Hekalo, Amar - Chair of Computer Science VI - Artificial Intelligence and Applied Computer Science

University of Würzburg
Chair of Artificial Intelligence
and Knowledge Systems
Am Hubland
D-97074 Würzburg

Room: B001
Phone: +49 931 / 31 - 86385

amar.hekalo@uni-wuerzburg.de

About Me

After completing my master's degree in physics, I joined the Chair of Artificial Intelligence and Knowledge Systems in November 2018. Since then, I have been researching, among others with the Uniklink Würzburg, to improve and make documentation in medicine more efficient with the help of language and image processing via Deep Learning and rule-based methods.

Research Interests

My research interests are especially in automated image processing and NLP. This includes in particular methods of information extraction, image classification, image segmentation as well as object recognition, specifically latest architectures, improvement of training (e.g. data augmentation and hyperparameter optimization) as well as medical applications.

Currently I am looking for HiWis with experience in frontend with Angular and backend with NodeJS to support me in the development of a documentation and e-learning tool for radiology. If you are interested, you can simply reach me by mail.

Projects

I am also a member of the ESF-ZDEX ("Center for Digital Experimentation") project, which aims to bring current research topics closer to SMEs in the Lower Franconia region. For this purpose, we are in constant exchange with our cooperation partners in the form of regular meetings, projects and events. New cooperation partners are always welcome.

Teaching

WS 21	Software Practical Course (Supervisor for Topics)
Since SS 20	Programming with Neural Networks (Lecture and exercise)
SS19	Software Engineering (Exercise)
Since SS 19	Seminar "Current Trends in Artificial Intelligence“ (Topics)
Since WS 18/19	Artificial Intelligence 2 (Exercise)

Publications

QBI: Quantile-based Bias Initialization for Efficient Private Data Reconstruction in Federated Learning. Nowak, Micha V.; Bott, Tim P.; Khachaturov, David; Puppe, Frank; Krenzer, Adrian; Hekalo, Amar. 2024.
- [ BibTeX ]
- [ URL ]
@misc{nowak2024qbiquantilebasedbiasinitialization, author = {Nowak, Micha V. and Bott, Tim P. and Khachaturov, David and Puppe, Frank and Krenzer, Adrian and Hekalo, Amar}, keywords = {myown}, title = {QBI: Quantile-based Bias Initialization for Efficient Private Data Reconstruction in Federated Learning}, year = 2024 }
Perfect Match: Radiomics and Artificial Intelligence in Cardiac Imaging. Baeßler, Bettina; Engelhardt, Sandy; Hekalo, Amar; Hennemuth, Anja; Hüllebrand, Markus; Laube, Ann; Scherer, Clemens; Tölle, Malte; Wech, Tobias. In Circulation: Cardiovascular Imaging, 17(6), p. e015490. 2024.
- [ Abstract ]
- [ BibTeX ]
- [ URL ]
Cardiovascular diseases remain a significant health burden, with imaging modalities like echocardiography, cardiac computed tomography, and cardiac magnetic resonance imaging playing a crucial role in diagnosis and prognosis. However, the inherent heterogeneity of these diseases poses challenges, necessitating advanced analytical methods like radiomics and artificial intelligence. Radiomics extracts quantitative features from medical images, capturing intricate patterns and subtle variations that may elude visual inspection. Artificial intelligence techniques, including deep learning, can analyze these features to generate knowledge, define novel imaging biomarkers, and support diagnostic decision-making and outcome prediction. Radiomics and artificial intelligence thus hold promise for significantly enhancing diagnostic and prognostic capabilities in cardiac imaging, paving the way for more personalized and effective patient care. This review explores the synergies between radiomics and artificial intelligence in cardiac imaging, following the radiomics workflow and introducing concepts from both domains. Potential clinical applications, challenges, and limitations are discussed, along with solutions to overcome them.

@article{doi:10.1161/CIRCIMAGING.123.015490, abstract = {Cardiovascular diseases remain a significant health burden, with imaging modalities like echocardiography, cardiac computed tomography, and cardiac magnetic resonance imaging playing a crucial role in diagnosis and prognosis. However, the inherent heterogeneity of these diseases poses challenges, necessitating advanced analytical methods like radiomics and artificial intelligence. Radiomics extracts quantitative features from medical images, capturing intricate patterns and subtle variations that may elude visual inspection. Artificial intelligence techniques, including deep learning, can analyze these features to generate knowledge, define novel imaging biomarkers, and support diagnostic decision-making and outcome prediction. Radiomics and artificial intelligence thus hold promise for significantly enhancing diagnostic and prognostic capabilities in cardiac imaging, paving the way for more personalized and effective patient care. This review explores the synergies between radiomics and artificial intelligence in cardiac imaging, following the radiomics workflow and introducing concepts from both domains. Potential clinical applications, challenges, and limitations are discussed, along with solutions to overcome them.}, author = {Baeßler, Bettina and Engelhardt, Sandy and Hekalo, Amar and Hennemuth, Anja and Hüllebrand, Markus and Laube, Ann and Scherer, Clemens and Tölle, Malte and Wech, Tobias}, journal = {Circulation: Cardiovascular Imaging}, keywords = {myown}, number = 6, pages = {e015490}, title = {Perfect Match: Radiomics and Artificial Intelligence in Cardiac Imaging}, volume = 17, year = 2024 }
A Real-Time Polyp-Detection System with Clinical Application in Colonoscopy Using Deep Convolutional Neural Networks. Krenzer, Adrian; Banck, Michael; Makowski, Kevin; Hekalo, Amar; Fitting, Daniel; Troya, Joel; Sudarevic, Boban; Zoller, Wolfgang G.; Hann, Alexander; Puppe, Frank. In Journal of Imaging, 9(2). 2023.
- [ Abstract ]
- [ BibTeX ]
- [ URL ]
Colorectal cancer (CRC) is a leading cause of cancer-related deaths worldwide. The best method to prevent CRC is with a colonoscopy. During this procedure, the gastroenterologist searches for polyps. However, there is a potential risk of polyps being missed by the gastroenterologist. Automated detection of polyps helps to assist the gastroenterologist during a colonoscopy. There are already publications examining the problem of polyp detection in the literature. Nevertheless, most of these systems are only used in the research context and are not implemented for clinical application. Therefore, we introduce the first fully open-source automated polyp-detection system scoring best on current benchmark data and implementing it ready for clinical application. To create the polyp-detection system (ENDOMIND-Advanced), we combined our own collected data from different hospitals and practices in Germany with open-source datasets to create a dataset with over 500,000 annotated images. ENDOMIND-Advanced leverages a post-processing technique based on video detection to work in real-time with a stream of images. It is integrated into a prototype ready for application in clinical interventions. We achieve better performance compared to the best system in the literature and score a F1-score of 90.24% on the open-source CVC-VideoClinicDB benchmark.

@article{jimaging9020026, abstract = {Colorectal cancer (CRC) is a leading cause of cancer-related deaths worldwide. The best method to prevent CRC is with a colonoscopy. During this procedure, the gastroenterologist searches for polyps. However, there is a potential risk of polyps being missed by the gastroenterologist. Automated detection of polyps helps to assist the gastroenterologist during a colonoscopy. There are already publications examining the problem of polyp detection in the literature. Nevertheless, most of these systems are only used in the research context and are not implemented for clinical application. Therefore, we introduce the first fully open-source automated polyp-detection system scoring best on current benchmark data and implementing it ready for clinical application. To create the polyp-detection system (ENDOMIND-Advanced), we combined our own collected data from different hospitals and practices in Germany with open-source datasets to create a dataset with over 500,000 annotated images. ENDOMIND-Advanced leverages a post-processing technique based on video detection to work in real-time with a stream of images. It is integrated into a prototype ready for application in clinical interventions. We achieve better performance compared to the best system in the literature and score a F1-score of 90.24% on the open-source CVC-VideoClinicDB benchmark.}, author = {Krenzer, Adrian and Banck, Michael and Makowski, Kevin and Hekalo, Amar and Fitting, Daniel and Troya, Joel and Sudarevic, Boban and Zoller, Wolfgang G. and Hann, Alexander and Puppe, Frank}, journal = {Journal of Imaging}, keywords = {myown}, number = 2, title = {A Real-Time Polyp-Detection System with Clinical Application in Colonoscopy Using Deep Convolutional Neural Networks}, volume = 9, year = 2023 }
Endoscopic Detection and Segmentation of Gastroenterological Diseases with Deep Convolutional Neural Networks. Krenzer, Adrian; Hekalo, Amar; Puppe, Frank. Vol. 2595, S. Ali; C. Daul; J. Rittscher; D. Stoyanov; E. Grisan (eds.). CEUR-WS, 2020.
- [ BibTeX ]
- [ URL ]
@proceedings{krenzer2020endoscopic, author = {Krenzer, Adrian and Hekalo, Amar and Puppe, Frank}, editor = {Ali, Sharib and Daul, Christian and Rittscher, Jens and Stoyanov, Danail and Grisan, Enrico}, keywords = {myown}, month = {04}, publisher = {CEUR-WS}, title = {Endoscopic Detection and Segmentation of Gastroenterological Diseases with Deep Convolutional Neural Networks}, volume = 2595, year = 2020 }