Bart Diricx
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Bart Diricx has a total of 15 co-authors. Here are the top co-authors by number of shared publications:
- Bart Dhoedt
Ghent University IDLab Ghent University IDLab Ghent University
2 shared publications - Bert Vankeirsbilck
IDLab, Department of Information and Technology, Ghent University, 9052 Gent, Belgium
2 shared publications - Pieter Simoens
Ghent University
2 shared publications - Pieter Van Molle
IDLab, Department of Information Technology Ghent University–IMEC Ghent Belgium
2 shared publications - Tim Verbelen
VERSES Research Lab, Los Angeles, USA
2 shared publications - Tom Kimpe
Healthcare Division, Barco NV, Kortrijk, Belgium
2 shared publications
Latest Publications
- Open Access
- 31 January 2023
- Published by Mattioli1885 in Dermatology Practical & Conceptual
- Vol. 13 (1) , e2023058
Can Multispectral Dermoscopy Help In Distinguishing Blue Color?
Citation: Janssen L, Van Kelst S, Diricx B, et al. Can Multispectral Dermoscopy Help In Distinguishing Blue Color? Dermatol Pract Concept. 2023;13(1):e2023058. DOI: https://doi.org/10.5826/dpc.1301a58 Accepted: May 31, 2022; Published: January 2023 Copyright: ©2023 Janssen et al. This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommercial License (BY-NC-4.0), https://creativecommons.org/licenses/by-nc/4.0/, which permits unrestricted noncommercial use, distribution, and reproduction in any medium, provided the original authors and source are credited. Funding: None. Competing interests: None. Authorship: All authors have contributed significantly to this publication. Corresponding author: Marjan Garmyn MD, PhD, Department of Dermatology, KU Leuven, Herestraat 49, 3000 Leuven, Belgium. Tel: +3216337950 Fax: +3216337951 E-mail: marjan.garmyn@kuleuven.be Introduction: The interpretation of colors is essential in the dermoscopic evaluation of skin lesions. The same blue color on white dermoscopy may indicate blood or pigment deep in the dermis. Skin parameter maps, generated from multispectral dermoscopic images, allow to more clearly visualize specific skin structures such as pigment distribution (pigment map) and vasculature (blood map). Objectives: Aim of this research is to investigate whether skin parameter maps can be used to objectively identify and distinguish the presence of pigment and blood. Methods: We retrospectively analyzed 24 blue naevi and 79 angiomas. The skin parameter maps of each of the lesions were independently reviewed by 3 expert dermoscopists, in the absence of the regular white-light dermoscopic image. Results: All the observers provided high levels of diagnostic accuracy for blue naevus and angioma based on skin parameter maps alone, and the dermoscopic diagnosis was considered substantially reliable because of the 79% of diagnostic K agreement. Percentages of blue naevi and angiomas that showed respectively deep pigment and blood were very high at 95.8% and 97.5%. There was a percentage of lesions that counterintuitively showed blood in blue naevi (37.5%) and deep pigment in angiomas (28.8%). Conclusions: Skin parameter maps based on multispectral images can help to objectify the presence of deep pigment or blood in blue naevi and angiomas. Application of these skin parameter maps could help in the differential diagnosis between pigmented and vascular lesions. Keywords : Dermoscopy, Multispectral, Blue naevus, Angioma, skin parameter maps This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
- 31 January 2023
- Published by Mattioli1885 in Dermatology Practical & Conceptual
- Vol. 13 (1) , e2023058
Open AccessCan Multispectral Dermoscopy Help In Distinguishing Blue Color?
Citation: Janssen L, Van Kelst S, Diricx B, et al. Can Multispectral Dermoscopy Help In Distinguishing Blue Color? Dermatol Pract Concept. 2023;13(1):e2023058. DOI: https://doi.org/10.5826/dpc.1301a58 Accepted: May 31, 2022; Published: January 2023 Copyright: ©2023 Janssen et al. This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommercial License (BY-NC-4.0), https://creativecommons.org/licenses/by-nc/4.0/, which permits unrestricted noncommercial use, distribution, and reproduction in any medium, provided the original authors and source are credited. Funding: None. Competing interests: None. Authorship: All authors have contributed significantly to this publication. Corresponding author: Marjan Garmyn MD, PhD, Department of Dermatology, KU Leuven, Herestraat 49, 3000 Leuven, Belgium. Tel: +3216337950 Fax: +3216337951 E-mail: marjan.garmyn@kuleuven.be Introduction: The interpretation of colors is essential in the dermoscopic evaluation of skin lesions. The same blue color on white dermoscopy may indicate blood or pigment deep in the dermis. Skin parameter maps, generated from multispectral dermoscopic images, allow to more clearly visualize specific skin structures such as pigment distribution (pigment map) and vasculature (blood map). Objectives: Aim of this research is to investigate whether skin parameter maps can be used to objectively identify and distinguish the presence of pigment and blood. Methods: We retrospectively analyzed 24 blue naevi and 79 angiomas. The skin parameter maps of each of the lesions were independently reviewed by 3 expert dermoscopists, in the absence of the regular white-light dermoscopic image. Results: All the observers provided high levels of diagnostic accuracy for blue naevus and angioma based on skin parameter maps alone, and the dermoscopic diagnosis was considered substantially reliable because of the 79% of diagnostic K agreement. Percentages of blue naevi and angiomas that showed respectively deep pigment and blood were very high at 95.8% and 97.5%. There was a percentage of lesions that counterintuitively showed blood in blue naevi (37.5%) and deep pigment in angiomas (28.8%). Conclusions: Skin parameter maps based on multispectral images can help to objectify the presence of deep pigment or blood in blue naevi and angiomas. Application of these skin parameter maps could help in the differential diagnosis between pigmented and vascular lesions. Keywords : Dermoscopy, Multispectral, Blue naevus, Angioma, skin parameter maps This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
- Open Access
Data-Efficient Sensor Upgrade Path Using Knowledge Distillation
Deep neural networks have achieved state-of-the-art performance in image classification. Due to this success, deep learning is now also being applied to other data modalities such as multispectral images, lidar and radar data. However, successfully training a deep neural network requires a large reddataset. Therefore, transitioning to a new sensor modality (e.g., from regular camera images to multispectral camera images) might result in a drop in performance, due to the limited availability of data in the new modality. This might hinder the adoption rate and time to market for new sensor technologies. In this paper, we present an approach to leverage the knowledge of a teacher network, that was trained using the original data modality, to improve the performance of a student network on a new data modality: a technique known in literature as knowledge distillation. By applying knowledge distillation to the problem of sensor transition, we can greatly speed up this process. We validate this approach using a multimodal version of the MNIST dataset. Especially when little data is available in the new modality (i.e., 10 images), training with additional teacher supervision results in increased performance, with the student network scoring a test set accuracy of 0.77, compared to an accuracy of 0.37 for the baseline. We also explore two extensions to the default method of knowledge distillation, which we evaluate on a multimodal version of the CIFAR-10 dataset: an annealing scheme for the hyperparameter and selective knowledge distillation. Of these two, the first yields the best results. Choosing the optimal annealing scheme results in an increase in test set accuracy of 6%. Finally, we apply our method to the real-world use case of skin lesion classification.Open AccessData-Efficient Sensor Upgrade Path Using Knowledge Distillation
Deep neural networks have achieved state-of-the-art performance in image classification. Due to this success, deep learning is now also being applied to other data modalities such as multispectral images, lidar and radar data. However, successfully training a deep neural network requires a large reddataset. Therefore, transitioning to a new sensor modality (e.g., from regular camera images to multispectral camera images) might result in a drop in performance, due to the limited availability of data in the new modality. This might hinder the adoption rate and time to market for new sensor technologies. In this paper, we present an approach to leverage the knowledge of a teacher network, that was trained using the original data modality, to improve the performance of a student network on a new data modality: a technique known in literature as knowledge distillation. By applying knowledge distillation to the problem of sensor transition, we can greatly speed up this process. We validate this approach using a multimodal version of the MNIST dataset. Especially when little data is available in the new modality (i.e., 10 images), training with additional teacher supervision results in increased performance, with the student network scoring a test set accuracy of 0.77, compared to an accuracy of 0.37 for the baseline. We also explore two extensions to the default method of knowledge distillation, which we evaluate on a multimodal version of the CIFAR-10 dataset: an annealing scheme for the hyperparameter and selective knowledge distillation. Of these two, the first yields the best results. Choosing the optimal annealing scheme results in an increase in test set accuracy of 6%. Finally, we apply our method to the real-world use case of skin lesion classification. - Open Access
- 1 March 2021
- Published by Springer Nature in Neural Computing and Applications
- Vol. 33 (16) , 10259-10275
Leveraging the Bhattacharyya coefficient for uncertainty quantification in deep neural networks
Modern deep learning models achieve state-of-the-art results for many tasks in computer vision, such as image classification and segmentation. However, its adoption into high-risk applications, e.g. automated medical diagnosis systems, happens at a slow pace. One of the main reasons for this is that regular neural networks do not capture uncertainty. To assess uncertainty in classification, several techniques have been proposed casting neural network approaches in a Bayesian setting. Amongst these techniques, Monte Carlo dropout is by far the most popular. This particular technique estimates the moments of the output distribution through sampling with different dropout masks. The output uncertainty of a neural network is then approximated as the sample variance. In this paper, we highlight the limitations of such a variance-based uncertainty metric and propose an novel approach. Our approach is based on the overlap between output distributions of different classes. We show that our technique leads to a better approximation of the inter-class output confusion. We illustrate the advantages of our method using benchmark datasets. In addition, we apply our metric to skin lesion classification—a real-world use case—and show that this yields promising results.
- 1 March 2021
- Published by Springer Nature in Neural Computing and Applications
- Vol. 33 (16) , 10259-10275
Open AccessLeveraging the Bhattacharyya coefficient for uncertainty quantification in deep neural networks
Modern deep learning models achieve state-of-the-art results for many tasks in computer vision, such as image classification and segmentation. However, its adoption into high-risk applications, e.g. automated medical diagnosis systems, happens at a slow pace. One of the main reasons for this is that regular neural networks do not capture uncertainty. To assess uncertainty in classification, several techniques have been proposed casting neural network approaches in a Bayesian setting. Amongst these techniques, Monte Carlo dropout is by far the most popular. This particular technique estimates the moments of the output distribution through sampling with different dropout masks. The output uncertainty of a neural network is then approximated as the sample variance. In this paper, we highlight the limitations of such a variance-based uncertainty metric and propose an novel approach. Our approach is based on the overlap between output distributions of different classes. We show that our technique leads to a better approximation of the inter-class output confusion. We illustrate the advantages of our method using benchmark datasets. In addition, we apply our metric to skin lesion classification—a real-world use case—and show that this yields promising results.