Network (HF-Net) and Low-Frequency Network (LF-Net). We employ pre-trained models ResNet and Res2Net as the feature backbone of the HF-Net, which makes the network capture the high-frequency details and texture information of wound images. To extract much low-frequency information, we utilize a Multi-Stream Dilation Convolution Residual Block (MSDCRB) as the backbone of the LF-Net. Moreover, a fusion module is proposed to fully explore informative features at the end of these two separate feature extraction branches, and obtain the final classification result. Extensive experiments demonstrate that HLG-Net can achieve maximum accuracy of 98.00%, 92.11%, and 82.61% in two-class, three-class, and four-class wound image classifications, respectively, which outperforms the previous state-of-the-art methods.

show abstract

Section: Comparison With Previous Workmentioning

confidence: 99%

Multi-Class Wound Classification via High and Low-Frequency Guidance Network

Guo,

Yi,

Dong

et al. 2023

Bioengineering

View full text Add to dashboard Cite

show abstract

“…Diabetic foot ulcers (DFUs) are the most common chronic wounds, and they constitute a serious complication of diabetes mellitus because of its association with pain, disability, and poor quality of life [2,3]. Their prevalence has been increasing significantly over the past decade, consuming scarce health care resources [4,5]. In diabetic foot ulcers, wound treatments should aim to alleviate symptoms, promote healing, and avoid adverse outcomes, especially lower extremity amputation.…”

Section: Introductionmentioning

confidence: 99%

Topical and Capsular RM Diabetes Wound Cure for Treatment of Diabetic Foot Ulcer

Eleje¹,

Ezeude²,

Oguejiofor³

et al. 2022

SCR

View full text Add to dashboard Cite

Background: Despite the prevalence and quality of life impact of diabetic foot ulcer (DFU), novel topical and oral therapy standard-of-care remain an important research priority. To date, the lack of available data has made it difficult to assess the efficacy of topical and capsular RM diabetes wound cure for DFUs. Case Presentation: We report the case of a 59-year-old Nigerian woman P2+0 with 2 living children, a known diabetes mellitus patient since 2006 and has been compliant with her antidiabetic medications, who presented with complaint of right foot ulcer. We also report the case of a 69-year-old Nigerian man that presented with DFU. The latter is a known diabetic and hypertensive diagnosed 15 years prior to presentation. For the first case, both topical and capsular RM wound cure were applied twice daily over an area of the ulcer for 8 weeks. For the second case, only RM diabetes wound cure capsule applied twice daily was received for 8 weeks. Efficacy was analysed by lesion measurements and photographs to determine overall response rate (ORR), and complete response (CR). All doses were well tolerated, and potentially efficacious. At 8 weeks across all dose levels, subjects achieved 100.0% ORR and 50.0% CR. Conclusion: Application of topical and ingestion of the oral capsule of RM diabetes wound cure appear safe and well tolerated with some reduction in lesion pain. Some greater levels of lesion stabilization was observed in the subjects, and this demonstrated great clinical benefit in foot ulcers. A randomized, placebo-controlled trial to confirm these findings is recommended.

show abstract

A review of artificial intelligence in wound care

Ganesan,

Morris,

Guo

et al. 2024

Art Int Surg

View full text Add to dashboard Cite

Our aging population, diabetes, and obesity have fueled the growth of chronic wounds seen throughout the world. Often, wounds are a marker of poor health that leads to increased mortality rates. However, the diagnosis and treatment of these wounds are challenging. Incorrectly differentiating between chronic wounds and other complex conditions can lead to adverse events. Artificial intelligence (AI) has been shown to offer some early benefits, and we hypothesized that it may enhance wound care but also carry some notable risks. We performed a detailed search using PubMed, Scopus, Cumulated Index in Nursing and Allied Health Literature, and Web of Science for AI applications in wound care. AI was found to be applied to wound diagnosis and characterization, wound monitoring for tissue change, daily therapy, and prevention and prognostics. AI made for more efficient and accurate wound assessments, less painful assessments of chronic wounds, more personalized treatment, and improved prognostic prediction capabilities. AI also allowed for more precise at-home observation and care, facilitating earlier wound treatment as needed. Challenges associated with AI included how to best allocate AI-assisted technologies equitably, how to safely maintain patient data, and how to diversify datasets for algorithm training. Because the algorithms are not transparent, validating findings may be challenging. AI presents a powerful tool in several aspects of advanced wound care and has the potential to improve diagnoses, accelerate healing, reduce pain, and improve the cost-effectiveness of wound care. More research needs to be done into how to best incorporate AI into daily clinical practice while keeping clinicians aware of the potential risks of using these evolving technologies.

show abstract

An Image Based Object Recognition System for Wound Detection and Classification of Diabetic Foot and Venous Leg Ulcers

Cited by 3 publications

References 5 publications

Multi-Class Wound Classification via High and Low-Frequency Guidance Network

Multi-Class Wound Classification via High and Low-Frequency Guidance Network

Topical and Capsular RM Diabetes Wound Cure for Treatment of Diabetic Foot Ulcer

A review of artificial intelligence in wound care

Contact Info

Product

Resources

About