Related Experiment Video
Updated: Jan 12, 2026
07:15
Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
Published on: August 16, 2020
7.4K
Investigating the potential of combining Raman spectroscopy and machine learning for gestational age classification
Metin Tekin1, Cemaleddin Simsek2, Selim Buyukkurt3
1Graduate School of Natural and Applied Sciences, Department of Electrical and Electronics Engineering, Karamanoglu Mehmetbey University, Karaman 70200, Turkiye.
Spectrochimica Acta. Part A, Molecular and Biomolecular Spectroscopy
|November 2, 2025
Summary
Raman spectroscopy combined with artificial neural networks accurately classifies rat cervical tissue by gestational age. This non-invasive method shows promise for monitoring pregnancy and assessing spontaneous preterm birth risk.
Area of Science:
- Biochemistry
- Medical Spectroscopy
- Computational Biology
Background:
- Spontaneous preterm birth (SPB) is a major global cause of infant mortality.
- Current SPB detection methods have limitations, including high false-positive rates.
- Cervical biochemical changes precede labor, offering potential biomarkers.
Purpose of the Study:
- To investigate Raman spectroscopy (RS) and machine learning (ML) for gestational age classification in rat cervical tissue.
- To evaluate the efficacy of various ML algorithms in analyzing RS data for pregnancy monitoring.
- To establish a foundation for using RS in clinical SPB risk assessment.
Main Methods:
- Ex vivo Raman spectra were acquired from rat cervical tissue across multiple gestational days (15-20).
- A dataset of 600 spectra was analyzed using diverse ML algorithms: decision trees, random forests, SVM, K-NN, and ANN.
- Classification accuracy was determined for each ML model.
Main Results:
- The Artificial Neural Network (ANN) model achieved the highest classification accuracy at 87.5%.
- Other ML models demonstrated lower accuracies, ranging from 58% to 69%.
- Increased misclassification was observed around gestational day 19, indicating heightened cervical biochemical variability.
Conclusions:
- Integrating Raman spectroscopy with ML, particularly ANN, is a promising approach for classifying gestational stage via cervical biochemical composition.
- RS offers a non-invasive tool for monitoring pregnancy progression.
- This research supports the potential of RS for future clinical applications in early spontaneous preterm birth risk assessment.
