EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Non-Invasive Jaundice Screening Using AI: Machine Learning Analysis of Sclera and Urine Images

Jungirl Seok iD, Jae Yeong Kim iD, Sang Gyune Kim iD, Young Seok Kim iD, Jeong-Ju Yoo iD

Journal of Clinical Medicine · 30.04.2025 · 14 (9) · 3125

DOI: 10.3390/jcm14093125 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Clinical Medicine, online erschienen am 30.04.2025. Verfasst von Jungirl Seok und 4 weiteren Autor:innen. Beteiligte Einrichtung: Department of Otorhinolaryngology-Head and Neck Surgery, Seoul National University Hospital, Seoul National University College of Medicine, Seoul 030…. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 25 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Soonchunhyang University Research Fund, National Cancer Center Grant, National Cancer Center Grant, Republic of Korea.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Jaundice commonly indicates liver dysfunction and is traditionally diagnosed via invasive blood tests.

Objective

This study aimed to develop an AI-based program utilizing images of sclera and urine for non-invasive jaundice screening and compared its accuracy to that of standard blood tests.

Methods

This retrospective study involved patients who underwent liver function and bilirubin tests. Scleral and urine images were collected and processed following a standardized protocol to ensure consistency; A4 paper was utilized for white balance correction. Various machine learning and deep learning algorithms, including Decision Tree, Random Forest, XGBoost, DeepSets, and ResNet, were applied to predict jaundice. A stratified five-fold cross-validation was employed, and jaundice was classified as present or absent based on the total bilirubin levels.

Results

In total, 57 patients with liver disease and 31 controls were included in the analysis. Various machine learning models were applied to analyze scleral and urine images. The DeepSets model exhibited the highest predictive performance, achieving an R2 of 0.782 in predicting bilirubin levels. For jaundice detection, the DeepSets model, using a threshold of 2.6 mg/dL, achieved an accuracy of 87.1% (AUC = 0.869), with a precision of 90.2% and a recall of 88.1%. In contrast, the Random Forest model, with a threshold of 2.9 mg/dL, achieved an accuracy of 84.2% (AUC = 0.833), with a precision of 86.0%, and a recall of 88.1%.

Conclusions

This study demonstrates that scleral images, when used with simple A4 white paper for color correction, have the potential to screen for jaundice and predict bilirubin levels. (Clinical Research Information Service of Republic of Korea, KCT0009915).

Schlagwörter

Stichwörter der Autor:innen

bilirubin level detectionmachine learningnon-invasive jaundice predictionsclera and urine image analysis

Autor:innen und Institutionen

  • Department of Otorhinolaryngology-Head and Neck Surgery, Seoul National University Hospital, Seoul National University College of Medicine, Seoul 03080, Republic of Korea.
  • Department of Otorhinolaryngology-Head and Neck Surgery, National Cancer Center, Goyang-si 10408, Republic of Korea.
  • Division of Gastroenterology and Hepatology, Department of Internal Medicine, Soonchunhyang University Bucheon Hospital, Bucheon 14584, Republic of Korea.

ORCID-Profile: Jungirl Seok, Jae Yeong Kim, Sang Gyune Kim, Young Seok Kim, Jeong-Ju Yoo

Förderung

  • Soonchunhyang University Research Fund
  • National Cancer Center Grant
  • National Cancer Center Grant, Republic of Korea

Zitieren

APA

Seok, J., Kim, J. Y., Kim, S. G., Kim, Y. S., & Yoo, J. J. (2025). Non-Invasive Jaundice Screening Using AI: Machine Learning Analysis of Sclera and Urine Images. Journal of Clinical Medicine, 14(9), 3125. https://doi.org/10.3390/jcm14093125

Vancouver

Seok J, Kim JY, Kim SG, Kim YS, Yoo JJ. Non-Invasive Jaundice Screening Using AI: Machine Learning Analysis of Sclera and Urine Images. Journal of Clinical Medicine. 2025;14(9):3125. doi:10.3390/jcm14093125

BibTeX
@article{Seok2025,
  author = {Seok, Jungirl and Kim, Jae Yeong and Kim, Sang Gyune and Kim, Young Seok and Yoo, Jeong-Ju},
  title = {Non-Invasive Jaundice Screening Using AI: Machine Learning Analysis of Sclera and Urine Images},
  journal = {Journal of Clinical Medicine},
  year = {2025},
  volume = {14},
  number = {9},
  pages = {3125},
  doi = {10.3390/jcm14093125}
}
Kostenlos starten →