Diabetic foot thermal image segmentation using Double Encoder-ResUnet (DE-ResUnet)

Bouallal, Doha; Douzi, Hassan; Harba, Rachid

doi:10.1080/03091902.2022.2077997

Cited by 8 publications

(4 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the study by Bouallal et al, the main aim was to develop an automated accurate algorithm for the segmentation of the diabetic foot [ 36 ], with the consideration that appropriate segmentation is important for the interpretation of thermal images, which can predict the onset of DFU. A dataset consisting of 398 pairs of thermal and RGB (Red, Green, Blue) images was studied, obtained from 145 diabetic patients and 54 healthy subjects.…”

Section: Artificial Intelligence In Diabetic Foot Syndrome: Methodolo...mentioning

confidence: 99%

Artificial Intelligence Methodologies Applied to Technologies for Screening, Diagnosis and Care of the Diabetic Foot: A Narrative Review

et al. 2022

View full text Add to dashboard Cite

Diabetic foot syndrome is a multifactorial pathology with at least three main etiological factors, i.e., peripheral neuropathy, peripheral arterial disease, and infection. In addition to complexity, another distinctive trait of diabetic foot syndrome is its insidiousness, due to a frequent lack of early symptoms. In recent years, it has become clear that the prevalence of diabetic foot syndrome is increasing, and it is among the diabetes complications with a stronger impact on patient’s quality of life. Considering the complex nature of this syndrome, artificial intelligence (AI) methodologies appear adequate to address aspects such as timely screening for the identification of the risk for foot ulcers (or, even worse, for amputation), based on appropriate sensor technologies. In this review, we summarize the main findings of the pertinent studies in the field, paying attention to both the AI-based methodological aspects and the main physiological/clinical study outcomes. The analyzed studies show that AI application to data derived by different technologies provides promising results, but in our opinion future studies may benefit from inclusion of quantitative measures based on simple sensors, which are still scarcely exploited.

show abstract

Section: Artificial Intelligence In Diabetic Foot Syndrome: Methodolo...mentioning

confidence: 99%

Artificial Intelligence Methodologies Applied to Technologies for Screening, Diagnosis and Care of the Diabetic Foot: A Narrative Review

et al. 2022

View full text Add to dashboard Cite

show abstract

“…The first task consists in segmenting the two feet and separating them from the rest of the background (Fig. 4), and in this study the images we are going to present are previously segmented using a neural architecture that is detailed in our previous work [23].…”

Section: B Df Thermal Analysis Methodsmentioning

confidence: 99%

Registration Methods for Thermal Images of Diabetic Foot Monitoring: A Comparative Study

Bouallal¹,

Douzi²,

Harba³

2022

IJACSA

Self Cite

View full text Add to dashboard Cite

This paper presents a comparative study of image registration techniques for Diabetic Foot (DF) thermal images. Four registration methods (Intensity-based algorithm, Iterative closest point (ICP), subpixel registration algorithm, which is mainly based on Fast Fourier Transform (FFT), and the pyramid approach for subpixel registration) have been implemented and analyzed. The performances of the four algorithms were evaluated using several overlap and symmetry metrics such as the Dice similarity coefficient (DSC), Root Mean Square Error (RMSE) and peak signal to noise ratio (PSNR). The methods were analyzed in a first step on the images of contralateral feet (right and left) of the same subject, which is called in this paper "contralateral registration" and in a second step on a pair of images of the same subject but acquired in two different times T0 and T10 after applying a cold stress test, which is called "multitemporal registration". Results showed that the intensity-based approach and the pyramid approach for subpixel registration algorithm give the best results in both types of registration (contralateral / multitemporal) and can be used efficiently for the registration of these types of images even under changing conditions.

show abstract

“…Thus, a challenge that remains in the literature is the lack of current datasets that utilize a wide variety of data modalities. It has been noted that fusing thermal and RGB images has improved the accuracy of DL methods for DFU segmentation compared to only using one or the other [39]. Additionally, including patient details such as age, sex, wound onset, ulceration history, as well as wound characteristics such as location, exudate, depth, odour, and pain have been proven critical in assessing wound status [27].…”

Section: Related Workmentioning

confidence: 99%

Protocol for Metadata and Image Collection at Diabetic Foot Ulcer Clinics: Enabling Research in Wound Analytics and Deep Learning

Basiri,

Manji,

LeLievre

et al. 2023

Preprint

View full text Add to dashboard Cite

Background: The escalating impact of diabetes and its complications, including diabetic foot ulcers (DFUs), presents global challenges in quality of life, economics, and resources, affecting around half a billion people. DFU healing is hindered by hyperglycemia-related issues and diverse diabetes-related physiological changes, necessitating ongoing personalized care. Artificial intelligence and clinical research strive to address these challenges by facilitating early detection and efficient treatments despite resource constraints. This study establishes a standardized framework for DFU data collection, introducing a dedicated case report form and a comprehensive dataset named Zivot. Results: Following this protocol, we created the Zivot dataset consisting of 269 patients with active DFUs, and about 3,700 RGB images and corresponding thermal and depth maps for the DFUs. The effectiveness of collecting a consistent and clean dataset was demonstrated using a bounding box prediction deep learning network that was constructed with EfficientNet as the feature extractor and UNet architecture. The network was trained on the Zivot dataset and the evaluation metrics showed promising values of 0.79 and 0.86 for F1-score and mAP segmentation metrics. Conclusions: This work and the Zivot database offer a foundation for further exploration of holistic and multi-modal approaches to DFU research.

show abstract

Diabetic foot thermal image segmentation using Double Encoder-ResUnet (DE-ResUnet)

Cited by 8 publications

References 32 publications

Artificial Intelligence Methodologies Applied to Technologies for Screening, Diagnosis and Care of the Diabetic Foot: A Narrative Review

Artificial Intelligence Methodologies Applied to Technologies for Screening, Diagnosis and Care of the Diabetic Foot: A Narrative Review

Registration Methods for Thermal Images of Diabetic Foot Monitoring: A Comparative Study

Protocol for Metadata and Image Collection at Diabetic Foot Ulcer Clinics: Enabling Research in Wound Analytics and Deep Learning

Contact Info

Product

Resources

About