EUVIMEDEUVIMED entdecken
Open Access

Automated measurement of hair density and projected apparent width via deep learning–based phototrichogram analysis

Sehun Pyo, Bomin Kim, Jaeeun Min, Byung Cheol Park, Sang-Il Choi iD

International Journal of Cosmetic Science · 2026-08 · 48 (4) · 852-863

DOI: 10.1111/ics.70132 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Cosmetic Science, online erschienen am 27.07.2026. Verfasst von Sehun Pyo und 4 weiteren Autor:innen. Beteiligte Einrichtung: Department of Artificial Intelligence-Based Convergence, Dankook University, Yongin-si, Gyeonggi-do, Republic of Korea. Hauptthemen laut MEDLINE: Hair, Deep Learning, Image Processing, Computer-Assisted. Die Arbeit nennt 35 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Objective

Phototrichogram (PTG) imaging enables quantitative assessment of hair density and projected apparent hair shaft width, but manual PTG analysis is labour‐intensive and can be sensitive to image quality and operator variability.

Methods

We present an automated PTG analysis pipeline that (i) performs hair instance segmentation to count individual hair shafts for density estimation and (ii) quantifies the projected apparent width of each segmented hair shaft using an image‐processing module coupled with principal component analysis (PCA) of segmented hair pixel distributions.

Results

Using 259 high‐resolution PTG images (6000 × 4000) from 92 patients with predominantly dark pigmented hair acquired in a dermatology clinic, the proposed segmentation model achieved an average precision of 0.914 and recall of 0.870 for hair detection. Method‐comparison analyses on the test set demonstrated a moderate‐to‐strong correlation between automated and clinician‐assessed density measurements ( r = 0.732, p < 0.001), with a mean bias of −12.08 hairs/cm 2 and 95% limits of agreement of [−53.30, 29.14]. For projected apparent hair shaft width, the proposed module yielded a mean error of 2.44 μm (4.74%) in average projected apparent hair width compared with clinician measurements.

Conclusion

These results suggest that deep learning–assisted PTG analysis can provide reproducible, time‐efficient quantification of key hair parameters for clinical and cosmetic research workflows, particularly for longitudinal monitoring under standardized imaging conditions.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

HairDeep LearningImage Processing, Computer-AssistedHumansPrincipal Component Analysis

Stichwörter der Autor:innen

hair densityhair lossinstance segmentationphototrichogramprincipal component analysisprojected apparent hair width

Autor:innen und Institutionen

  • Department of Artificial Intelligence-Based Convergence, Dankook University, Yongin-si, Gyeonggi-do, Republic of Korea.
  • Department of Computer Engineering, Dankook University, Yongin-si, Gyeonggi-do, Republic of Korea.
  • Department of Dermatology, College of Medicine, Dankook University, Cheonan-si, Chungcheongnam-do, Republic of Korea.

ORCID-Profile: Sang-Il Choi

Zitieren

APA

Pyo, S., Kim, B., Min, J., Park, B. C., & Choi, S. I. (2026). Automated measurement of hair density and projected apparent width via deep learning–based phototrichogram analysis. International Journal of Cosmetic Science, 48(4), 852-863. https://doi.org/10.1111/ics.70132

Vancouver

Pyo S, Kim B, Min J, Park BC, Choi SI. Automated measurement of hair density and projected apparent width via deep learning–based phototrichogram analysis. International Journal of Cosmetic Science. 2026;48(4):852-863. doi:10.1111/ics.70132

BibTeX
@article{Pyo2026,
  author = {Pyo, Sehun and Kim, Bomin and Min, Jaeeun and Park, Byung Cheol and Choi, Sang-Il},
  title = {Automated measurement of hair density and projected apparent width via deep learning–based phototrichogram analysis},
  journal = {International Journal of Cosmetic Science},
  year = {2026},
  volume = {48},
  number = {4},
  pages = {852--863},
  doi = {10.1111/ics.70132}
}
Kostenlos starten →