Evaluation of machine learning models for classification of fertilization status of African clawed frog (xenopus laevis) by use of synthetic images
Evaluation von Machine-Learning-Modellen zur Klassifikation des Befruchtungsstatus des afrikanischen Krallenfroschs (Xenopus laevis) anhand synthetischer Bilder
PLOS One · 11.11.2026 · 21 (11) · e0360384
DOI: 10.1371/journal.pone.0360384 ↗Auf einen Blick
Eine CNN-basierte YOLOv10b-Anwendung klassifizierte befruchtete und unbefruchtete X. laevis-Eier in synthetischen und experimentellen Bildern. Die Modelle erreichten dabei unterschiedliche Genauigkeiten.
Fragestellung
Wie gut klassifizieren drei Optimierer eines CNN-basierten YOLOv10b-Modells den Befruchtungsstatus von X. laevis anhand synthetischer Bilder?
Methode
Entwicklung und Test eines CNN-basierten Tools mit YOLOv10b; synthetische Datengenerierung für 1.194 Trainings- und Validierungsbilder; Test auf 120 synthetischen und 20 experimentellen Bildern.
Ergebnisse
Test mit synthetischen Daten: 87 % (Adam), 89 % (AdamW), 90 % (SGD). Median der Embryo-Klassifikationsgenauigkeit: 97 % (Adam), 95 % (AdamW), 96 % (SGD).
Schlussfolgerung der Autor:innen
Die Autor:innen sehen die schnelle Entwicklung von Machine-Learning-Tools zur Klassifikation der Befruchtung als machbar und nützlich für Laborabläufe.
Kernaussagen
- YOLOv10b klassifizierte befruchtete und unbefruchtete Eier in Bildern.
- Synthetische Daten reduzierten den Bedarf an manueller Annotation.
- Die Modellgenauigkeit variierte je nach Optimierer.
Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.
Kurz gefasst
Fachartikel in der Zeitschrift PLOS One, online erschienen am 11.11.2026. Verfasst von Gopal Narayan Srivastava und 9 weiteren Autor:innen. Beteiligte Einrichtung: Aquatic Germplasm and Genetic Resources Centers, Louisiana State University Agricultural Center, Louisiana, United States of America. Hauptthemen laut MEDLINE: Xenopus laevis, Fertilization, Machine Learning, Image Processing, Computer-Assisted. Die Arbeit nennt 41 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NIH HHS, National Institutes of Health, National Science Foundation.
Abstract (Originalsprache)
The African clawed frog ( Xenopus laevis ) has been a pivotal model organism providing valuable insights into cell biology, developmental biology, and human disease. Fast and accurate identification of fertilized embryos versus unfertilized X. laevis eggs is crucial for time sensitive experimental applications such as assessing the efficacy of sperm cryopreservation and artificial fertilization. Traditionally, this task has relied on manual annotation, a labor-intensive process. To address this challenge, we developed a convolutional neural network (CNN)-based tool using the YOLOv10b (“You only look once version 10 balanced accuracy”) architecture to detect and quantify fertilized and unfertilized eggs in images. Our tool leveraged synthetic data generation to automate the creation of 1,194 training and validation images and annotations, significantly reducing the need for manual annotation. Through testing on 120 synthetic and 20 experimental images, we evaluated the performance of three optimizer models: Adam, AdamW, and SGD. The accuracy when testing the synthetic data was 87% for the Adam model, 89% for the AdamW model, and 90% for the SGD. On the other hand, the median percentage embryo classification accuracy (ECA) values were 97% (Adam), 95% (AdamW), and 96% (SGD). Our results demonstrate the feasibility of rapid development of machine learning tools for classification of fertilization making it a valuable tool for improving the efficiency in laboratory operation for biomedical researchers and resource centers that use X. laevis as a model organism. The codes to image synthesizing, training, and testing along with the Graphical User Interface (GUI) were made publicly available on GitHub and Open Science Framework (OSF).
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- Aquatic Germplasm and Genetic Resources Centers, Louisiana State University Agricultural Center, Louisiana, United States of America.
- Department of Biological Sciences, Louisiana State University, Baton Rouge, Louisiana, United States of America.
- Department of Mathematics, Louisiana State University, Baton Rouge, Louisiana, United States of America.
- Department of Computer Science, Louisiana State University, Baton Rouge, Louisiana, United States of America.
- National Xenopus Resource (NXR), Marine Biological Laboratory, Massachusetts, United States of America.
- National Aquatic Germplasm Center (NAGC), Louisiana State University Agricultural Center, Louisiana, United States of America.
ORCID-Profile: Gopal Narayan Srivastava, Monika Pandey, Marko E Horb, Yue Liu
Förderung
- NIH HHS
- National Institutes of Health
- National Science Foundation
- U.S. Department of Agriculture
Zitieren
APA
Srivastava, G. N., Pandey, M., Bist, K., Chapagain, S., Anderson, C., Shaidani, N. I., Horb, M. E., Wolenski, P., Tiersch, T. R., & Liu, Y. (2026). Evaluation of machine learning models for classification of fertilization status of African clawed frog (xenopus laevis) by use of synthetic images. PLOS One, 21(11), e0360384. https://doi.org/10.1371/journal.pone.0360384
Vancouver
Srivastava GN, Pandey M, Bist K, Chapagain S, Anderson C, Shaidani NI, et al. Evaluation of machine learning models for classification of fertilization status of African clawed frog (xenopus laevis) by use of synthetic images. PLOS One. 2026;21(11):e0360384. doi:10.1371/journal.pone.0360384
BibTeX
@article{Srivastava2026,
author = {Srivastava, Gopal Narayan and Pandey, Monika and Bist, Kiran and Chapagain, Sandesh and Anderson, Carl and Shaidani, Nikko-Ideen and Horb, Marko E and Wolenski, Peter and Tiersch, Terrence R and Liu, Yue},
title = {Evaluation of machine learning models for classification of fertilization status of African clawed frog (xenopus laevis) by use of synthetic images},
journal = {PLOS One},
year = {2026},
volume = {21},
number = {11},
pages = {e0360384},
doi = {10.1371/journal.pone.0360384}
}
EUVIMED