DETR and YOLOv5: Exploring Performance and Self-Training for Diabetic Foot Ulcer Detection

Brüngel, Raphael; Friedrich, Christoph M.

doi:10.1109/cbms52027.2021.00063

Cited by 19 publications

(9 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Automated detection of DFUs based on deep learning methods allows the monitoring task to be performed without clinical intervention[92]. Brüngel et al[88] compare the DFU detection performance of YOLOv5 and DETR. Also, they compare the test results using the base model and the self-training model at different confidence levels.…”

mentioning

confidence: 99%

A Survey of Wound Image Analysis Using Deep Learning: Classification, Detection, and Segmentation

Zhang

Tian

et al. 2022

IEEE Access

View full text Add to dashboard Cite

Wounds not only harm the physical and mental health of patients, but also introduce huge medical costs. Meanwhile, there is a shortage of physicians in some areas, and clinical examinations are sometimes unreliable in wound diagnosis. Reliable wound analysis is of great importance in its diagnosis, treatment, and care. Currently, deep learning has developed rapidly in the field of computer vision and medical imaging and has become the most commonly used technique in wound image analysis. This paper studies the current research on deep learning in the field of wound image analysis, including classification, detection, and segmentation. We first review the publicly available datasets from various research, and study the preprocessing methods used in wound image analysis. Second, various models used in different deep learning tasks (classification, detection, and segmentation) and their applications in different types of wounds (e.g., burns, diabetic foot ulcers, pressure ulcers) are investigated. Finally, we discuss the challenges in the field of wound image analysis using deep learning, and provide an outlook on the research and development prospects.

show abstract

mentioning

confidence: 99%

A Survey of Wound Image Analysis Using Deep Learning: Classification, Detection, and Segmentation

Zhang

Tian

et al. 2022

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Then, the predictive model was iteratively retrained with its own most confident predictions over the unlabeled data. Self-training is a simple and effective approach and has been applied to several semi-supervised learning tasks with biomedical data [24, 25, 26, 27].…”

Section: Methodsmentioning

confidence: 99%

Estimating the Prevalence of Dementia in India Using a Semi-Supervised Machine Learning Approach

et al. 2023

View full text Add to dashboard Cite

Introduction. Accurate estimation of dementia prevalence is essential for making effective public and social care policy to support individuals and families suffering from the disease. The purpose of this paper is to estimate the prevalence of dementia in India using a semi-supervised machine learning approach based on a large nationally representative sample. Methods. The sample of this study is adults 60 years or older in the wave 1 (2017-2019) of the Longitudinal Aging Study in India (LASI). A subsample in LASI received extensive cognitive assessment and clinical consensus ratings and therefore have diagnoses of dementia. A semi-supervised machine learning model was developed to predict the status of dementia for LASI participants without diagnoses. After obtaining the predictions, sampling weights and age standardization to the World Health Organization (WHO) standard population were applied to generate the estimate for prevalence of dementia in India. Results. The prevalence of dementia for those aged 60 years and older in India was 8.44% (95% CI: 7.89%~9.01%). The age-standardized prevalence was estimated to be 8.94% (95% CI: 8.36%~9.55%). The prevalence of dementia was greater for those who were older, were females, received no education, and lived in rural areas. Discussion/Conclusion. The prevalence of dementia in India may be higher than prior estimates derived from local studies. These prevalence estimates provide the information necessary for making long-term planning of public and social care policy. The semi-supervised machine learning approach adopted in this paper may also be useful for other large population ageing studies that have a similar data structure.

show abstract

“…It ensured that our results could be easily reproducible and comparable against existing benchmarks in the field of object detection, reinforcing the reliability and stability of our approach. The YOLOv5 detection algorithm excels in achieving an optimal balance between speed and accuracy, featuring a refined architecture suitable for implementation on resource-constrained microcontrollers [30]. From the compact YOLOv5 nano model to larger variants, these models exhibit memory weights conducive to diverse applications, including military contexts.…”

Section: Methodsmentioning

confidence: 99%

Military Decision-Making Process Enhanced by Image Detection

Žigulić,

Glučina,

Lorencin

et al. 2023

Information

View full text Add to dashboard Cite

This study delves into the vital missions of the armed forces, encompassing the defense of territorial integrity, sovereignty, and support for civil institutions. Commanders grapple with crucial decisions, where accountability underscores the imperative for reliable field intelligence. Harnessing artificial intelligence, specifically, the YOLO version five detection algorithm, ensures a paradigm of efficiency and precision. The presentation of trained models, accompanied by pertinent hyperparameters and dataset specifics derived from public military insignia videos and photos, reveals a nuanced evaluation. Results scrutinized through precision, recall, map@0.5, mAP@0.95, and F1 score metrics, illuminate the supremacy of the model employing Stochastic Gradient Descent at 640 × 640 resolution: 0.966, 0.957, 0.979, 0.830, and 0.961. Conversely, the suboptimal performance of the model using the Adam optimizer registers metrics of 0.818, 0.762, 0.785, 0.430, and 0.789. These outcomes underscore the model’s potential for military object detection across diverse terrains, with future prospects considering the implementation on unmanned arial vehicles to amplify and deploy the model effectively.

show abstract

DETR and YOLOv5: Exploring Performance and Self-Training for Diabetic Foot Ulcer Detection

Cited by 19 publications

References 24 publications

A Survey of Wound Image Analysis Using Deep Learning: Classification, Detection, and Segmentation

A Survey of Wound Image Analysis Using Deep Learning: Classification, Detection, and Segmentation

Estimating the Prevalence of Dementia in India Using a Semi-Supervised Machine Learning Approach

Military Decision-Making Process Enhanced by Image Detection

Contact Info

Product

Resources

About