idw – Informationsdienst Wissenschaft
Nachrichten, Termine, Experten
Nachrichten, Termine, Experten
A research team from Helmholtz Munich, the Technical University of Munich (TUM) and the TUM University Hospital has developed DeepNeo, an AI-powered algorithm that automates the process of analyzing coronary stents after implantation. The tool matches medical expert accuracy while significantly reducing assessment time. With strong validation in both human and animal models, DeepNeo has the potential to standardize monitoring after stent implantation and thus improve cardiovascular treatment outcomes.
Challenges in Stent Healing Monitoring
Each year, more than three million people worldwide are treated with stents to open blocked blood vessels caused by heart disease. However, monitoring the healing process after implantation remains a challenge. If the tissue growing over the stent develops irregularly – either becoming too thick or forming deposits – it can lead to complications, such as re-narrowing or occlusion of the blood vessel. Currently, analyzing these healing patterns in intravascular optical coherence tomography (OCT) images is time-consuming and impracticable for routine clinical practice.
Automatic Assessment of Stent Healing
A research team from Helmholtz Munich and the TUM University Hospital has now developed DeepNeo, an artificial intelligence (AI) algorithm that can automatically assess stent healing in OCT images. DeepNeo differentiates between different healing patterns with an accuracy comparable to clinical experts – but in a fraction of the time. The AI tool also provides precise measurements, e.g. regarding tissue thickness and stent coverage, offering valuable insights for patient management.
“With DeepNeo, we can achieve an automated, standardized, and highly accurate analysis of stent and vascular healing that was previously only possible through extensive manual effort,” says Valentin Koch, first author of the study introducing the algorithm. “DeepNeo is as good as a doctor, but much faster.”
Validated with Strong Performance
To train DeepNeo, researchers used 1,148 OCT images from 92 patient scans, manually annotated to classify different types of tissue growth. They then tested the AI algorithm in an animal model, where it correctly identified unhealthy tissue in 87 percent of cases when compared to detailed laboratory analysis, the current gold standard. When analyzing human scans, DeepNeo also demonstrated high precision, closely matching expert assessments.
“DeepNeo demonstrates how machine learning can support clinicians in making quicker, more informed treatment decisions. The next step is now to effectively integrate AI algorithms like DeepNeo into clinical practice,” explains Dr. Carsten Marr, Director at the Institute of AI for Health at Helmholtz Munich. His colleague Prof. Julia Schnabel, who leads the Institute of Machine Learning in Biomedical Imaging and is Professor of Computational Imaging and AI in Medicine at TUM, envisions DeepNeo as part of an AI-powered healthcare system that could offer unprecedented certainty for clinical decision-making.
Towards Clinical Implementation
The project has received a Helmholtz Innovation Grant, and a patent application has been filed. Ascenion, technology transfer partner in the life sciences, is supporting the DeepNeo team in identifying potential industry partners. “DeepNeo facilitates and standardizes OCT imaging assessment after stent implantation and thus improves clinical decision-making,” say PD Dr. med. Philipp Nicol and Prof. Dr. med. Michael Joner, cardiologists at the TUM University Hospital, who led the project from the clinical side. “This has the potential to not only reduce healthcare costs but pave the way for more effective and personalized cardiovascular treatments.”
About the Researchers
Valentin Koch, Researcher at Helmholtz Munich and TUM until October 2024, now AI lead at the startup Floy.
Prof. Julia Schnabel, Director of the Institute of Machine Learning in Biomedical Imaging at Helmholtz Munich, and Professor of Computational Imaging and AI in Medicine at TUM.
Prof. Dr. med. Michael Joner, Deputy Director Department of Cardiology, TUM University Hospital German Heart Center.
Dr. Carsten Marr, Director of the Institute of AI for Health at Helmholtz Munich.
PD Dr. med. Philipp Nicol, Cardiologist at the TUM University Hospital German Heart Center.
About Helmholtz Munich
Helmholtz Munich is a leading biomedical research center. Its mission is to develop breakthrough solutions for better health in a rapidly changing world. Interdisciplinary research teams focus on environmentally triggered diseases, especially the therapy and prevention of diabetes, obesity, allergies, and chronic lung diseases. With the power of artificial intelligence and bioengineering, researchers accelerate the translation to patients. Helmholtz Munich has around 2,500 employees and is headquartered in Munich/Neuherberg. It is a member of the Helmholtz Association, with more than 43,000 employees and 18 research centers the largest scientific organization in Germany. More about Helmholtz Munich (Helmholtz Zentrum München Deutsches Forschungszentrum für Gesundheit und Umwelt GmbH): www.helmholtz-munich.de/en
Dr. Carsten Marr
E-Mail: carsten.marr@helmholtz-munich.de
Koch et al., year: 2025. Deep learning model DeepNeo predicts neointimal tissue characterization using optical coherence tomography. Nature Communications Medicine. DOI: 10.1038/s43856-025-00835-5
https://www.nature.com/articles/s43856-025-00835-5
The bottom panel shows DeepNeo’s classification: blue indicates non-analyzable regions, green marks ...
Valentin Koch
Helmholtz Munich / Valentin Koch
Philipp Nicol and Valentin Koch working with DeepNeo.
T. Vitadello
T. Vitadello
Criteria of this press release:
Journalists, Scientists and scholars
Biology, Medicine
transregional, national
Research results
English
The bottom panel shows DeepNeo’s classification: blue indicates non-analyzable regions, green marks ...
Valentin Koch
Helmholtz Munich / Valentin Koch
Philipp Nicol and Valentin Koch working with DeepNeo.
T. Vitadello
T. Vitadello
You can combine search terms with and, or and/or not, e.g. Philo not logy.
You can use brackets to separate combinations from each other, e.g. (Philo not logy) or (Psycho and logy).
Coherent groups of words will be located as complete phrases if you put them into quotation marks, e.g. “Federal Republic of Germany”.
You can also use the advanced search without entering search terms. It will then follow the criteria you have selected (e.g. country or subject area).
If you have not selected any criteria in a given category, the entire category will be searched (e.g. all subject areas or all countries).
