Hi! I am currently a post-doctorat at IMT Atlantique and Lab-STICC in Brest, France. 

Biography

I graduated from the Hanoi National University of Education (Hanoi, Vietnam) in June 2014 and then I got a three-year contract at the Institute of Mathematic, Vietnam  academy of science and technology (VAST) as a junior researcher in July 2014.

With the support of the Institute of Mathematics, VAST  I got a two-year scholarship from the International Centre for Mathematics and Computer Science in Toulouse (CIMI) to study Applied Mathematics Master at Paul Sabatier University in 2015-2017.

Then I got third-year PhD contract (2017-2020) with Continental Digital Services France and LAAS-CNRS, in SARA team, under supervision of Balakrishna Prabhu and Olivier Brun. My PhD  program is financed by Continental.

I am a co-founder of Torus Actions which is a company established in March 2019, I joined the company officially from 2020-2023 after finishing my PhD. The company's objective is to provide tools, algorithms and innovative systems for life, especially for healthcare. My work in the company is under advisor of Prof. Nguyen Tien Zung. 

Since July 2023, I started a new position as a post-doctorat at IMT Atlantique and Lab-STICC  in Brest, France, under supervision of Ronan Fablet.

Personal Awards

• 2015-2017: Master Scholarship, CIMI, Toulouse, France.

• 2014 and 2013:  Award for excellent mathematical students, National Program for the development of mathematics.

• 2013: Award for doing Mathematical research of HNUE for Undergraduate Students.

• 2011:  Award of ranking the 3rd of the National Mathematical Olympics for Undergraduate Students.

• 2010:  Top 1 of the entrance exam to Faculty of Mathematics of HNUE.

Team Awards

• 2022:  DermatologIA  selected one of the 5 winners of the Call for Expressions of Interest "Data, AI and Health Pathways" by the French Sector Strategy Committee for Industries and Health Technologies. It is an app that uses AI algorithms to detect all types of Skin Diseases using clinical/dermoscopic images. The app involves a multidisciplinary team, including dermatologists, production team, AI scientists, data engineerings, annotators, and business professionals, ... It was an honor for me as a part of the project, and I participated as leader of the AI scientists and data engineering team.

• 2019: won first runner up award ISIC competition 2019 to detect Skin Cancer using dermoscopic images. I participated as a team member, contributing to build AI classification models and ensembling the results.

• 2019: received best paper award of ASMTA 2019 conference for the paper "An algorithm for improved proportional fairness of vehicular users". I co-authored this paper, together with my two PhD supervisors.

• 2018: Achieved silver medal (Team “AI Toulouse”) on TGS Salt Identification Challenge 2018 on Kaggle. I participated as a team member, contributing to build AI segmentation models and ensembling the results. 

Research Interests

 My research interests span across different fields, primarily focusing on Computer Vision and its applications in Healthcare and Environmental studies. I have a keen interest in several aspects of Computer Vision, including image segmentation, classification, object detection and image reconstruction. 

• In my current role, I am working with time series image reconstruction through the lens of Machine Learning, with applications in Ocean color remote sensing.  I am studying about meta-learning, bi-level optimization, end-to-end data assimilation neural networks and diffusion score-based generative models applied to time series images. 

• Previously, during the time at our startup, my focus was on using Computer Vision for the detection and assessment of skin diseases. The aim was to solve real-world problems by developing optimized AI core models that would be useful for dermatologists, general physicians, practitioners, and patients. The balance between accuracy and resource utilization (GPU infrastructure and computing time) was a key area of study. By communicating with dermatologists and patients, a deep understanding of their needs was gained, which then guided the modeling and addressing of these needs using Machine Learning.

• My PhD and Master study was centered around Stochastic Modeling, Reinforcement Learning, Convex Optimization, and Machine Learning, particularly in the context of channel allocation for vehicular users in Wireless Networks. 

• During my Bachelor I was interested in Abstract Algebra, which led me to explore existing results through two small projects. These projects focused on Geometric Properties of Orbital Closures in Nilpotent Matrices Over Algebraically Closed Fields; and Hilbert Functions of Monomial Ideals in Commutative Algebra.

Publications

Preprints

1. [Co-Author among the Team]. Gap-Free GNSS-R Wind Field Reconstruction: A Neural Mapping Scheme and Initial Validation, 2025.

2. [Co-Author among the Team]. Towards Quasi-global Gap-free Ocean Wind Speed Mapping Enabled by Rapid-Sampling GNSS Reflectometry Satellites, 2025.

Articles

3. [Co-Author among the Team]. Physics-Informed Neural Data Assimilation for High-Resolution Coastal SPM Reconstruction from Model and Satellite data, Applied Ocean Research 2025.

4. [Co-Author among the Team].  Generalization performance of neural mapping schemes for the space-time interpolation of satellite-derived ocean colour datasets, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing (JSTARS), 2025. 

5. [Co-Author among the Team].  Observation-only learning of neural mapping schemes for gappy satellite-derived ocean colour parameters, IEEE Transactions on Geoscience and Remote Sensing, 2025. 

6. [Co-Author among the Team].  Score-based diffusion models for space-time interpolation of satellite-derived images: a sea surface turbidity case study, IEEE International Geoscience and Remote Sensing Symposium, 2025.[pdf]

7. [Co-Author among the Team].  Adaptive Spatial and Multi-Variable Generalization of 4DVarNet in ocean colour Remote Sensing, IEEE International Geoscience and Remote Sensing Symposium, 2024.[pdf]

8. T.T.N. Nguyen, O. Brun, B. Prabhu. A Learning-based Scheme for Channel Allocation to Vehicular Users in Wireless Networks, Performance of Evaluation, 2023.

9. T.T.N. Nguyen, O. Brun, B. Prabhu. Using channel predictions for improved proportional-fair utility for vehicular users, Computer Networks, 2022.

10. T.T.N. NGUYEN, B.P. Le. Topological voting method for image segmentation, Journal of Imaging, 2022.

11. T.T.N. Nguyen, O. Brun, B. Prabhu. Learning Resource Allocation Algorithms for Cellular Networks, MLN 2021.

12. T.T.N. Nguyen, O. Brun, B. Prabhu. Joint downlink power control and channel allocation based on a partial view of future channel conditions, 18th International Symposium on Modeling and Optimization in Mobile, Ad Hoc, and Wireless Networks, 2020.

13. T.T.N. Nguyen, O. Brun, B. Prabhu. An algorithm for improved proportional-fair utility for vehicular users, ASMTA 2019. 

14. T.T.N. Nguyen, U. Ayesta, B. Prabhu.Scheduling users in drive-thru Internet: a multi-armed bandit approach, Wiopt 2019.

Patents

1. [Co-Inventor Among the Team].  Compute system with Skin Disease Identification mechanism and method of operation thereof, U.S. Patent, U.S. Patent, Patent number US-12217422-B1, Publication date 2025-02-04.

2. [Co-Inventor Among the Team]. Compute system with Hidradenitis suppurativa severity diagnostic mechanism and method of operation thereof, U.S. Patent, Public number 20240194352, Publication date 2024-06-13.

3. [Co-Inventor Among the Team]. Compute system with Severity Diagnostic mechanism and method of operation thereof, U.S. Patent, Patent number US-20240130669-A1, Publication date 2024-04-25.

4.  [Co-Inventor Among the Team]. Compute system with Eczema Diagnostic mechanism and method of operation thereof, U.S. Patent, Patent number US-20240135533-A1, Publication date 2024-04-25.

5.  [Co-Inventor Among the Team].  Compute system with Psoriasis Diagnostic mechanism and method of operation thereof, U.S. Patent, Patent number US-11890105-B1, Publication date 2024-02-06.

6. (Patent Pending)  [Co-Inventor Among the Team].  Compute system with nail ailment diagnostic mechanism and method of operation thereof, U.S. Patent, (date filed: February, 2025)

Abstracts and Reports

1. [Co-Author among the Team].  Hybrid Modeling Approach for Enhancing Sea Surface Turbidity Mapping Using Neural Data Assimilation Networks: Applications in the Wadden Sea, abstract in Living Planet Symposium 2025.

2. [Co-Author among the Team].  Probabilistic Diffusion Models for Ocean Chlorophyll-a Prediction, abstract in EGU April 2025.

3. [Co-Author among the Team].  Score-based Diffusion Models for the Space-Time Interpolation of Sea Surface Turbidity, abstract in EGU April 2025.

4. [Co-Author among the Team].  High Resolution Reconstruction Of Suspended Particulate Matter Satellite Images Using 4DVarNet For The European Digital Twin Ocean: An Observing System Simulation Experiment (OSSE) In The Dutch Wadden Sea, abstract in AGU December 2024.

5. [Co-Author among the Team].  Spatial Generalization of 4DVarNet in ocean colour Remote Sensing, abstract in EGU April 2024.

6. T.T.N NGUYEN, et al. AI-aided automatic severity scoring system for Hidradenitis Suppurativa, abstract accepted for EHSF conference, 2023.

7. Tat Dat Tô, Dinh Thi Lan, Thi Thu Hang Nguyen, Thi Thuy Nga Nguyen, Hoang-Phuong Nguyen, Tien Zung Nguyen, Ensembled skin cancer classification (ISIC 2019 challenge submission)   hal-02335240

8. Q.H. Lu, T.T.H. Nguyen, T.T.N. Nguyen, N.T. Zung, T.D. To. Skin Lesion Analysis Towards Melanoma Detection for ISIC 2018.  hal-01847743

Talks & Poster & Activities

Below are some of my talks, presentations or posters that I have done recently:

• Participated as an AI Model Mentor (PyTorch) at the EDITO Model Lab Hackathon, Toulouse, France, 22–24 October 2025.

• Score-based diffusion models for space-time interpolation of satellite-derived images: a sea surface turbidity case study, IGARSS 2025, Brisbane, Australia, 3-8 August 2025.

• Hybrid Modeling Approach for Enhancing Sea Surface Turbidity Mapping Using Neural Data Assimilation Networks: Applications in the Wadden Sea, Living Planet Symposium, Vienna, Austria, 23-27 June 2025. 

• Score-based Diffusion Models for the Space-Time Interpolation of Sea Surface Turbidity, EGU 2025, Vienna, Austria, 27 April-02 May 2025.

• Score-based Generative Models for the Space-Time Interpolation of Sea Surface Turbidity: a case study in the Wadden Sea, Deepdive Seminar, IMT Atlantique, Brest, France, 26th March 2025.

• Operational deployment of Sea surface turbidity mapping with 4DVarNet for the Wadden Sea - a demonstration on EDITO platform, EDITO Model Lab AG, Mallorca, Spain, 19-21 March 2025.

• Bridging AI and Oceanography for Advanced Ocean Modeling, AI Action Summit, Polytechnique, Paris, 6-7 February 2025. 

• Improving Satellite Observation Gap Filling with End-to-end Data Assimilation Neural Network: Sea Surface Turbidity Case Studies, Workshop on Few-Shot Learning, IMT Atlantique, Plouzané, France, december 3rd 2024. 

• Enhancing Real Satellite Observation Mapping with End-to-end Data Assimilation Neural Network using Simulation Data, AI Wine Seminar, Torus AI, Toulouse, 14th November 2024. 

• Enhancing Satellite Observation Mapping with 4DVarNet Using SCHISM and Delft3D-FM Simulation Data: Sea Surface Turbidity Case Studies, Work Package 2 EDITO Workshop, Grenoble, France, 30th September - 2nd October 2024. 

• Adaptive Spatial and Multi-Variable Generalization of 4DVarNet in ocean colour Remote Sensing, IEEE International Geoscience and Remote Sensing Symposium, Athens, Greece, 7-12 July, 2024.

• Hybrid model approach with SCHISM model data, Institute of Coastal Systems-Analysis and Modelling Helmholtz-Zentrum Hereon,  Geesthacht, Germany, during my visit from 14-23 February 2024.

• Generalization performance of 4DVarNet for ocean color dataset, Emulation and Data Assimilation Workshop, WP2 EDITO Model Lab, Hamburg, Germany, 19-20 February 2024.

• Topological Voting Method for Image Segmentation, deepdive, IMT Atlantique, Plouzané,  France, 5th June 2024.

• A data-driven scheme for channel allocation to connected vehicles in wireless networks, deepdive, IMT Atlantique, Plouzané,  France, 27th March 2024.

• Hybrid model approach with Delft3D model data, Deltares, Delft, The Netherlands, during my visit from 13-24 November 2024.

• A Data-Driven Scheme for channel allocation to connected vehicles in Wireless Networks, ITC 35th conference - Turin, Italy 03- 05 October 2023.

• A journey on a Kaggle competition, AI Tasting Seminar, Toulouse, 20th April 2023.

• A Data-Driven Optimization Algorithm for channel allocation to vehicular users, AI Tasting Seminar, Torus AI, Toulouse, France, 23rd February 2023. 

• AI-aided automatic severity scoring system for Hidradenitis Suppurativa, EHSF conference, Florence, Italy, 9th February 2023. 

• Supervised learning for Skin Disease Detection,  a tutorial at Torus AI, Toulouse, France, January 2023.

• AI core for detecting all types of skin diseases: the data and AI algorithm, Torus AI team buiding, Andorra la Vella, January 2023.

Teaching

• TD Performance Evaluation for 4th-year student at INSA Toulouse.

• Course for 11th grade at Hung Vuong Gifted High School (Phu Tho, Vietnam) for 1 month in 2013.

• Course for 12th grade at Hung Vuong Gifted High School (Phu Tho, Vietnam) for 2 months in 2014.

Software

Python (Pytorch, Tensorflow, Pandas, Xarray, OpenCV, Numpy, Streamlit,...), Git, SUMO, Scilab