Breakthrough segmentation-free technique enhances plant root analysis using AI pose estimation
en-GBde-DEes-ESfr-FR

Breakthrough segmentation-free technique enhances plant root analysis using AI pose estimation

10/11/2024 TranSpread

The innovative approach, powered by SLEAP software, eliminates the limitations of traditional segmentation, enhancing phenotyping for crop improvement and climate resilience studies.

Plant roots are vital for water and nutrient uptake, anchorage, and carbon sequestration, playing a key role in stress resilience and climate change mitigation. Understanding root system architecture (RSA) is crucial for optimizing resource use and identifying traits for genetic enhancement. However, traditional root phenotyping methods, which rely on image segmentation, are labor-intensive and prone to errors that hinder large-scale analysis.

A study (DOI:10.34133/plantphenomics.0175 ) published in Plant Phenomics on 12 April 2024, facilitates large-scale phenotypic screening, accelerating crop breeding for stress tolerance and efficient resource use.

Researchers utilized pose estimation models to detect and localize root system landmarks across multiple plant species, bypassing traditional segmentation methods. They assessed model performance by measuring localization errors on 2D images across diverse datasets. The models achieved high precision, with median errors below 1% of root length. The rice primary root model at 3 days after germination (DAG) exhibited the lowest error at 0.087% (0.540 mm), while the Arabidopsis lateral root model at 7 DAG had the highest at 0.984% (0.843 mm), demonstrating strong reliability in morphological landmark detection. To evaluate the method's effectiveness for trait extraction, researchers compared pose estimation-derived phenotypic traits against manually annotated ground truth. High correlations were observed, with R² values ranging from 0.980 to 0.998 for key traits in rice and soybean, underscoring the method's accuracy. Despite minor discrepancies in root length measurements—primarily due to differences in node allocation—pose estimation consistently approximated root dimensions accurately. The study further showed that proofreading landmarks reduced trait errors, with 96.2% to 99.5% of traits falling within one standard deviation of manually corrected values. Compared to segmentation-based methods, pose estimation required significantly fewer annotations, reduced labeling time by 32.8%, and achieved up to a 90% reduction in training time. The extracted traits enabled robust genotype classification and phenotypic trait mapping, with key traits like convex hull area and lateral root angles driving variance. This landmark-based approach, validated across large datasets, offers a rapid and precise alternative for high-throughput plant phenotyping.

According to the study's lead researcher, Dr. Wolfgang Busch, “Our approach revolutionizes root phenotyping by leveraging deep learning-based landmark detectio. By eliminating segmentation, we’ve made root analysis more efficient and scalable, paving the way for advancements in sustainable agriculture.”

This segmentation-free AI approach represents a paradigm shift in plant phenotyping, offering a faster, more accurate, and scalable solution. By providing open access to their data and tools, the researchers hope to accelerate innovations in crop science and environmental research.

###

References

DOI

10.34133/plantphenomics.0175

Original Source URL

https://spj.science.org/doi/10.34133/plantphenomics.0175

Funding information

This work was supported by gifts from the Bezos Earth Fund, the Hess Corporation, through the TED Audacious Project, and the National Institutes of Health (1RF1MH132653).

About Plant Phenomics

Science Partner Journal Plant Phenomics is an online-only Open Access journal published in affiliation with the State Key Laboratory of Crop Genetics & Germplasm Enhancement, Nanjing Agricultural University (NAU) and distributed by the American Association for the Advancement of Science (AAAS). Like all partners participating in the Science Partner Journal program, Plant Phenomics is editorially independent from the Science family of journals. Editorial decisions and scientific activities pursued by the journal's Editorial Board are made independently, based on scientific merit and adhering to the highest standards for accurate and ethical promotion of science. These decisions and activities are in no way influenced by the financial support of NAU, NAU administration, or any other institutions and sponsors. The Editorial Board is solely responsible for all content published in the journal. To learn more about the Science Partner Journal program, visit the SPJ program homepage.

The mission of Plant Phenomics is to publish novel research that will advance all aspects of plant phenotyping from the cell to the plant population levels using innovative combinations of sensor systems and data analytics. Plant Phenomics aims also to connect phenomics to other science domains, such as genomics, genetics, physiology, molecular biology, bioinformatics, statistics, mathematics, and computer sciences. Plant Phenomics should thus contribute to advance plant sciences and agriculture/forestry/horticulture by addressing key scientific challenges in the area of plant phenomics.

Title of original paper: Fast and Efficient Root Phenotyping via Pose Estimation
Authors: Elizabeth M. Berrigan, Lin Wang, Hannah Carrillo, Kimberly Echegoyen,Mikayla Kappes, Jorge Torres, Angel Ai-Perreira, Erica McCoy, Emily Shane,Charles D. Copeland, Lauren Ragel, Charidimos Georgousakis,Sanghwa Lee, Dawn Reynolds, Avery Talgo, Juan Gonzalez, Ling Zhang,Ashish B. Rajurkar, Michel Ruiz, Erin Daniels, Liezl Maree, Shree Pariyar,Wolfgang Busch*, and Talmo D. Pereira*
Journal: Plant Phenomics
Original Source URL: 10.34133/plantphenomics.0175
DOI: https://spj.science.org/doi/10.34133/plantphenomics.0175
Latest article publication date: 12 April 2024
Subject of research: Not applicable
COI statement: The authors declare that they have no competing interests.
Attached files
  • Fig. 1. Overview of the high-throughput phenotyping pipeline using SLEAP.
  • Fig. 4. (A to H) Schematics of root trait computation methods from landmark data.
10/11/2024 TranSpread
Regions: North America, United States
Keywords: Applied science, Engineering

Disclaimer: AlphaGalileo is not responsible for the accuracy of news releases posted to AlphaGalileo by contributing institutions or for the use of any information through the AlphaGalileo system.

Testimonials

For well over a decade, in my capacity as a researcher, broadcaster, and producer, I have relied heavily on Alphagalileo.
All of my work trips have been planned around stories that I've found on this site.
The under embargo section allows us to plan ahead and the news releases enable us to find key experts.
Going through the tailored daily updates is the best way to start the day. It's such a critical service for me and many of my colleagues.
Koula Bouloukos, Senior manager, Editorial & Production Underknown
We have used AlphaGalileo since its foundation but frankly we need it more than ever now to ensure our research news is heard across Europe, Asia and North America. As one of the UK’s leading research universities we want to continue to work with other outstanding researchers in Europe. AlphaGalileo helps us to continue to bring our research story to them and the rest of the world.
Peter Dunn, Director of Press and Media Relations at the University of Warwick
AlphaGalileo has helped us more than double our reach at SciDev.Net. The service has enabled our journalists around the world to reach the mainstream media with articles about the impact of science on people in low- and middle-income countries, leading to big increases in the number of SciDev.Net articles that have been republished.
Ben Deighton, SciDevNet

We Work Closely With...


  • BBC
  • The Times
  • National Geographic
  • The University of Edinburgh
  • University of Cambridge
  • iesResearch
Copyright 2024 by AlphaGalileo Terms Of Use Privacy Statement