My group mines artificial intelligence methodologies for efficiency and robustness to allow eventual clinical applicability.
We focus on the analysis and integration of biomedical imaging and multisensory data to compute quantitative clinically relevant features.
Research Projects
Diagnostik und Therapieverlauf durch den Einsatz von KI-Methoden
KI-basierte multimodale Datenerhebung, insbesondere von Audio- und Bildmaterial
Optimierung und Generalisierung von neuronalen Netzen für die erfolgreiche Anwendung in der Klinik
Erstellung von Benchmark-Datensätzen für umfassende Reproduzierbarkeit
Neuronale Architekturoptimierung für die Anwendung auf Mikrocontrollern
Hardware- und Softwareseitige Implementierung von Hardwarebeschleunigern für die effiziente Datenverarbeitung
Architekturoptimierung für den Einsatz in der Medizin
Maximierung von Robustheit und Generalität neuronaler Netze
Balancierte neuronale Netze durch ganzheitliche Ressourcenoptimierung
Entwicklung von Software für effizientes überwachtes Lernen
Do stability of hippocampal neuronal excitatory synaptic connectivity and representations support spatial learning?
(Third Party Funds Single)
Project leader: Andreas Kist Term: 1. August 2025 - 31. July 2028 Funding source: DFG-Einzelförderung / Sachbeihilfe (EIN-SBH)
Hippocampal pyramidal neurons encode spatial locations through localized firing patterns, or place cells. Studies in humans and animals have shown a critical role for the hippocampus in spatial and episodic memory. Several theories have sought to explain how the hippocampus supports these functions and have considered place cells as basic substrates of long-term spatial memory. In many such frameworks, a central tenet is that long-term spatial memory arises from ensembles of cells that retain their spatial coding properties over time periods relevant to long-term memory. However, hippocampal spatial representations exhibit a surprising level of instability, or drift, which seems counterintuitive for a brain region important for memory formation. A few hypotheses exist on the function of this phenomenon, but the cellular and circuit mechanisms underlying hippocampal representational drift representations are completely unknown. By using two-photon time-lapse imaging to track simultaneously excitatory structural plasticity and activity patterns of the same pyramidal neurons over several days in the hippocampus of live mice, we want to shed light on the link between the stability connectivity and the drift of hippocampal representations. Moreover, we will study the relationship between structural plasticity, representation turnover, and learning by using a behavioral learning task, concomitant to imaging, and by perturbing the activity patterns of hippocampal dorsal CA1 pyramidal neurons by increasing local inhibition. In parallel, - and to increase the efficiency and reliability of our data analysis – we will develop a method based on machine and deep learning to consistently and automatically track dendritic spines in two-photon image time lapses and to match these to the activity of single pyramidal neurons in hippocampal dorsal CA1.
Neubig, L., Larsen, D., Ikuma, T., Kopp, M., Kunduk, M., & Kist, A. (2026). Markerless Tracking-Based Registration for Medical Image Motion Correction. In Lina Felsner, Thomas Küstner, Andreas Maier, Chen Qin, Seyed-Ahmad Ahmadi, Anees Kazi, Xiaoling Hu (Eds.), Lecture Notes in Computer Science (pp. 34-43). Daejeon, KOR: Springer Science and Business Media Deutschland GmbH.
2025
Arjomandi, J., Neubig, L., & Kist, A. (2025). LLM-driven Baselines for Medical Image Segmentation: A Systematic Analysis. In Christoph Palm, Katharina Breininger, Thomas Deserno, Heinz Handels, Andreas Maier, Klaus H. Maier-Hein, Thomas M. Tolxdorff (Eds.), Bildverarbeitung für die Medizin 2025. Proceedings, German Conference on Medical Image Computing, Regensburg March 09-11, 2025 (pp. 50-56). Regensburg, DE: Cham: Springer.
Lojo Rodríguez, M.B., Lopes Borges, G., López-Brea García, M., Schützenberger, A., & Kist, A. (2024). Neural Radiance Fields for 3D Reconstruction in Monoscopic Laryngeal Endoscopy. In 2024 IEEE International Symposium on Biomedical Imaging (ISBI). Athen, GR: Institute of Electrical and Electronics Engineers, Inc..
Neubig, L., & Kist, A. (2023). Dataset Pruning using Evolutionary Optimization. In Thomas M. Deserno, Heinz Handels, Andreas Maier, Klaus Maier-Hein, Christoph Palm, Thomas Tolxdorff (Eds.), Informatik aktuell (pp. 134-139). Braunschweig, DEU: Springer Science and Business Media Deutschland GmbH.
Neubig, L., & Kist, A. (2023). Evolutionary Normalization Optimization Boosts Semantic Segmentation Network Performance. In Hayit Greenspan, Hayit Greenspan, Anant Madabhushi, Parvin Mousavi, Septimiu Salcudean, James Duncan, Tanveer Syeda-Mahmood, Russell Taylor (Eds.), Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (pp. 703-712). Vancouver, BC, CAN: Springer Science and Business Media Deutschland GmbH.
Sun, Y., & Kist, A. (2023). Compact Convolutional Transformers on Edge TPUs. In Proceedings of the Bildverarbeitung für die Medizin Workshop, BVM 2023 (pp. 141-146). Springer Science and Business Media Deutschland GmbH.
Neubig, L., Groh, R., Kunduk, M., Larsen, D., Leonard, R., & Kist, A. (2022). Efficient Patient Orientation Detection in Videofluoroscopy Swallowing Studies. In Klaus Maier-Hein, Thomas M. Deserno, Heinz Handels, Andreas Maier, Christoph Palm, Thomas Tolxdorff (Eds.), Informatik aktuell (pp. 129-134). Heidelberg, DEU: Springer Science and Business Media Deutschland GmbH.
Schleiss, J., Hense, J., Kist, A., Schlingensiepen, J., & Stober, S. (2022). TEACHING AI COMPETENCIES IN ENGINEERING USING PROJECTS AND OPEN EDUCATIONAL RESOURCES. In Hannu-Matti Jarvinen, Santiago Silvestre, Ariadna Llorens, Balazs Vince Nagy (Eds.), SEFI 2022 - 50th Annual Conference of the European Society for Engineering Education, Proceedings (pp. 1592-1600). Barcelona, ES: European Society for Engineering Education (SEFI).
My group mines artificial intelligence methodologies for efficiency and robustness to allow eventual clinical applicability.
We focus on the analysis and integration of biomedical imaging and multisensory data to compute quantitative clinically relevant features.
Research Projects
Do stability of hippocampal neuronal excitatory synaptic connectivity and representations support spatial learning?
(Third Party Funds Single)
Term: 1. August 2025 - 31. July 2028
Funding source: DFG-Einzelförderung / Sachbeihilfe (EIN-SBH)
Hippocampal pyramidal neurons encode spatial locations through localized firing patterns, or place cells. Studies in humans and animals have shown a critical role for the hippocampus in spatial and episodic memory. Several theories have sought to explain how the hippocampus supports these functions and have considered place cells as basic substrates of long-term spatial memory. In many such frameworks, a central tenet is that long-term spatial memory arises from ensembles of cells that retain their spatial coding properties over time periods relevant to long-term memory. However, hippocampal spatial representations exhibit a surprising level of instability, or drift, which seems counterintuitive for a brain region important for memory formation. A few hypotheses exist on the function of this phenomenon, but the cellular and circuit mechanisms underlying hippocampal representational drift representations are completely unknown. By using two-photon time-lapse imaging to track simultaneously excitatory structural plasticity and activity patterns of the same pyramidal neurons over several days in the hippocampus of live mice, we want to shed light on the link between the stability connectivity and the drift of hippocampal representations. Moreover, we will study the relationship between structural plasticity, representation turnover, and learning by using a behavioral learning task, concomitant to imaging, and by perturbing the activity patterns of hippocampal dorsal CA1 pyramidal neurons by increasing local inhibition. In parallel, - and to increase the efficiency and reliability of our data analysis – we will develop a method based on machine and deep learning to consistently and automatically track dendritic spines in two-photon image time lapses and to match these to the activity of single pyramidal neurons in hippocampal dorsal CA1.
2026
2025
2024
2023
2022
2021
Related Research Fields
Contact: