Kibrom Berihu Girum
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Kibrom Berihu Girum has a total of 37 co-authors. Here are the top co-authors by number of shared publications:
Raabid Hussain
ImViA laboratory, University of Burgundy, Dijon, France
5 shared publicationsAlain Lalande
University of Burgundy
4 shared publicationsGilles Créhange
University of Burgundy
4 shared publications- Alain Lalande
Université Bourgogne Europe, CHU Dijon Bourgogne, Pôle Imagerie, CNRS, ICMUB UMR 6302
2 shared publications Youssef Skandarani
ImViA laboratory, University of Burgundy, Dijon, France
2 shared publicationsAbdul Qayyum
ImViA laboratory, University of Burgundy, Dijon, France
1 shared publication
Latest Publications
- 4 April 2026
- Published by Agence Bibliographique de l'Enseignement Supérieur
Artificial intelligence for image-guided prostate brachytherapy procedures
Utilisation de l'intelligence artificielle pour les procédures de curiethérapie prostatique guidée par l'image: Les procédures de radiothérapie visent à exposer les cellules cancéreuses aux rayonnements ionisants. L'implantation permanente de sources radioactives à proximité des cellules cancéreuses est une technique classique pour guérir le cancer de la prostate à un stade précoce. Le processus implique l'acquisition d'images du patient, la délimitation des volumes cibles et des organes à risque à l'aide de l'imagerie, la planification du traitement, l’implantation de grains radioactifs guidées par l'image et l'évaluation post-implantatoire. L'analyse d'images médicales basée sur l'intelligence artificielle peut être bénéfique pour des procédures de radiothérapie. Elle peut aider à faciliter et à améliorer l'efficacité des procédures en segmentant automatiquement les organes cibles les images et en extrapolant des informations cliniquement pertinentes. Cependant, la délimitation manuelle des volumes cibles est toujours la routine standard pour la plupart des centres cliniques, ce qui prend du temps et n'est pas à l'abri de variations intra et inter-observateurs. Dans cette thèse, nous visons à développer des solutions de traitement d'images médicales pour automatiser divers étapes des procédures actuelles de curiethérapie de la prostate guidée par l'image, notamment l'identification des grains radioactifs à partir d'images de scanner X et la segmentation du volume cible clinique à partir d'images médicales.Dans la première application, nous avons développé et évalué une nouvelle technique de détection et d'identification des grains radioactifs implantés sur des scanner X post-implantatoire en rapport avec la curiethérapie prostatique. Cela permet aux experts d'évaluer la qualité du positionnement de grains radioactifs guidées par l'image en calculant les paramètres dosimétriques, en particulier le calcul de dosimétrie post-implantoire de la curiethérapie de rattrapage de la prostate réalisée des années après la curiethérapie initiale dans le cadre de récidive de cancer de la prostate. La deuxième application impliquait le développement de méthodes d'apprentissage profond pour délimiter automatiquement les volumes cibles cliniques. Nous avons évalué les méthodes proposées sur une base de données cliniques d'échographie transrectale peropératoire et des images scanner X post-implantoires de la curiethérapie prostatique guidée par l'image. L'évaluation de notre méthode a été ensuite étendue à d'autres applications d'analyse d'images médicales. Nos méthodes ont donné des résultats prometteurs menant à une perspective essentielle pour des tâches d'analyse d'images médicales efficaces et précises. Elles peuvent être rebuvant être appliquées pour automatiser la gestion des procédures de curiethérapie prostatique guidée par l'image.
- 4 April 2026
- Published by Agence Bibliographique de l'Enseignement Supérieur
Artificial intelligence for image-guided prostate brachytherapy procedures
Utilisation de l'intelligence artificielle pour les procédures de curiethérapie prostatique guidée par l'image: Les procédures de radiothérapie visent à exposer les cellules cancéreuses aux rayonnements ionisants. L'implantation permanente de sources radioactives à proximité des cellules cancéreuses est une technique classique pour guérir le cancer de la prostate à un stade précoce. Le processus implique l'acquisition d'images du patient, la délimitation des volumes cibles et des organes à risque à l'aide de l'imagerie, la planification du traitement, l’implantation de grains radioactifs guidées par l'image et l'évaluation post-implantatoire. L'analyse d'images médicales basée sur l'intelligence artificielle peut être bénéfique pour des procédures de radiothérapie. Elle peut aider à faciliter et à améliorer l'efficacité des procédures en segmentant automatiquement les organes cibles les images et en extrapolant des informations cliniquement pertinentes. Cependant, la délimitation manuelle des volumes cibles est toujours la routine standard pour la plupart des centres cliniques, ce qui prend du temps et n'est pas à l'abri de variations intra et inter-observateurs. Dans cette thèse, nous visons à développer des solutions de traitement d'images médicales pour automatiser divers étapes des procédures actuelles de curiethérapie de la prostate guidée par l'image, notamment l'identification des grains radioactifs à partir d'images de scanner X et la segmentation du volume cible clinique à partir d'images médicales.Dans la première application, nous avons développé et évalué une nouvelle technique de détection et d'identification des grains radioactifs implantés sur des scanner X post-implantatoire en rapport avec la curiethérapie prostatique. Cela permet aux experts d'évaluer la qualité du positionnement de grains radioactifs guidées par l'image en calculant les paramètres dosimétriques, en particulier le calcul de dosimétrie post-implantoire de la curiethérapie de rattrapage de la prostate réalisée des années après la curiethérapie initiale dans le cadre de récidive de cancer de la prostate. La deuxième application impliquait le développement de méthodes d'apprentissage profond pour délimiter automatiquement les volumes cibles cliniques. Nous avons évalué les méthodes proposées sur une base de données cliniques d'échographie transrectale peropératoire et des images scanner X post-implantoires de la curiethérapie prostatique guidée par l'image. L'évaluation de notre méthode a été ensuite étendue à d'autres applications d'analyse d'images médicales. Nos méthodes ont donné des résultats prometteurs menant à une perspective essentielle pour des tâches d'analyse d'images médicales efficaces et précises. Elles peuvent être rebuvant être appliquées pour automatiser la gestion des procédures de curiethérapie prostatique guidée par l'image.
- 24 March 2022
- Published by Elsevier in Medical Image Analysis
- Vol. 79, 102428
Deep learning methods for automatic evaluation of delayed enhancement-MRI. The results of the EMIDEC challenge
- 24 March 2022
- Published by Elsevier in Medical Image Analysis
- Vol. 79, 102428
Deep learning methods for automatic evaluation of delayed enhancement-MRI. The results of the EMIDEC challenge
- 19 February 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 40 (6) , 1542-1554
Learning With Context Feedback Loop for Robust Medical Image Segmentation
Deep learning has successfully been leveraged for medical image segmentation. It employs convolutional neural networks (CNN) to learn distinctive image features from a defined pixel-wise objective function. However, this approach can lead to less output pixel interdependence producing incomplete and unrealistic segmentation results. In this paper, we present a fully automatic deep learning method for robust medical image segmentation by formulating the segmentation problem as a recurrent framework using two systems. The first one is a forward system of an encoder-decoder CNN that predicts the segmentation result from the input image. The predicted probabilistic output of the forward system is then encoded by a fully convolutional network (FCN)-based context feedback system. The encoded feature space of the FCN is then integrated back into the forward system’s feed-forward learning process. Using the FCN-based context feedback loop allows the forward system to learn and extract more high-level image features and fix previous mistakes, thereby improving prediction accuracy over time. Experimental results, performed on four different clinical datasets, demonstrate our method’s potential application for single and multi-structure medical image segmentation by outperforming the state of the art methods. With the feedback loop, deep learning methods can now produce results that are both anatomically plausible and robust to low contrast images. Therefore, formulating image segmentation as a recurrent framework of two interconnected networks via context feedback loop can be a potential method for robust and efficient medical image analysis.
- 19 February 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 40 (6) , 1542-1554
Learning With Context Feedback Loop for Robust Medical Image Segmentation
Deep learning has successfully been leveraged for medical image segmentation. It employs convolutional neural networks (CNN) to learn distinctive image features from a defined pixel-wise objective function. However, this approach can lead to less output pixel interdependence producing incomplete and unrealistic segmentation results. In this paper, we present a fully automatic deep learning method for robust medical image segmentation by formulating the segmentation problem as a recurrent framework using two systems. The first one is a forward system of an encoder-decoder CNN that predicts the segmentation result from the input image. The predicted probabilistic output of the forward system is then encoded by a fully convolutional network (FCN)-based context feedback system. The encoded feature space of the FCN is then integrated back into the forward system’s feed-forward learning process. Using the FCN-based context feedback loop allows the forward system to learn and extract more high-level image features and fix previous mistakes, thereby improving prediction accuracy over time. Experimental results, performed on four different clinical datasets, demonstrate our method’s potential application for single and multi-structure medical image segmentation by outperforming the state of the art methods. With the feedback loop, deep learning methods can now produce results that are both anatomically plausible and robust to low contrast images. Therefore, formulating image segmentation as a recurrent framework of two interconnected networks via context feedback loop can be a potential method for robust and efficient medical image analysis.
- 29 January 2021
- Published by Springer Nature
- p. 378-384
Automatic Myocardial Infarction Evaluation from Delayed-Enhancement Cardiac MRI Using Deep Convolutional Networks
- 29 January 2021
- Published by Springer Nature
- p. 378-384
Automatic Myocardial Infarction Evaluation from Delayed-Enhancement Cardiac MRI Using Deep Convolutional Networks
- 20 July 2020
- Published by Springer Nature in International Journal of Computer Assisted Radiology and Surgery
- Vol. 15 (9) , 1467-1476
A deep learning method for real-time intraoperative US image segmentation in prostate brachytherapy
Purpose This paper addresses the detection of the clinical target volume (CTV) in transrectal ultrasound (TRUS) image-guided intraoperative for permanent prostate brachytherapy. Developing a robust and automatic method to detect the CTV on intraoperative TRUS images is clinically important to have faster and reproducible interventions that can benefit both the clinical workflow and patient health. Methods We present a multi-task deep learning method for an automatic prostate CTV boundary detection in intraoperative TRUS images by leveraging both the low-level and high-level (prior shape) information. Our method includes a channel-wise feature calibration strategy for low-level feature extraction and learning-based prior knowledge modeling for prostate CTV shape reconstruction. It employs CTV shape reconstruction from automatically sampled boundary surface coordinates (pseudo-landmarks) to detect the low-contrast and noisy regions across the prostate boundary, while being less biased from shadowing, inherent speckles, and artifact signals from the needle and implanted radioactive seeds. Results The proposed method was evaluated on a clinical database of 145 patients who underwent permanent prostate brachytherapy under TRUS guidance. Our method achieved a mean accuracy of \( 0.96 \pm 0.01\) and a mean surface distance error of \(0.10 \pm 0.06 \, \hbox {mm}\). Extensive ablation and comparison studies show that our method outperformed previous deep learning-based methods by more than 7% for the Dice similarity coefficient and 6.9 mm reduced 3D Hausdorff distance error. Conclusion Our study demonstrates the potential of shape model-based deep learning methods for an efficient and accurate CTV segmentation in an ultrasound-guided intervention. Moreover, learning both low-level features and prior shape knowledge with channel-wise feature calibration can significantly improve the performance of deep learning methods in medical image segmentation.- 20 July 2020
- Published by Springer Nature in International Journal of Computer Assisted Radiology and Surgery
- Vol. 15 (9) , 1467-1476
A deep learning method for real-time intraoperative US image segmentation in prostate brachytherapy
Purpose This paper addresses the detection of the clinical target volume (CTV) in transrectal ultrasound (TRUS) image-guided intraoperative for permanent prostate brachytherapy. Developing a robust and automatic method to detect the CTV on intraoperative TRUS images is clinically important to have faster and reproducible interventions that can benefit both the clinical workflow and patient health. Methods We present a multi-task deep learning method for an automatic prostate CTV boundary detection in intraoperative TRUS images by leveraging both the low-level and high-level (prior shape) information. Our method includes a channel-wise feature calibration strategy for low-level feature extraction and learning-based prior knowledge modeling for prostate CTV shape reconstruction. It employs CTV shape reconstruction from automatically sampled boundary surface coordinates (pseudo-landmarks) to detect the low-contrast and noisy regions across the prostate boundary, while being less biased from shadowing, inherent speckles, and artifact signals from the needle and implanted radioactive seeds. Results The proposed method was evaluated on a clinical database of 145 patients who underwent permanent prostate brachytherapy under TRUS guidance. Our method achieved a mean accuracy of \( 0.96 \pm 0.01\) and a mean surface distance error of \(0.10 \pm 0.06 \, \hbox {mm}\). Extensive ablation and comparison studies show that our method outperformed previous deep learning-based methods by more than 7% for the Dice similarity coefficient and 6.9 mm reduced 3D Hausdorff distance error. Conclusion Our study demonstrates the potential of shape model-based deep learning methods for an efficient and accurate CTV segmentation in an ultrasound-guided intervention. Moreover, learning both low-level features and prior shape knowledge with channel-wise feature calibration can significantly improve the performance of deep learning methods in medical image segmentation.- 11 July 2020
- Published by Springer Nature in International Journal of Computer Assisted Radiology and Surgery
- Vol. 15 (9) , 1437-1444
Fast interactive medical image segmentation with weakly supervised deep learning method
Purpose To achieve accurate image segmentation, which is the first critical step in medical image analysis and interventions, using deep neural networks seems a promising approach provided sufficiently large and diverse annotated data from experts. However, annotated datasets are often limited because it is prone to variations in acquisition parameters and require high-level expert’s knowledge, and manually labeling targets by tracing their contour is often laborious. Developing fast, interactive, and weakly supervised deep learning methods is thus highly desirable. Methods We propose a new efficient deep learning method to accurately segment targets from images while generating an annotated dataset for deep learning methods. It involves a generative neural network-based prior-knowledge prediction from pseudo-contour landmarks. The predicted prior knowledge (i.e., contour proposal) is then refined using a convolutional neural network that leverages the information from the predicted prior knowledge and the raw input image. Our method was evaluated on a clinical database of 145 intraoperative ultrasound and 78 postoperative CT images of image-guided prostate brachytherapy. It was also evaluated on a cardiac multi-structure segmentation from 450 2D echocardiographic images. Results Experimental results show that our model can segment the prostate clinical target volume in 0.499 s (i.e., 7.79 milliseconds per image) with an average Dice coefficient of 96.9 ± 0.9% and 95.4 ± 0.9%, 3D Hausdorff distance of 4.25 ± 4.58 and 5.17 ± 1.41 mm, and volumetric overlap ratio of 93.9 ± 1.80% and 91.3 ± 1.70 from TRUS and CT images, respectively. It also yielded an average Dice coefficient of 96.3 ± 1.3% on echocardiographic images. Conclusions We proposed and evaluated a fast, interactive deep learning method for accurate medical image segmentation. Moreover, our approach has the potential to solve the bottleneck of deep learning methods in adapting to inter-clinical variations and speed up the annotation processes.
- 11 July 2020
- Published by Springer Nature in International Journal of Computer Assisted Radiology and Surgery
- Vol. 15 (9) , 1437-1444
Fast interactive medical image segmentation with weakly supervised deep learning method
Purpose To achieve accurate image segmentation, which is the first critical step in medical image analysis and interventions, using deep neural networks seems a promising approach provided sufficiently large and diverse annotated data from experts. However, annotated datasets are often limited because it is prone to variations in acquisition parameters and require high-level expert’s knowledge, and manually labeling targets by tracing their contour is often laborious. Developing fast, interactive, and weakly supervised deep learning methods is thus highly desirable. Methods We propose a new efficient deep learning method to accurately segment targets from images while generating an annotated dataset for deep learning methods. It involves a generative neural network-based prior-knowledge prediction from pseudo-contour landmarks. The predicted prior knowledge (i.e., contour proposal) is then refined using a convolutional neural network that leverages the information from the predicted prior knowledge and the raw input image. Our method was evaluated on a clinical database of 145 intraoperative ultrasound and 78 postoperative CT images of image-guided prostate brachytherapy. It was also evaluated on a cardiac multi-structure segmentation from 450 2D echocardiographic images. Results Experimental results show that our model can segment the prostate clinical target volume in 0.499 s (i.e., 7.79 milliseconds per image) with an average Dice coefficient of 96.9 ± 0.9% and 95.4 ± 0.9%, 3D Hausdorff distance of 4.25 ± 4.58 and 5.17 ± 1.41 mm, and volumetric overlap ratio of 93.9 ± 1.80% and 91.3 ± 1.70 from TRUS and CT images, respectively. It also yielded an average Dice coefficient of 96.3 ± 1.3% on echocardiographic images. Conclusions We proposed and evaluated a fast, interactive deep learning method for accurate medical image segmentation. Moreover, our approach has the potential to solve the bottleneck of deep learning methods in adapting to inter-clinical variations and speed up the annotation processes.
- 10 October 2019
- Published by Springer Nature
- p. 119-127
Deep Generative Model-Driven Multimodal Prostate Segmentation in Radiotherapy
- 10 October 2019
- Published by Springer Nature
- p. 119-127
Deep Generative Model-Driven Multimodal Prostate Segmentation in Radiotherapy