MultipleXLab: A high-throughput portable live-imaging root phenotyping platform using deep learning and computer...

Area of Science:

• Plant Biology
• Agricultural Science
• Biotechnology

Background:

• Efficient crop selection for water and nutrient uptake requires detailed plant root architecture profiling.
• Current high-performance root imaging tools are often expensive, stationary, and lack non-destructive high-throughput capabilities.
• Extracting structural and morphological root features non-destructively at high-throughput remains a significant challenge.

Purpose of the Study:

• To develop a modular, mobile, and cost-effective system for high-throughput plant root phenotyping.
• To overcome the limitations of existing stationary and costly imaging solutions.
• To enable automated analysis of root development from germination onwards.

Main Methods:

• Development of the MultipleXLab system: a mobile setup with a conventional camera, multi-axis rotational stage, and integrated LED lighting.
• Utilizing custom-built 3D-printed plate holders for seed monitoring.
• Implementation of a deep learning-based image segmentation model for automated data analysis.

Main Results:

• The MultipleXLab system successfully monitored seed germination and root growth non-invasively and at high-throughput.
• The system demonstrated robust data acquisition for Arabidopsis mutants.
• Precise evaluation of germination index and hourly growth rates was achieved, differentiating between mutant lines.

Conclusions:

• MultipleXLab offers a flexible, user-friendly, and mobile alternative to expensive, stationary root phenotyping platforms.
• The platform facilitates numerous applications for plant biologists, the seed industry, crop scientists, and breeding companies.
• It provides a valuable tool for advancing research in crop resilience and development.