Research Associate
Dr.-Ing. Ee Heng Chen
Technical University of Munich (TUM)
Machine Vision and Perception Group (MVP)
Munich Institute of Robotics and Machine Intelligence (MIRMI)
German Heart Center Munich (DHM)
Work address 1
Machine Vision and Perception Group (MVP)
Parkring 13 (8101)
85748
Garching b. München
Work address 2
Munich Institute of Robotics and Machine Intelligence (MIRMI)
Georg-Brauchle-Ring 60_62(2941)/I
80992
München
Work address 3
German Heart Center Munich (DHM)
Lazarettstraße 36
80636
München
Biography
Dr.-Ing. Ee Heng Chen received his PhD degree in computer science from the Technical University of Munich in collaboration with BMW AG. During his PhD, he worked on various topics related to camera-based scene understanding and decision making for self-driving vehicles at the R&D department of BMW AG. Since 2021 he joined MIRMI and is currently working on the AI-Heart Center project in collaboration with the German Heart Center Munich.
Research Interests
- Modeling of decision making processes
- Object detection
- Semantic segmentation
- Optical flow estimation
- And any scene understanding approaches that enables the creation of better decision making systems !!!
Current Projects
Please do not hesitate to contact me if you are interested in any of the following projects !!!
Digitalization of the Intensive Care Unit at the German Heart Center Munich
This project aims to assist the daily work of clinicians by digitalizing the Intensive Care Unit at the German Heart Center Munich. The focus of this project is to collect, analyze and visualize data from various electronic medical devices in the ICU. Intelligent analysis and visualization of data will enable clinicians to have a better understanding of a patient's current health status. Currently, we are looking into the following 4 topics:
Topic #1 : Intelligent web-based data collection and visualization
We aim to collect real time data from medical devices and visualize them in an intutive and concise manner for clinicians. This enables clinicians to focus on just one screen instead of multiple device screens. Currently we have a prototype running using the django framework.
- Data Processing and Visualization Pipeline for Patient Data in Intensive Care Unit
- Visualization Framework for Patient Data in Intensive Care Unit
Topic #2 : Multivariate medical data analysis
We aim to analyse (real time) multivariate medical data from medical devices to identify early signs of clinical deterioration. The challenge here is to create a prediction framework that is able to handle time series data of varying frequency as well as sporadic aperiodic data.
Topic #3 : Skeleton-based action recognition in the ICU
We aim to identify the action of both the clinicians as well as the patients in the ICU. The recognized actions can not only be used to help document what clinicians has done, but can also be used as an alarm to identify patients that are in need of assistance.
- Robust Skeleton-based Action Recognition under Occlusion in Intensive Care Unit
- Robust Realtime Multiperson 3D-Skeleton Tracking under Occlusion in Intensive Care Unit
Topic #4 : Scene understanding in the ICU
We aim to understand the context of the scene in the ICU, which includes the interaction between clinicians and patients and thre relationship between objects in the scene. The goal here is to provide high level information about a patient's stay in the ICU both for better long-term analysis and identification of correlation between action and patient health.
2022
- Ee Heng Chen "Development of Vision-based Driving Strategies for Traffic Junction Scenarios" Doctoral dissertation, Technische Universität München, 2022. Link
- Kai Wu, Ee Heng Chen, Xing Hao, Felix Wirth, Keti Vitanova, Rüdiger Lange and Darius Burschka. "Adaptable Action-Aware Vital Modelsfor Personalized Intelligent Patient Monitoring" 2022 IEEE International Conference on Robotics and Automation (ICRA), pages 826-832. IEEE, 2022. Link
2021
- Ee Heng Chen, Joeran Zeisler, and Darius Burschka. "Direct Image Based Traffic Junction Crossing System for Autonomous Vehicles". In 2021 24th IEEE International Conference on Intelligent Transportation (ITSC), pages 334-340. IEEE, 2021. Link
- Ee Heng Chen, Manoj Vemparala, Nael Fasfous, Alexander Frickenstein, Ahmed Mzid, Naveen Shankar Nagaraja, Joeran Zeisler, Walter Stechele. "Investigating Binary Neural Networks for Traffic Sign Detection and Recognition". In 2021 32nd IEEE Intelligent Vehicles Symposium (IV), pages 1400-1405. IEEE, 2021. Link
- Ee Heng Chen, Jöran Zeisler, and Darius Burschka. "Estimating Dense Optical Flow of Objects for Autonomous Vehicles". In 2021 32nd IEEE Intelligent Vehicles Symposium (IV), pages 1393-1399. IEEE, 2021. Link
2020
- Ee Heng Chen, Hanbo Hu, Jöran Zeisler, and Darius Burschka. "Pixelwise Traffic Junction Segmentation for Urban Scene Understanding". In 2020 IEEE 23rd International Conference on Intelligent Transportation Systems (ITSC), pages 1–8. IEEE, 2020. Link
2019
- Ee Heng Chen, Philipp Röthig, Jöran Zeisler, and Darius Burschka. "Investigating Low Level Features in CNN for Traffic Sign Detection and Recognition". In 2019 IEEE Intelligent Transportation Systems Conference (ITSC), pages 325–332. IEEE, 2019. Link
2018
- Ee Heng Chen, and Darius Burschka. "Object-Centric Approach to Prediction and Labeling of Manipulation Tasks". In 2018 IEEE International Conference on Robotics and Automation (ICRA), pages 6931–6938. IEEE, 2018. Link