ZiMT Journal Club December 2021: Prof. Dr.-Ing. Thomas Seel / Tracking the Unmeasurable and Achieving the Infeasible – Dynamic Inference and Learning in Plug-and-Play Sensor Networks, Continuous Sphygmomanometry and Intelligent Neuroprostheses

Symbolic picture for the article. The link opens the image in a large view.

Prof. Dr.-Ing. Thomas Seel, Lehrstuhl für Daten, Sensoren und Geräte, Department Artificial Intelligence in Biomedical Engineering (AIBE), FAU

Even if something cannot be measured directly, it might still be determined and tracked accurately by combining information that is hidden in the measured data with nonrestrictive prior knowledge. Likewise, a given task might be infeasible at first but become achievable by learning from repeated attempts. This talk is about novel methods for Sensor Fusion and Dynamic Learning and about the diagnosis and treatment solutions that are enabled by these methods. The rationale of each method is explained, and three example applications are discussed. (1) Algorithms for autonomous calibration and drift compensation are used to overcome long standing limitations of inertial motion tracking and thereby enable ubiquitous sensing and unsupervised use of plug-and-play IMU networks. (2) Noninvasive yet continuous measurements of the arterial blood pressure are accomplished by rapid adjustments of a dynamic cuff pressure pattern based on ultrasonic flow measurements. (3) Advanced learning control methods facilitate real-time pattern adaptation for artificial muscle recruitment after stroke or spinal cord injury, which has led to the design of intelligent neuroprostheses with biomimetic motor learning abilities.


D. Laidig, A. J. Jocham, B. Guggenberger, K. Adamer, M. Fischer, T. Seel. Calibration-Free Gait Assessment by Foot-Worn Inertial Sensors. Frontiers in Digital Health, 3 pages 1–21, 2021.

C. Wiesener, L. Spieker, J. Axelgaard, R. Horton, A. Niedeggen, N. Wenger, T. Seel, T. Schauer. Supporting Front Crawl Swimming in Paraplegics Using Electrical Stimulation: A Feasibility Study. Journal of NeuroEngineering and Rehabilitation, 17 (51):1–14, 2020.

D. Weber, C. Guehmann, T. Seel. RIANN–A Robust Neural Network Outperforms Attitude Estimation Filters. Artificial Intelligence, 2 pages 444–463, 2021.

A. Passon, T. Schauer, T. Seel. Inertial-Robotic Motion Tracking in End-Effector-Based Rehabilitation Robots. Frontiers in Robotics and AI, 7 pages 167, 2020.

F. Olsson, M. Kok, T. Seel, K. Halvorsen. Robust Plug-and-Play Joint Axis Estimation Using Inertial Sensors. Sensors, 20 (12):1–30, 2020.

T. Seel, C. Werner, T. Schauer. The Adaptive Drop Foot Stimulator – Multivariable Learning Control of Foot Pitch and Roll Motion in Paretic Gait. Medical Engineering Physics, 38 (11):1205–1213, 2016.