Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Review
. 2025 Jul 8:13:1584136.
doi: 10.3389/fpubh.2025.1584136. eCollection 2025.

Application of artificial intelligence in the analysis of asbestos fibers

Richard Lee  1 , Drew Van Orden  2 , Suzanne Blanda  1 , John Mihalick  1 , David Bickford  1 , Patrick Metsch  1
Affiliations
Review

Application of artificial intelligence in the analysis of asbestos fibers

Richard Lee et al. Front Public Health. .

Abstract

Automated asbestos fiber detection and identification has been the goal of asbestos microscopists for decades. The advent of inexpensive memory, fast digital processing, machine learning, and microscope automation provide the enabling platform for success. This paper will review recent developments in fiber detection and identification by PCM and SEM and will present recent progress in employing artificial intelligence in the TEM classification of asbestos and non-asbestos amphiboles in the evaluation of elongated minerals in raw materials. To date, this project has been self-funded.

Keywords: amphibole; artificial intelligence; asbestos; automation; identification.

PubMed Disclaimer

Conflict of interest statement

RL, SB, JM, DB, and PM were employed by the RJ Lee Group, Inc. The remaining author declares that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Figures

Figure 1
Figure 1
Comparison of images from two different samples as seen in the transmission electron microscope. The left image shows non-asbestos particles (black arrow) while the right shows asbestos particles (red arrow). These morphological classifications originated with an evaluation of the minerals in hand samples and comply with definitions found in accepted protocols (18, 19).
Figure 2
Figure 2
The morphology of each particle (shape, surface, sides, and ends) is examined and characterized using accepted terminology (27).
Figure 3
Figure 3
An example of data uploaded to a database showing the image, diffraction pattern, and elemental composition of each particle.
Figure 4
Figure 4
Image showing the AI system application of several discrimination procedures resulting in a consensus classification of the particle as “Non-Asbestos.” The EDS scan, TEM image, and diffraction pattern of the classified particle are shown on the left, while the results of different classification procedures are shown on the right.
Figure 5
Figure 5
An example of a decision tree (capped at two levels) showing the application of artificial intelligence to an example dataset.
Figure 6
Figure 6
An example of a decision tree (capped at two levels) showing the application of artificial intelligence to an example dataset.
See this image and copyright information in PMC

Similar articles

See all similar articles

References

    1. Lynch JR, Howard EA. Measurement of asbestos exposure. J Occup Med. (1968) 10:21–4. - PubMed
    1. Langer AM. Research perspectives concerning Asbestos minerals and their effects on biological systems. Environ Health Perspect. (1974) 9:335–8. doi: 10.1289/ehp.749335, PMID: - DOI - PMC - PubMed
    1. Lee R, Kelly J, Walker J. Automated methods for fiber measurement and identification. VDI-Berichte Nr. (1983) 475:71–89.
    1. Baron BA, Shulman SA. Evaluation of the Magiscan image analyzer for asbestos fiber counting. Am Ind Hyg Assoc J. (1987) 48:39–46. doi: 10.1080/15298668791384346 - DOI
    1. Inoue I, Kaga A, Yamaguchi K, Kamoi S. Development of an automatic system for counting asbestos fibers using image processing. Part Sci Technol. (1999) 16:263–79. doi: 10.1080/02726359808906799 - DOI

LinkOut - more resources