Artificial intelligence for decoding human brain organisation
Understanding cognition through AI
Facing the challenges of brain complexity and big data analytics, research at INM-1 addresses human brain organization on its different scales, hand in hand with brain simulation and the investigation of learning in artificial and biological neural networks to advance our understanding of cognition, which opens perspectives to discover novel mechanisms for AI.
The research group at INM-1 is linked to the Big Data Analytics group headed by Timo Dickscheid, which develops methods for high throughput analysis and 3D reconstruction of microscopic image data, and neuroinformatics solutions for managing high-resolution brain atlases. The focus of this Helmholtz AI research group will be on deep learning methods for microstructural object detection and classification with limited training data, supported by shape and topological constraints. Driven by continuously increasing image resolutions and data volumes in high-throughput settings, methods are designed for distributed operation on HPC systems.
- Machine Learning and Computer Vision for biomedical image analysis
- High throughput imaging and HPC environments for microscopy
- Schiffer C, Spitzer H, Kiwitz K, Unger N, Wagstyl K, Evans AC, et al. Convolutional neural networks for cytoarchitectonic brain mapping at large scale. Neuroimage. 2021;240(118327):118327. DOI: 10.1016/j.neuroimage.2021.118327
- Schiffer C, Harmeling S, Amunts K, Dickscheid T. 2D histology meets 3D topology: Cytoarchitectonic brain mapping with graph neural networks. In: Medical Image Computing and Computer Assisted Intervention – MICCAI 2021. Cham: Springer International Publishing; 2021. p. 395–404. DOI: 10.1007/978-3-030-87237-3_38
- Kiwitz K, Schiffer C, Spitzer H, Dickscheid T, Amunts K. Deep learning networks reflect cytoarchitectonic features used in brain mapping. Sci Rep. 2020;10(1):22039. DOI: 10.1038/s41598-020-78638-y
- Dickscheid T, Haas S, Bludau S, Glock P, Huysegoms M, Amunts K. Towards 3D Reconstruction of Neuronal Cell Distributions from Histological Human Brain Sections. Advances in Parallel Computing. 2019;34(Future Trends of HPC in a Disruptive Scenario):223–239. DOI: 10.3233/APC190016
- Spitzer H, Kiwitz K, Amunts K, Harmeling S, Dickscheid T. Improving Cytoarchitectonic Segmentation of Human Brain Areas with Self-supervised Siamese Networks. In: Frangi AF, Schnabel JA, Davatzikos C, Alberola-López C, Fichtinger G, eds. Medical Image Computing and Computer Assisted Intervention – MICCAI 2018. Springer International Publishing; 2018:663–671.
- Dickscheid T, Förstner W. A Trainable Markov Random Field for Low-Level Image Feature Matching with Spatial Relationships.Photogrammetrie - Fernerkundung - Geoinformation. 2013;2013(4):269–283. DOI: 10.1127/1432-8364/2013/0176
- Amunts K, Lepage C, Borgeat L, et al. BigBrain: An Ultrahigh-Resolution 3D Human Brain Model. Science. 2013;340(6139):1472-1475. DOI: 10.1126/science.1235381
- Dickscheid T, Schindler F, Förstner W. Coding Images with Local Features. International Journal of Computer Vision. 2011;94(2):154–174. DOI: 10.1007/s11263-010-0340-z
- Development of an online accessible multimodal human brain atlas with cellular resolution
- AI-driven image registration for micrometer resolution biomedical images
- Deep Learning methods for brain mapping and instance segmentation of microstructural objects