FULL End-to-end multimodal 3D imaging and machine learning workflow for non-destructive diagnosis of vine trunk diseases

Worldwide, vineyards sustainability is threatened by grapevine ( Vitis vinifera L.) trunk diseases (GTD), which spread insidiously, irreversibly degrading internal trunk tissues and ultimately entailing vine death. Foliar symptoms can erratically appear, but the sanitary status of vines cannot be ascertained without injuring the plants. To tackle this challenge, we developed a novel approach based on multimodal 4D imaging and artificial intelligence (AI)-based image processing that allowed a non-invasive GTD diagnosis. Each imaging modality contribution to tissue discrimination was evaluated, and we identified quantitative structural and physiological markers characterizing wood degradation. The combined study of the anatomical distribution of degraded tissues and the foliar symptom history of plants collected in a vineyard in Champagne, France, demonstrated that white rot and intact tissue contents were key measurements. We finally proposed a model for an accurate GTD diagnosis. This work opens new routes for precision agriculture by permitting field monitoring of GTD and surveying plant health in situ .

[1]  S. Farine,et al.  Trunk Surgery as a Tool to Reduce Foliar Symptoms in Diseases of the Esca Complex and Its Influence on Vine Wood Microbiota , 2021, Journal of fungi.

[2]  R. Laubenbacher,et al.  Using digital twins in viral infection , 2021, Science.

[3]  Romain Fernandez,et al.  Fijiyama: a registration tool for 3D multimodal time-lapse imaging , 2020, Bioinform..

[4]  S. Ahola,et al.  Characterization of the decay process of Scots pine caused by Coniophora puteana using NMR and MRI , 2020, Holzforschung.

[5]  C. Windt,et al.  In vivo pressure gradient heterogeneity increases flow contribution of small diameter vessels in grapevine , 2019, Nature Communications.

[6]  J. Domec,et al.  Ecophysiological impacts of Esca, a devastating grapevine trunk disease, on Vitis vinifera L. , 2019, PloS one.

[7]  S. Jansen,et al.  Exploring the Hydraulic Failure Hypothesis of Esca Leaf Symptom Formation1[OPEN] , 2019, Plant Physiology.

[8]  L. Mugnai,et al.  Grapevine trunk disease in European and Mediterranean vineyards: occurrence, distribution and associated disease-affecting cultural factors , 2019 .

[9]  E. Abou-Mansour,et al.  Vitis Methods to Understand and Develop Strategies for Diagnosis and Sustainable Control of Grapevine Trunk Diseases. , 2019, Phytopathology.

[10]  A. McElrone,et al.  In vivo quantification of plant starch reserves at micrometer resolution using X-ray microCT imaging and machine learning. , 2018, The New phytologist.

[11]  C. Clément,et al.  Grapevine Trunk Diseases: A Review of Fifteen Years of Trials for Their Control with Chemicals and Biocontrol Agents. , 2017, Plant disease.

[12]  H. Sebastian Seung,et al.  Trainable Weka Segmentation: a machine learning tool for microscopy pixel classification , 2017, Bioinform..

[13]  Ian H. Witten,et al.  Data Mining, Fourth Edition: Practical Machine Learning Tools and Techniques , 2016 .

[14]  M. Pétrissans,et al.  Variations in the natural density of European oak wood affect thermal degradation during thermal modification , 2016, Annals of Forest Science.

[15]  Wanzhao Li,et al.  Relating MOE decrease and mass loss due to fungal decay in plywood and MDF using resonalyser and X-ray CT scanning , 2016 .

[16]  A. McElrone,et al.  Genes Expressed in Grapevine Leaves Reveal Latent Wood Infection by the Fungal Pathogen Neofusicoccum parvum , 2015, PloS one.

[17]  Jean-Luc Verdeil,et al.  Visualization of the 3D structure of the graft union of grapevine using X-ray tomography , 2012 .

[18]  Marc Fermaud,et al.  New Insights into Esca of Grapevine: The Development of Foliar Symptoms and Their Association with Xylem Discoloration. , 2012, Plant disease.

[19]  J. Vallance,et al.  Wood necrosis in esca-affected vines: types, relationships and possible links with foliar symptom expression , 2012 .

[20]  Matthieu Boone,et al.  Three-Dimensional X-Ray Imaging and Analysis of Fungi on and in Wood , 2009, Microscopy and Microanalysis.

[21]  J. Péros,et al.  Symptoms, wood lesions and fungi associated with esca in organic vineyards in Languedoc-Roussillon (France). , 2008 .

[22]  S. D. Marco,et al.  Wood Discoloration and Decay in Grapevines with Esca Proper and Their Relationship with Foliar Symptoms , 2007, Phytopathologia Mediterranea.

[23]  R. B. Pearce,et al.  Decay Development and its Restriction in Trees , 2000, Arboriculture & Urban Forestry.

[24]  L. Mugnai,et al.  Esca (Black Measles) and Brown Wood-Streaking: Two Old and Elusive Diseases of Grapevines. , 1999, Plant disease.

[25]  T. A. Carpenter,et al.  Non-invasive imaging of fungal colonization and host response in the living sapwood of sycamore (Acer pseudoplatanus L.) using nuclear magnetic resonance , 1994 .

[26]  Nancy Chinchor,et al.  MUC-4 evaluation metrics , 1992, MUC.

[27]  J. Wery,et al.  Current knowledge on Grapevine Trunk Diseases with complex etiology: a systemic approach , 2020, Phytopathologia Mediterranea.

[28]  A. McElrone,et al.  In vivo visualization of the final stages of xylem vessel refilling in grapevine (Vitis vinifera) stems. , 2018, The New phytologist.

[29]  D. Gramaje,et al.  Managing Grapevine Trunk Diseases With Respect to Etiology and Epidemiology: Current Strategies and Future Prospects. , 2018, Plant disease.

[30]  D. Gramaje,et al.  Grapevine trunk diseases. A review , 2016 .

[31]  E. Gelhaye,et al.  Density mapping of decaying wood using X-ray computed tomography , 2014 .

[32]  Henk Van As,et al.  MRI of plants and foods. , 2013, Journal of magnetic resonance.

[33]  W. Gubler,et al.  Grapevine trunk diseases , 2005 .