[Post-doc] Khac-Lan Nguyen

[Post-doc] Khac-Lan Nguyen: High-throughput phenotyping of fruit tree genetic diversity for better adaptation to climate change

Post-doc topic funded by #DigitAg

High-throughput phenotyping of fruit tree genetic diversity for better adaptation to climate change

KL Nguyen

I am a postdoctoral researcher in the EMMAH unit at the INRAE center in Avignon. My work involves developing advanced digital-based phenotyping techniques to better understand how the use of genetic diversity can help improve the resilience of fruit trees to climate change. 
As part of the PEPR "Agroecology and Digital" program, the teams from the EMMAH, GAFL, and AGAP research units are developing tools and methods based on digital technologies to characterize complex cropping systems. In addition to methodological developments enabling the conversion of stereovision and RGB (red, green, blue) data into usable vegetation traits, for example by ecophysiologists or geneticists, this project focuses on the fusion of different information sources, as well as on the generalization and transferability of methods and traits between species. Several approaches (machine learning, deep learning, mathematical morphology, statistics) will be combined. The project relies on the acquisition of data sets from core collections of three important fruit species (peach, apricot, apple), each containing over 150 different genotypes, ensuring a representation of variability in flowering and tree structure through differences in age, species, and contrasting environments. It also aims to establish temporal coherence between past measurements (visual notations) and digital phenotyping.
I am drawn to digital agriculture for its potential to revolutionize the agricultural sector through innovative technologies such as machine learning and deep learning, thereby optimizing efficiency and sustainability. My motivation stems from my passion for using technology to address real agricultural challenges and contribute to global food security. With a background in applied mathematics, I bring specific skills in deep learning, statistics, and image processing, along with a commitment to solving problems collaboratively. My goal is to further develop professionally by expanding my knowledge of emerging technologies and making significant contributions to the advancement of agricultural practices.

  • Starting date:  1st April 2024
  • Scientific fields: deep learning, applied mathematics, statistics, image processing
  • Co-Supervisor(s): Marie Weiss, UMR Emmah, Inrae et Evelyne Costes, UMR Agap, Inrae
  • #DigitAg:  Axe 2 : Innovations en agriculture numérique, Axe 3 : Capteurs, acquisition et gestion de données, Challenge 2: Le phénotypage rapide, Challenge 1 : Le challenge agroécologique, Challenge 3 : La protection des cultures

Keywords: Digital phenotyping, genetic screening, fruit trees, transferability between species

Abstract: This project aims to characterize the resilience of fruit trees through digital phenotyping of a number of traits related to flowering and tree architecture by combining the expertise in image analysis (AGAP-PHENOMEN, EMMAH-CAPTE), genetics (GAFLPrunus), and architectural analysis (AGAP-AFEF) from the four involved teams. In addition to methodological developments for processing stereovision and RGB data to access these traits, this project focuses on the fusion of information, as well as the generalization and transferability of methods and traits between species. Several approaches (machine and deep learning, mathematical morphology, statistics) will be combined. This project relies on the acquisition of data sets in core collections of three important fruit species (peaches, apricots, apples), each containing more than 150 different genotypes, thus ensuring representativeness of the variability in flowering and tree structure through differences in age, species, and contrasting environments. It also aims to establish temporal consistency between past measurements (visual notations) and digital phenotyping. From the perspective of expected outcomes, we aim to contribute to the phenotyping of complex and integrative traits (improving accuracy and throughput, accessing new traits not manually accessible), and to the genetic screening of resilience by determining a typology of trees that can maintain production in the face of environmental disturbances.

Contact: khac-lan.nguyen [AT] inrae.fr
Social network: LinkedIn

See also

Communications / Papers:

Journal articles 

  • A. Almhdie-Imjabbar, K. L. Nguyen, H. Toumi, R. Jennane, and E. Lespessailles. "Prediction of knee osteoarthritis progression using radiological descriptors obtained from bone texture analysis and Siamese neural networks : data from OAI and MOST cohorts." Arthritis Research & Therapy, 24(1):66, 2022. (NB: A. Almhdie-Imjabbar and K. L. Nguyen are both listed as first authors, see the section below the abstract on the first page of the article †Ahmad Almhdie-Imjabbar and Khac-Lan Nguyen contributed equally to this paper as first authors) (Open access)
  • K. L. Nguyen, P. Delachartre, and M. Berthier. "Multi-Grid Phase Field Skin Tumor Segmentation in 3D Ultrasound Images." IEEE Transactions on Image Processing, (99):1-1, 2019.
  • K. L. Nguyen, Mohamed M. Tekitek, P. Delachartre, and Michel Berthier. "Multiple Relaxation Time Lattice Boltzmann Models for Multigrid Phase-Field Segmentation of Tumors in 3D Ultrasound Images." SIA.

    National conferences 
  • K. L. Nguyen, A. Almhdie-Imjabbar, H. Toumi, R. Jennane, and E. Lespessailles. "Combinaison de la texture trabéculaire osseuse et des réseaux de neurones convolutifs pour la prédiction de la progression de la gonarthrose." Revue du Rhumatisme, (87) A90, 2020.
  • K. L. Nguyen, B. Sciolla, P. Delachartre, and M. Berthier. "Phase field segmentation of high frequency 3D ultrasound images using log-likelihood." In XXVI colloque GRETSI, 2017.

    International conferences 
  • K. L. Nguyen, B. Jamet, C. Bailly, C. Bodet-Milin, F. Kraeber-Bodéré, P. Moreau, C. Touzeau, T. Carlier, D. Mateus. "A multi-task learning approach for prediction of treatment response in multiple myeloma." IEEE Nuclear Science Symposium and Medical Imaging Conference, submitted in May 2022.
  • K. L. Nguyen, A. Xavier, A. Almhdie-Imjabbar, H. Toumi, R. Jennane, and E. Lespessailles. "Interest of texture analysis and neural networks for the characterization of knee osteoarthritis radiographic progression in OAI and MOST cohorts." Bone Reports, (13) 100696, 2020.