As pets become part of everyday family life, scientists are asking a new question: could the hair shed by cats and dogs help reveal the chemical pollutants circulating inside our homes?
A new study published in Artificial Intelligence & Environment suggests the answer is yes. Researchers developed an intelligent non-target screening framework that combines text mining, machine learning, and high-resolution mass spectrometry to map the “chemical space” of contaminants found in pet hair and indoor dust. The study shows that pet hair may serve as a complementary exposure matrix, helping scientists better understand the complex mixtures of pollutants present in human-pet shared indoor environments.
“Modern households are not simple indoor spaces. They are dynamic chemical environments shaped by building materials, consumer products, human activities, and increasingly, companion animals,” said corresponding author Fei Cheng. “Our study shows that pet hair can capture many of the same contaminant signals found in indoor dust, offering a new way to understand exposure in homes where people and pets live closely together.”
The team began by mining 16,692 publications related to indoor air pollution. From this large body of literature, they identified 3,684 indoor-relevant chemicals and built a molecular ion library containing 2,661 compounds. They also used artificial neural networks and convolutional capsule neural networks to generate a diagnostic ion library containing 30 fragment ions. These two libraries were then applied to high-resolution mass spectrometry data from 14 pet hair samples and 10 indoor dust samples.
The results were striking. Compared with conventional target analysis, the scenario-specific libraries increased the number of annotated compounds in major chemical categories by as much as 28-fold and reduced analysis time by about 47-fold. This means the approach could make complex indoor chemical screening faster, broader, and more relevant to real-world exposure scenarios.
Non-target screening revealed that pet hair and indoor dust shared more than 50% of their molecular features. Both matrices showed similar chemical profiles, dominated by commercial and industrial chemicals, pharmaceuticals and personal care products, flame retardants, pesticides, food additives, and transformation products. These findings suggest that pet hair does not simply represent a separate contaminant reservoir. Instead, it reflects many of the same indoor chemical sources that shape dust contamination.
Using machine learning, the researchers further identified the molecular properties that influence contaminant accumulation. Key factors included hydrophobicity, ionization-related properties, aromatic atom count, molecular flexibility, and polarizability. In simpler terms, compounds that are moderately hydrophobic, weakly polar, poorly soluble in water, and structurally suited for binding to organic-rich surfaces are more likely to accumulate in both dust and pet hair.
This has important implications for exposure science. Indoor dust has long been used to assess household contamination, but pet hair may provide additional insight because it is a biological material that has prolonged contact with indoor air, dust, skin oils, and household surfaces. The authors emphasize that larger studies are needed because the current work pooled samples, which may mask household-to-household and pet-to-pet differences.
Still, the study demonstrates a powerful new framework for exploring emerging exposure scenarios.
“By combining artificial intelligence with high-resolution chemical analysis, we can move beyond predefined pollutant lists and discover a wider range of chemicals that may matter for indoor health,” Cheng said. “Pet hair offers a promising and practical matrix for future environmental monitoring, especially in homes where people and animals share the same living space every day.”
===
Journal reference:
He LW; Cheng F; Shi JW; et al. Exploring emerging exposure scenarios via intelligent non-target screening: Chemical space characterization of pet hair contaminants. AI Environ. 2026, 1(1): 40-51. DOI: 10.66178/aie-0026-0006
https://www.the-newpress.com/aie/article/doi/10.66178/aie-0026-0006
===
About the Journal:
Artificial Intelligence & Environment is an international multidisciplinary platform for communicating advances in fundamental and applied research on the intersection of environmental science and artificial intelligence (AI). It is dedicated to serving as an innovative, efficient and professional platform for researchers in the cross-discipline fields of earth and environmental sciences, big data science and AI around the world to deliver findings from this rapidly expanding field of science. It is a peer-reviewed, open-access journal that publishes critical review, original research, rapid communication, view-point, commentary and perspective papers.
Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.