Pregnancy detection based on blood serum sample Raman spectroscopy.

Abstract

In this research, women were diagnosed as pregnant based on blood serum samples Raman spectroscopy. Raman spectroscopy is a vibrational technique that provides information on the chemical composition of samples. The Raman techniques have significantly impacted the study of various degenerative diseases, particularly cancer detection, using less invasive methods such as the analysis of blood serum samples. Additionally, these techniques have been used to study the health status of patients, which is often difficult to monitor using conventional techniques. This study obtained serum samples from 12 women diagnosed as pregnant and 11 non-pregnant volunteers (controls). Spectra were collected using a LabRAM HR800 Raman Spectrometer (Horiba Jobin-Yvon) with an 830 nm wavelength laser. For each serum sample, 10 Raman spectra were obtained by focusing the laser, using a $100\times$ microscope objective, on different points of the sample with an exposure time of 40 seconds and an irradiation power of 17 mW. The raw spectra were processed using baseline correction and smoothing to remove noise, fluorescence, and shot noise. Subsequently, the spectra were normalized and analyzed using the multivariate statistical method of Principal Component Analysis (PCA). In these spectra, the characteristic bands of main blood serum biomolecules such as phenylalanine (Phe), tyrosine (Tyr), glutathione, amide III, phospholipid, carotene, and tryptophan (Trp) can be observed. Nevertheless, when analyzing the average spectra of pregnant and non-pregnant women, the main spectral differences were associated with variations in molecules like glutathione, tryptophan (Trp), tyrosine, and phenylalanine, which occur during the first trimesters of pregnancy. This aligns with findings previously reported by other studies. Furthermore, the serum samples from pregnant and non-pregnant patients can be effectively discriminated using PCA applied to the Raman spectra, revealing two distinct clusters in the PCA plot corresponding to each group's status. The results demonstrate that pregnancy can be determined based on blood serum samples Raman spectroscopy with $92.5\%$ sensitivity and $97.3\%$ specificity. Although there are very effective devices on the market to determine pregnancy based on the Human Chorionic Gonadotropin (HCG) hormone detection in urine samples, these preliminary results indicate an alternative method known as Raman spectroscopy. On the other hand, the results could also suggest the possibility of carrying out other gynecological or fetal-related studies in women using these Raman techniques.