Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2021 Jan 26;12(2):1020-1035.
doi: 10.1364/BOE.412715. eCollection 2021 Feb 1.

Malignant and benign thyroid nodule differentiation through the analysis of blood plasma with terahertz spectroscopy

Affiliations

Malignant and benign thyroid nodule differentiation through the analysis of blood plasma with terahertz spectroscopy

Maria R Konnikova et al. Biomed Opt Express. .

Abstract

The liquid and lyophilized blood plasma of patients with benign or malignant thyroid nodules and healthy individuals were studied by terahertz (THz) time-domain spectroscopy and machine learning. The blood plasma samples from malignant nodule patients were shown to have higher absorption. The glucose concentration and miRNA-146b level were correlated with the sample's absorption at 1 THz. A two-stage ensemble algorithm was proposed for the THz spectra analysis. The first stage was based on the Support Vector Machine with a linear kernel to separate healthy and thyroid nodule participants. The second stage included additional data preprocessing by Ornstein-Uhlenbeck kernel Principal Component Analysis to separate benign and malignant thyroid nodule participants. Thus, the distinction of malignant and benign thyroid nodule patients through their lyophilized blood plasma analysis by terahertz time-domain spectroscopy and machine learning was demonstrated.

PubMed Disclaimer

Conflict of interest statement

The authors declare that there are no conflicts of interest.

Figures

Fig. 1.
Fig. 1.
The algorithm of the blood plasma sample measurements.
Fig. 2.
Fig. 2.
The two-stage scheme of the data separation.
Fig. 3.
Fig. 3.
Average absorption coefficient a) and refractive index b) spectra of water (1, blue), healthy individuals (2, green), and patients with thyroid nodules (3, red) with error bars. Averaging was carried out over 30 measurements for blood plasma of patients with thyroid nodules, over 15 measurements for both blood plasma healthy individuals and water.
Fig. 4.
Fig. 4.
Average difference spectra of absorption coefficient а) and refractive index b) of blood plasma healthy individuals (green squares) and patients with thyroid nodules (red squares) with error bars. Averaging was carried out over 30 measurements for blood plasma of patients with thyroid nodules and over 15 measurements for healthy individuals’ blood plasma (each measurement with an independent reference signal obtained by passing through a cell with water).
Fig. 5.
Fig. 5.
Averaged difference spectra of absorption coefficient and refractive index for the liquid plasma of two groups: malignant thyroid nodules (purple squares) and benign thyroid nodules (orange squares) normalized to water. Averaging was carried out over 15 measurements (each measurement with an independent reference signal obtained by passing through a cell with water).
Fig. 6.
Fig. 6.
The absorption coefficient of thick (a) and thin (b) pellets of lyophilized human blood plasma: green line - a group of healthy individuals, 1h-6h; orange line - group 1, 1b-5b; purple line - group 2, 1m-5m.
Fig. 7.
Fig. 7.
The absorption coefficient (a) and refractive index (b) with each sample averaging over thick and thin pellets: green line - a group of healthy individuals, 1h-6h; orange line - group 1, 1b-5b; purple line - group 2, 1m-5m.
Fig. 8.
Fig. 8.
The absorption coefficient of lyophilized blood plasma of two groups with malignant thyroid nodules (purple squares) and benign thyroid nodules (orange squares).
Fig. 9.
Fig. 9.
Dependence of the absorption coefficient of liquid a) and lyophilized b) blood plasma of two groups with malignant thyroid nodules (purple squares) and benign thyroid nodules (orange squares) at a frequency of 1 THz on the concentration of glucose in the samples.
Fig. 10.
Fig. 10.
Visualization of linear kernel PCA (a), multidimensional scaling (b), UMAP (c), and CMSE with linear PCA (d).
Fig. 11.
Fig. 11.
Linear separability of the healthy, benign, and malignant thyroid nodus data by the SVM with a linear kernel. Regularization parameter C=1 (a); Mean ROC curve for linear SVMs for the 3 splits (b).
Fig. 12.
Fig. 12.
Feature IDs sorted by relative features importance (a). Feature ID corresponds to a specific THz absorption frequency. The first most informative THz absorption frequencies (b).
Fig. 13.
Fig. 13.
Thin (a) and thick (b) benign and malignant thyroid nodules plasma pellets transformed by the OU kernel PCA. The numbers show the sample ID.

Similar articles

Cited by

References

    1. Kim D.-L., Song K.-H., Kim S. K., “High prevalence of carcinoma in ultrasonography-guided fine needle aspiration cytology of thyroid nodules,” Endocr. J. 55(1), 135–142 (2008).10.1507/endocrj.K07-120 - DOI - PubMed
    1. Brito J. P., Yarur A. J., Prokop L. J., McIver B., Murad M. H., Montori V. M., “Prevalence of thyroid cancer in multinodular goiter versus single nodule: a systematic review and meta-analysis,” Thyroid 23(4), 449–455 (2013).10.1089/thy.2012.0156 - DOI - PubMed
    1. Bin Saeedan M., Aljohani I. M., Khushaim A. O., Bukhari S. O., Elnaas S. T., “Thyroid computed tomography imaging: pictorial review of variable pathologies,” Insights Imaging 7(4), 601–617 (2016).10.1007/s13244-016-0506-5 - DOI - PMC - PubMed
    1. Kang T., Kim D. W., Lee Y. J., Cho Y. J., Jung S. J., Park H. K., Baek H. J., “Magnetic resonance imaging features of normal thyroid parenchyma and incidental diffuse thyroid disease: a single-center study,” Front. Endocrinol. 9, 746 (2018).10.3389/fendo.2018.00746. - DOI - PMC - PubMed
    1. Aghaghazvini L., Pirouzi P., Sharifian H., Yazdani N., Kooraki S., Ghadiri A., Assadi M., “3 T magnetic resonance spectroscopy as a powerful diagnostic modality for assessment of thyroid nodules,” Arch. Endocrinol. Metab. 62(5), 501–505 (2018).10.20945/2359-3997000000069 - DOI - PMC - PubMed

LinkOut - more resources