Our Mission
Enable latest AI technology for our partners in industry and clinical research
Outside the university, research projects are often tied to the goodwill and innovative strength of entrepreneurs and decision-makers. A major obstacle can also be the proper setup of a suitable project infrastructure. Especially in the field of AI research, large computing capacities and solid data management are of utmost importance.
Our mission is to make Machine Learning and Deep Learning usable not only for companies, but also for research projects.
For this purpose, we offer our longterm project expertise as well as the necessary hardware and software.
Because we believe that innovations are created by supporting the advance of AI research projects that contribute to healthcare, economy, and society.
We are specialists in the development of innovative AI solutions for medical image processing and machine learning applications.
History
The idea for Chimaera was born during the PhD studies of the co-founders Dr. Marcus Prümmer, Dr. Dieter Hahn and Dr. Volker Daum together with their mentor and advisor Prof. Dr. Joachim Hornegger. Based on their longtearm experience in machine learning research, they recognized the challenges of developing the right software tools and services for artificial intelligence projects already ten years ago. The experience has been extended also to annotating and curating training data for AI projects.
Starting with customized software development, Chimaera's portfolio expanded steadily. Today, Chimaera is certified according to DIN EN ISO 13485 and a reliable partner for research, industry and medium-sized businesses. Our team of experts in image processing, AI and hardware optimization creates customized solutions for a wide range of use cases.
As a research-oriented high-tech company, we are in constant exchange with the important specialists in the current research landscape. This enables Chimaera to draw on the latest expertise from research practice and adapt it for its customers. We maintain close contacts with our alma mater, the FAU Erlangen-Nuremberg, but also with our contacts in the international research scene.
Chimaera Timeline
Milestones in our company history
2024
Distributed processing
The Chimaera SDK now includes built-in features for remote processing (Docker, Kubernetes,…)
2023
Commercial SDK use
The Chimaera SDK is increasingly being used by customers for various applications.
2022
New website and AI services
Complete rework of website and new AI Services for small and medium businesses
2021
10 Years ISO 13485
We have successfully improved
and maintained our QMS for 10 years
2021
Chimaera AI Building Blocks
Enabling AI with tailored Software,
Infrastructure and Services
2020
Chimaera Annotation Client
The new stand-alone annotation
tool for images
2018
Chimaera SDK
Our framework for image
processing and app development
2015
Company Relocation
We moved from Effeltrich to
Erlangen
2011
ISO
Certification
Certification of
ISO 13485 and ISO 9001
2007
Company
Foundation
From an idea at the
University FAU to a high tech startup
Team
Become part of our team
From researcher to researcher – at eye level, that's Chimaera. We are looking for ambitious young professionals so that we can continue to work on cutting-edge technology.
You don't do just "something with AI" – you develop sustainable and applicable AI projects for research and industry!
Checkout current positions:
Management
From researcher to researcher – at eye level, that's Chimaera. Both in our projects and within our company, we work to the highest quality standards. Through our own experience, we know the requirements of our customers very well and meet them as sparring partners and consultants at eye level.
We treat our employees in the same way. Everybody has his special field, but a strong team achieves more than the sum of the individual performances.
Dr.-Ing. Volker Daum
After completing his university degree (with honours), Volker was with the Fraunhofer IIS until starting his PhD, financed by a grant from the international Max-Planck research school (IMPRS) for optics and photonics. In his doctoral thesis, his research was focusing on non-rigid image registration problems for multi-modal medical images.
- University of Erlangen-Nuremberg – Computer Science: focus on numerical methods and hardware optimization
- Visited the Scientific Computing and Imaging Institute (SCI) at the University of Utah – worked there as a research assistant on a project for large scale numerical simulation
Dr.-Ing. Dieter Hahn
Dieter studied Computer Science at the Friedrich-Alexander University Erlangen-Nuremberg (FAU) and made his diploma in cooperation with Siemens Corporate Research, Princeton, USA. During his time at the Pattern Recognition Lab of the FAU, he was teaching exercises for Pattern Recognition and Computer Vision with applications in Augmented Reality. He was also head of the Image Registration and Time-of-Flight Group.
- Obtained his PhD (with honours) on statistical similarity measures for medical image registration
- Diploma thesis in Computer Science (with honours) on feature-based image registration in cooperation with Siemens Corporate Research, Princeton, NJ, USA
Dr.-Ing. Marcus Prümmer
While being a researcher at the FAU, Marcus was teaching exercises on Medical Image Processing at the Pattern Recognition Lab of the FAU. Between 2005-2010 he was head of the medical image processing group at the Pattern Recognition Lab, in 2009 he became Senior researcher. In cooperation with Siemens Healthcare he developed reconstruction algorithms for cardiac C-arm CT at Stanford University, Department of Radiology, Lucas Center, CA, USA.
- University of Mannheim – Computer Science: diploma thesis on vessel segmentation and analysis
- Freelancer for Siemens Healthcare after his diploma thesis
- University of Erlangen-Nuremberg: PhD thesis on motion estimation and compensation for cardiac C-arm CT reconstruction
Advisory Board
On the pulse of research:
As a research-oriented high-tech company, we are in constant exchange with research groups and specialists in the current research landscape. This enables Chimaera to draw on the latest expertise from research practice and adapt it for its customers.
We maintain close contacts with our alma mater, the FAU Erlangen-Nuremberg, but also with our contacts in the international research scene.
This also involves researchers and clinicians at a number of renowned institutions and university hospitals.
Achim is currently the President of Germany's most innovative university – the FAU Erlangen-Nürnberg – and passionate to drive innovation and to better harness the potential of FAU's full research spectrum. While being the head of FAU's Pattern Recognition Lab, he co-founded Chimaera together with Volker, Dieter and Marcus.
Development Team
Our development team consists of experienced people with appropriate academic backgrounds in computer science, computational engineering, software development and project management.
Knowledge here ranges from standard programming languages (e.g. C/C++, Python, CMake, Shell, Batch, JavaScript, Java) to specialized hardware programming (e.g. Cuda, OpenGL, OpenCL).
Annotation Experts
We employ our own annotation experts who are highly trained in using our customized annotation software.
Through extensive project experience, this allows us to handle annotation tasks in a wide variety of healthcare and industry settings.
Sales and Marketing
Our competent sales team will be happy to answer your inquiries about services or the use of our products, quotes, or licensing options.
Contact us via this website or send an email at: sales(at)chimaera.de
References
Our partners in industry and research
Research Grants
Surgical AI Hub Germany
EUREKA joint project: Objective measurement of surgical quality through the development of a data-based value network
BMBF funding program: Development of new digital services for data-oriented value creation (DigiLeistDAT)
Our involvement: Project management sub-project “Provision of annotation tools for the preparation of surgical data, training of robust AI models”
Project partners:
- TU Dresden
- University of Heidelberg
- German Cancer Research Center DKFZ
- Karl Storz SE & Co. KG
- Chimaera GmbH
- mbits imaging GmbH
- formigas GmbH
Project term: 2024-2026 (3 years)
Multiscale HCC
Diagnosis and treatment of hepatocellular carcinoma (HCC) using omics methods
BMBF funding program: “Personalized medicine, Module I: Research consortia for systems medicine”
Our involvement: Project management sub-project “Image analysis and mining for the development of predictive and prognostic models for HCC”
Project partners:
- Eberhard Karls University of Tübingen
- University of Stuttgart
- Chimaera GmbH
Project term: 2017-2019 (3 years)
Development and implementation of new clinical procedures for MRI-based myelin quantification for the objectifiable diagnosis of neurological diseases such as multiple sclerosis or Alzheimer's disease
Central Innovation Program for SMEs (ZIM) of the Federal Ministry for Economic Affairs and Climate Action (BMWi)
Our involvement: Project management sub-project “Modular development of new algorithms and certifiable software components based on highly parallel graphics hardware for the realization of complex diagnostic procedures on MRI data for (pre-)clinical use”
Project partners:
- University of Applied Sciences Koblenz
- Chimaera GmbH
- SOVAmed GmbH
Project term: 2013-2014 (1 year)
MoBiMed
Joint project: Mechanism of targeting – angiogenesis for diagnostics and and therapy
BMBF funding program: “Molecular Imaging”
Our involvement: Associated partner of the FAU
Project partners:
- German Cancer Research Center (DKFZ)
- University of Erlangen-Nuremberg FAU
- University of Münster WWU
- University of Tübingen
- Max Planck Institute for Neurological Research
Project term: 2008-2012 (3 years)
Publications
The following list shows an excerpt of our publications in the field of medical image processing. Several peer reviewed journal articles and contributions in international top conferences underline the novelty of the acquired results for the medical imaging community.
Journals
- Ali, Mohammad; Zetzsche, Maximilian; Scholz, Michael; Hahn, Dieter; Gaffling, Simone; Heichel, Jens; Hammer, Christian; Bräuer, Lars; Paulsen, Friedrich. "New insights into the lacrimal pump", The Ocular Surface, 18, 2020, 10.1016/j.jtos.2020.07.013
- Gaffling, Simone; Daum, Volker; Steidl, Stefan; Maier, Andreas; Köstler, Harald; Hornegger, Joachim, A Gauss-Seidel Iteration Scheme for Reference-Free 3-D Histological Image Reconstruction, IEEE Transactions on Medical Imaging, vol. 34, no. 2, pp. 514-530, 2015
- Mualla, Firas; Prümmer, Marcus; Hahn, Dieter; Hornegger, Joachim, Toward Automatic Detection of Vessel Stenoses in Cerebral 3D DSA Volumes, Physics in Medicine and Biology, vol. 57, no. 9, pp. 2555-2573, 2012
- Bennewitz, Christian; Kuwert, Torsten; Han, Jingfeng; Ritt, Philipp; Hahn, Dieter; Thimister, Willie; Hornegger, Joachim; Uder, Michael; Schmidt, Daniela. Computer-aided evaluation of the anatomical accuracy of hybrid SPECT/spiral-CT imaging of lesions localized in the neck and upper abdomen. Nuclear Medicine Communications. 33(11):1153-1159, November 2012
- Kiefer, Alexander; Kuwert, Torsten; Hahn, Dieter; Hornegger, Joachim; Uder, Michael; Ritt, Philipp. Anatomical Accuracy of Abdominal Lesion Localization. Nuklearmedizin 50 (2011) No. 4 pp. 141-173
- Rohkohl, Christopher; Lauritsch, Günter; Biller, Lisa; Prümmer, Marcus; Boese, Jan; Hornegger, Joachim, Interventional 4D motion estimation and reconstruction of cardiac vasculature without motion periodicity assumption, Medical Image Analysis, vol. 14, no. 5, pp. 687-694, 2010
- Hahn, Dieter; Daum, Volker; Hornegger, Joachim. Automatic Parameter Selection for Multi-Modal Image Registration. In: IEEE Transactions on Medical Imaging 29 (2010) No. 5 pp. 1140-1155
- Prümmer, Marcus; Hornegger, Joachim. Cardiac C-arm CT: a unified framework for motion estimation and dynamic CT. In: IEEE Transactions on Medical Imaging 28 (2009) No. 11 pp. 1836-49
- Spiegel, Martin; Hahn, Dieter; Daum, Volker; Wasza, Jakob; Hornegger, Joachim. Segmentation of kidneys using a new active shape model generation technique based on non-rigid image registration. In: Computerized Medical Imaging and Graphics 33 (2009) No. 1 pp. 29-39
- Kollorz, Eva; Hahn, Dieter; Linke, Rainer; Goecke, Tamme; Hornegger, Joachim; Kuwert, Torsten. Quantification of Thyroid Volume Using 3-D Ultrasound Imaging. In: IEEE Transactions on Medical Imaging 27 (2008) No. 4 pp. 457-466
- Scherl, Holger; Hornegger, Joachim; Prümmer, Marcus; Lell, Michael. Semiautomatic Level-Set Based Segmentation and Stenosis Quantification of the Internal Carotid Artery in 3D CTA Data Sets. In: Medical Image Analysis 11 (2007) No. 1 pp. 21-34
Conferences
- Fischer, Peter; Daum, Volker; Hahn, Dieter; Prümmer, Marcus; Hornegger, Joachim, Regression Forest-Based Organ Detection in Normalized PET Images, Bildverarbeitung für die Medizin 2014, Aachen, 18.03.2014, pp. 384-389, 2014
- Eckl, Jutta; Daum, Volker; Hornegger, Joachim; Pohl, Kilian, Efficient Non-rigid Registration Algorithm Guided by Landmarks and Principle Components of a Deformation Field, International Symposium on Biomedical Imaging, Barcelona, 02.05.2012, pp. 704-707, 2012
- Gaffling, Simone, Daum, Volker, Hornegger, Joachim, Landmark-constrained 3-D Histological Imaging: A Morphology-preserving Approach, VMV 2011: Vision, Modeling & Visualization (16th International Workshop on Vision, Modeling and Visualization 2011), Berlin, 04-06.10.2011, pp. 309-316, 2011, ISBN 978-3-905673-85-2
- Holub, Wolfgang; Rohkohl, Christopher; Schuldhaus, Dominik; Prümmer, Marcus; Lauritsch, Günter; Hornegger, Joachim, 4D motion animation of coronary arteries from rotational angiography Proc. SPIE 7964 (Medical Imaging 2011: Visualization, Image-Guided Procedures, and Modeling), Lake Buena Vista, Florida, USA, February, vol. 7964, pp. 79641S, 2011
- Siegl, Christian; Hofmann, Hannes; Keck, Benjamin; Prümmer, Marcus; Hornegger, Joachim, Hardware-unabhängige Beschleunigung von Medizinischer Bildverarbeitung mit OpenCL Bildverarbeitung für die Medizin 2011, Lübeck, 20 - 22 März 2011, pp. 449-453, 2011, ISBN 978-3-642-19335-4
- Siegl, Christian; Hofmann, Hannes; Keck, Benjamin; Prümmer, Marcus; Hornegger, Joachim, OpenCL, a Viable Solution for High-performance Medical Image Reconstruction? Proceedings of SPIE (Medical Imaging 2011: Physics of Medical Imaging), Lake Buena Vista, Florida, USA, 12 - 17 Feb 2011, vol. 7961, pp. 79612Q, 2011
- Schwemmer, Chris; Prümmer, Marcus; Daum, Volker; Hornegger, Joachim, High-Density Object Removal from Projection Images using Low-Frequency-Based Object Masking Bildverarbeitung für die Medizin 2010 - Algorithmen - Systeme - Anwendungen (Bildverarbeitung für die Medizin 2010), Aachen, 14.-16.3.2010, pp. 365-369, 2010, ISBN 978-3-642-11967-5
- Paulus, Jan; Bock, Rüdiger; Daum, Volker; Hornegger, Joachim. Non-Rigid Registration to Capture Optic Nerve Head Variability In: Bildverarbeitung für die Medizin, Heidelberg, 2010, pp. 221-225
- Hahn, Dieter; Daum, Volker; Hornegger, Joachim; Kuwert, Torsten. Data-Driven Density Estimation applied to SPECT Subtraction Imaging for Epilepsy Diagnosis. In: Proceedings of the MICCAI Workshop on Probabilistic Models For Medical Image Analysis - MICCAI 2009, London, UK, 2009, pp. 115-126
- Daum, Volker; Hahn, Dieter; Hornegger, Joachim; Kuwert, Torsten. PCA Regularized Nonrigid Registration for PET/MRI Attenuation Correction. In: Proceedings of the MICCAI Workshop on Probabilistic Models For Medical Image Analysis - MICCAI 2009, London, UK, 2009, pp. 127-138
- Gaffling, Simone; Jäger, Florian; Daum, Volker; Tauchi, Miyuki; Lütjen-Drecoll, Elke. Interpolation of Histological Slices by Means of Non-rigid Registration In: Bildverarbeitung für die Medizin, Berlin, 2009, pp. 267-271
- Prümmer, Marcus; Hornegger, Joachim. 3D-Reconstruction in the Presence of Motion. In: Proceedings of the 4th Russian-Bavarian Conference on Biomedical Engineering, Moscow, Russia, 2008, pp. 9-13
- Daum, Volker; Hahn, Dieter; Hornegger, Joachim. A Nonlinear Projection Scheme for Fast Rigid Registration. In: IEEE Nuclear Science Symposium and Medical Imaging Conference Record, Honolulu, USA, 2007, pp. 4022-4026
- Hahn, Dieter; Daum, Volker; Hornegger, Joachim; Bautz, Werner; Kuwert, Torsten. Difference Imaging of Inter- and Intra-Ictal SPECT Images for the Localization of Seizure Onset in Epilepsy. In: IEEE Nuclear Science Symposium and Medical Imaging Conference Record, Honolulu, USA, 2007, pp. 4331-4335
- Hahn, Dieter; Daum, Volker; Hornegger, Joachim; Kuwert, Torsten. Comparison of Differences between Intra- and Inter-Ictal SPECT Images with MRI using Registration Techniques. In: Deutsche Gesellschaft für Nuklearmedizin (DGN), Hannover, Germany, 2007, pp. A59
- Daum, Volker; Helbig, Holger; Janka, Rolf; Eckardt, Kai-Uwe; Zeltner, Raoul. Quantitative Measurement of Kidney and Cyst Sizes in Patients with Autosomal Dominant Polycystic Kidney Disease (ADPKD). In: 3rd Russian-Bavarian Conference on Biomedical Engineering, Erlangen, Germany, 2007, Vol. 1, pp. 111-115
- Prümmer, Marcus; Fahrig, R.; Wigström, L.; Boese, J.; Lauritsch, G.; Strobel, N.; Hornegger, Joachim. Cardiac C-arm CT: 4D non-model based heart motion estimation and its application. In: Proceedings of the SPIE on Medical Imaging, San Diego, USA, Vol. 6510, 2007, pp. 651015
- Prümmer, Marcus; Wigstroem, Lars; Fahrig, R.; Lauritsch, G.; Hornegger, Joachim. Cardiac C-Arm CT: SNR Enhancement by Combining Multiple Retrospectively Motion Corrected FDK-like Reconstructions. In: Bildverarbeitung für die Medizin 2007, München, Germany, 2007, pp. 222-226
- Hahn, Dieter; Wolz, Gabriele; Sun, Yiyong; Hornegger, Joachim; Sauer, Frank; Kuwert, Torsten; Xu, Chenyang. A Practical Salient Region Feature Based 3D Multi-Modality Registration Method for Medical Images. In: Proceedings of SPIE on Medical Imaging, San Diego, USA, Vol. 6144, 2006, pp. 870-879
- Hahn, Dieter; Wolz, Gabriele; Sun, Yiyong; Sauer, Frank; Hornegger, Joachim; Kuwert, Torsten; Xu, Chenyang. Utilizing Salient Region Features for 3D Multi-Modality Medical Image Registration. In: Bildverarbeitung für die Medizin, Hamburg, Germany, 2006, pp. 221-225
- Prümmer, Marcus; Wigstroem, Lars; Hornegger, Joachim; Boese, Jan; Lauritsch, Guenter; Strobel, Norbert; Fahrig, Rebecca. Cardiac C-arm CT: Efficient Motion Correction for 4D-FBP. In: IEEE Medical Imaging Conference, San Diego, USA, 2006, pp. 2620-2628
- Prümmer, Marcus; Hornegger, Joachim; Pfister, Marcus; Dörfler, Arnd. Multi-modal 2D-3D non-rigid registration. In: Proceedings of SPIE on Medical Imaging, San Diego, USA, Vol. 6144, 2006, pp. 297-308
- Hahn, Dieter; Hornegger, Joachim; Bautz, Werner; Kuwert, Torsten; Römer, Wolfgang. Unbiased Rigid Registration Using Transfer Functions. In: Proceedings of SPIE on Medical Imaging, San Diego, USA, 2005, pp. 151-162
- Hornegger, Joachim; Hahn, Dieter; Bautz, Werner; Kuwert, Torsten; Römer, Wolfgang. Tumor Therapy Monitoring Based on Monomodal Image Fusion. In: Proceedings of the Radiological Society of North America (RSNA), Chicago, USA, 2005, pp. 355
- Prümmer, Marcus; Han, Jingfeng; Hornegger, Joachim. 2D-3D Non-rigid Registration using Iterative Reconstruction. In: Vision Modeling and Visualization, Erlangen, Germany, 2005, pp. 187-194
