Fei Zheng
Post-doctoral researcher at CREATIS-Lyon, France
✉ fei.zheng@creatis.insa-lyon.fr
Post-doctoral researcher at CREATIS-Lyon, France
✉ fei.zheng@creatis.insa-lyon.fr
I graduated from Xidian University in 2011, and received my master degree in electronics and telecommunications engineering at
Chinese National Laboratory of Radar Signal Processing
in 2014. I completed my PhD in computer science in the field of statistical signal processing at LIRIS, Ecole Centrale de Lyon, in 2017.
After thesis, I worked in team Statify, INRIA on anomaly detection algorithms for diabetes diagnosis and advanced artificial pancreas from 2018-2020. Currently, I'm a post-doctoral researcher at CREATIS-Lyon, working on magnetic resonance image data for myocardial infarct characterization.
Statistical learning for signal processing, medical image analysis,
dimension reduction, clustering, classification and anomaly detection.
Non-linear non-Gaussian system modeling, time-series filtering and prediction, Markov models.
Phased array radar signal processing.
Cardiac imaging plays a central role for myocardial infarct diagnosis, and patient follow-up after revascularization. Collaborating with clinical researchers from CHU Saint-Etienne, this project aims to retrospectively explore Magnetic Resonance Imaging (MRI) studies that include multi-parametric (LGE, T1, T2 etc) imaging of myocardial damages and regular follow-up, exploit computational atlases tools and statistical learning methods to better understand infarct evolution with therapy.
Most used artificial pancreas system implements an hybrid closed-loop requiring declaration of the time and amount of carbohydrates (CHO) intakes. Consequently, such systems are prone to CHO under estimations due to missing declarations and may lead to inadequate insulin dosing. This project is in collaboration with CEA-LETI and Diabeloop SA, aims to develop an unanounced meal detection and quantification method, thus to reduce such risk.
Glycemic variability (GV) is an important component of glycemic control for patients with type 1 diabetes. However, dozens of existing GV metrics views the variability from different aspects, and no consensus has been reached among physicians as to which metrics to use in practice. In collaboration with CEA-LETI and CHU Grenoble, this project aims to provide physicians a comprehensive index for daily GV evaluation and classification, which takes into account multiple aspects of the variability of glycemie.
Switching Hidden Markov models (SHMMs) are often used to approach non-Gaussian non-linear systems, and widely applied in the fields such as target tracking, seismic signal processing and finance. However, recursive filtering and smoothing are not feasible in SHMMs due to the unknown switches, and need to be approched by e.g. Markov Chain Monte-Carlo methods. This doctoral project aims to find an alternative restoration solution, introduce an original family of SHMMs defined with copulas which allows recursive restoration, and develop corresponding identification algorithms.
Anti-jamming modules are essential for the robustness of radar functions. adaptive side lobe cancellation (ASLC) is such a system that helps radar to suppress high duty cycle and noise-like interferences received through the side lobes of the radar. This master project of China Jinjiang Info Industrial Co.,LTD aims to design and realize an ASLC system for a phased array radar receiver, which enables its ability to againt interferences from up to three unknown spatial directions at the same time.
Sea clutter greatly impact the target detection and tracking performance of maritime radar. Grasping the characteristics of sea clutter has practical significance on reducing false detection alarms of the targets. This undergraduate project aims to analyze the statistical characteristics of the acquired real sea clutter data, and find out the best-fit distribution of the sea clutter.
F. Zheng, S. Derrode, W. Pieczynski.
Semi-supervised Optimal Recursive Filtering and Smoothing in Non-Gaussian Markov Switching Models.Signal Processing, vol. 171, Elsevier, 2020. link
F. Zheng, M. Jalbert, F. Forbes et al.
Characterization of Daily Glycemic Variability in Subject with Type 1 Diabetes Using Mixture of Metrics.Diabetes Technology & Therapeutics, 22(4): 301-313, 2019. link
F. Zheng, S. Derrode, W. Pieczynski.
Parameter Estimation in Switching Markov Systems and Unsupervised Smoothing.IEEE Transactions on Automatic Control, 64(4): 1761-1767, 2019. link
M. Jalbert, F. Zheng, A, Wojtusciszyn et al.
Glycemic Variability Indices Can Be Used to Diagnose Islet Transplantation Success in Type 1 Diabetic Patients.Acta Diabetologica, 57(3): 335-345, 2019. link
J. Dong, C. Wang, F. Zheng, F. Luo.
Effect of Channel Error on GPS Nulling Algorithm Performance.Radio Engineering of China 43(9): 24-27, 2013 (in Chinese). link
F. Zheng, S. Bonnet, E. Villeneuve et al.
Unannounced Meal Detection for Artificial Pancreas Systems Using Extended Isolation Forest.IEEE International Engineering in Medicine and Biology Conference (EMBC), Jul, 2020. link
F. Zheng, S. Bonnet, F. Forbes et al.
Caractérisation de la Variabilité Glycémique par Analyse Statistique Multivariée.27ème Colloque Francophone de Traitement du Signal et des Images (GRETSI), Lille, France, Aug, 2019. link
F. Zheng, M. Jalbert, F. Forbes et al.
Caractérisation de la Variabilité Glycémique Journalière chez le Patient avec Diabète de Type 1.Congrès de la Société Francophone du Diabète (SFD), Marsaille, France, Mar, 2019. link
F. Zheng, S. Derrode, W. Pieczynski.
Fast Exact Filtering in Generalized Conditionally Observed Markov Switching Models with Copulas.Traitement et Analyse de l'Information Méthodes et Applications (TAIMA), Hammamet, Tunisie, 2018. link
F. Zheng, S. Derrode, W. Pieczynski.
Parameter Estimation in Conditionally Gaussian Pairwise Markov Switching Models and Unsupervised Smoothing.IEEE 26th International Workshop on Machine Learning for Signal Processing (MLSP), Salerno, Italy, Sept, 2016. link
You can also contact me at my personal email address or on Linkedin
✉ feizheng0209@gmail.com