High-throughput robotic collection, imaging, and machine learning analysis of salt patterns: composition and concentration from dried droplet photos

2025/01/01 by Bruno C. Batista, S Amrutha, Jie Yan +2 · 2 voices · 4 citations
Engineering · #Advanced Chemical Sensor Technologies

paper · pdf · doi:10.1039/d4dd00333k

openalex publication_date 2025/01/01 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/13

Abstract

A robotic drop imager collected over 23 000 images of dried drop stains to train a multi-layer perceptron capable of identifying salt type and concentration with >90% accuracy.

Citations

• The chemical basis of morphogenesis

Cited by

• Democratizing self-driving labs: advances in low-cost 3D printing for laboratory automation
• Identification of 2D colloidal assemblies in images: a threshold processing method versus machine learning
• Water Hardness from Photos of Dried Drop Stains: Reporter Salts Enable Low-Concentration Detection
• Machine Learning-Enabled Image Analysis of Complex Chemical Mixtures: Synthetic Urine Droplets as a Test System

Discussions

• We can determine the composition and concentration of solutions from photos of dried drops! Our latest paper in @digital-discovery.rsc.org describes how RODI 🤖 collects a 23,500-image library and ML [bsky, 7 points, 1 comments]
• In our latest featured article, Batista, Steinbock et al. introduce their robotic drop imager for generating datasets of dried sample deposits, which are then used to create a predictive model for sal [bsky, 2 points, 0 comments]

Related

• Evaluation of GPT-4o and GPT-4o-mini's Vision Capabilities for Compositional Analysis from Dried Solution Drops
• Water Hardness from Photos of Dried Drop Stains: Reporter Salts Enable Low-Concentration Detection
• Machine Learning-Enabled Image Analysis of Complex Chemical Mixtures: Synthetic Urine Droplets as a Test System
• Machine Learning-Enabled Image Analysis of Complex Chemical Mixtures: Synthetic Urine Droplets as a Test System
• High-throughput robotic colourimetric titrations using computer vision