Automated Data Analysis for Chiral Pooling with a YOLO Machine Learning Model

By converting chromatograms into machine-readable inputs, this approach enables fast, consistent identification of peaks, retention times and resolution, supporting more efficient selection of optimal chiral separation conditions.

The results: significant time savings, increased throughputs, and improved decision-making in chiral pooling workflows.

What the Poster Covers:

Automated analysis of chiral pooling data
Applying computer vision to detect chromatographic peaks across large screening datasets

YOLO machine learning model architecture
A 1D adaptation of object detection algorithms for chromatographic signal interpretation

High-throughput screening optimisation
Rapid ranking of separation conditions based on retention time and resolution

Performance and validation
Evaluation of model accuracy, false positive/negative rates, and detection limits

Scalable workflow integration
A continuously improving system using real data and analyst feedback