Prediction of Neonatal Respiratory Distress Biomarker Concentration by Application of Machine Learning to Mid-Infrared Spectra

Ahmed, Waseem; Veluthandath, Aneesh V.; Rowe, David J.; Madsen, Jan; Clark, Howard; Postle, Anthony D.; Wilkinson, James S.; Murugan, Ganapathy Senthil

doi:10.3390/s22051744

Cited by 16 publications

(20 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Up to now, prenatal assessment of fetal lung maturity to predict the risk of NRDS is very important to prevent the occurrence of NRDS [ 10 ]. At present, the clinical evaluation of fetal lung maturity (FLM) mainly depends on amniocentesis for amniotic fluid analysis, including determination of amniotic fluid lecithin/sphingomyelin ratio, detection of phosphatidylglycerol, and the optical density value (OD650) of 650 nm measured by foam test and spectrophotometer.…”

Section: Introductionmentioning

confidence: 99%

The Role and Clinical Value of Optimized Fetal Main Pulmonary Artery Doppler Parameters in the Diagnosis and Prognosis Monitoring of Neonatal Respiratory Distress Syndrome

Zhang

2022

Computational and Mathematical Methods in Medicine

View full text Add to dashboard Cite

Objective. A retrospective study was conducted to explore the role and clinical value of optimized fetal main pulmonary artery Doppler parameters in the diagnosis and prognosis monitoring of neonatal respiratory distress syndrome (NRDS). Methods. Pregnant women admitted to our hospital for delivery from March 2019 to June 2021 were included in the study. There were 372 singleton pregnant women who voluntarily underwent ultrasound and Doppler examinations with informed consent for the study. The AT/ET ratio of the middle segment of the main pulmonary artery, the middle part of the left (or right) pulmonary artery, and the proximal segment of the intrapulmonary artery was measured by ultrasonic Doppler. According to the outcome of the newborn, it was divided into NRDS group ( n =90) and non-NRDS group ( n =282) to compare the differences of fetal AT/ET ratio at three Doppler sampling locations and the differences of fetal AT/ET ratio in different gestational weeks. Using Pearson’s correlation analysis, the correlation among the Doppler parameters of fetal main pulmonary artery, the ratio of AT and AT/ET, gestational weeks, and the efficiency of prenatal prediction of NRDS was evaluated by receiver-operating characteristic (ROC). In addition, the diagnostic efficiency of fetal main pulmonary artery Doppler parameter AT/ET ratio and amniotic fluid method for prenatal prediction of NRDS was compared. Results. Compared with the non-NRDS group, the gestational age, gestational age at delivery, and birth weight in the NRDS group were significantly lower than those in the non-NRDS group. The 1 min Apgar score of newborns was less than 7, and the proportion of newborns transferred to NICU was significantly higher than that in the control group. Compared with non-NRDS group, AT was significantly shortened, AT/ET ratio was significantly decreased, and ET was significantly prolonged in NRDS group. The fetal aortic blood flow parameters AT, ET, and AT/ET ratio in different gestational weeks were significantly different among the three groups. The AT/ET ratio in the middle segment of the aorta in NRDS group was lower than that in non-NRDS group. The ratio of AT/ET was still correlated with the occurrence of postpartum NRDS. The AUC value of NRDS predicted by AT/ET ratio was 0.984. In the combined determination of amniotic fluid Lh and S ratio and PG, the diagnostic accuracy of prenatal prediction of neonatal respiratory distress syndrome was 100.005%, 98.43%, and 95.56%, respectively. The accuracy of fetal aortic AT/ET ratio in prenatal prediction of neonatal respiratory distress syndrome was 100.005%, 95.28%, and 95.57%, respectively. The order of Kappa coefficient from high to low was amniotic fluid PG/L/S ratio, fetal main pulmonary artery blood flow parameter AT/ET ratio, amniotic fluid L/S ratio, OD650, amniotic fluid PG, and foam test. Conclusion. Fetal main pulmonary artery Doppler AT/ET ratio may be a very potential parameter for prenatal noninvasive prediction of NRDS. The fetal main pulmonary artery Doppler flow spectrum may be a very potential alternative to amniocentesis to evaluate FLM.

show abstract

Section: Introductionmentioning

confidence: 99%

The Role and Clinical Value of Optimized Fetal Main Pulmonary Artery Doppler Parameters in the Diagnosis and Prognosis Monitoring of Neonatal Respiratory Distress Syndrome

Zhang

2022

Computational and Mathematical Methods in Medicine

View full text Add to dashboard Cite

show abstract

“…The reported values of the concentration of lipids in lung aspirates were 0.12-0.6 μmol/mL. 22 Because we used a centrifugation step in our sample preparation, it is difficult to predict the actual concentration of lipids in the measured sample. To establish that Raman spectroscopy can be useful on samples with physiologically relevant concentration, we have prepared samples with concentration 10, 1, 0.5, 0.1, and 0.01 μmol/mL 1 mL aqueous solution.…”

Section: Ratiometric Analysis Of Dppc/sm Liposomesmentioning

confidence: 97%

“…We chose PLSR because it addresses issues related to multicollinearity in spectral data and allows us to develop robust prediction models. 22 Raman spectra were baseline corrected using Spectrogryph software, 35 and the PLSR analysis was performed using the quant module of Operant LLC, Peak ® spectroscopy software. Figure 6A shows the work flow used in the machine learning methods.…”

Section: Plsr Of Raman Spectramentioning

confidence: 99%

See 1 more Smart Citation

Quantification of lung surfactant lipid (dipalmitoylphosphatidylcholine/sphingomyelin) ratio in binary liposomes using Raman spectroscopy

Veluthandath,

Ahmed,

Madsen

et al. 2023

J Raman Spectroscopy

Self Cite

View full text Add to dashboard Cite

Early diagnosis of neonatal respiratory distress syndrome (nRDS) is important in reducing the mortality of preterm babies. Knowledge of the ratio of two components of lung surfactant, dipalmitoylphosphatidylcholine (DPPC), and sphingomyelin (SM) can be used as biomarkers of lung maturity and inform treatment. Raman spectroscopy is a powerful tool to analyze vibrational spectra of organic molecules which requires only limited sample preparation steps and, unlike IR spectroscopy, is not masked by water absorption. In this paper, we explore the potential of using Raman spectroscopy as a tool to estimate the ratio of DPPC and SM from aqueous vesicles of binary mixture of DPPC and SM. We demonstrate that the ratio of DPPC and SM can be estimated by estimating the ratio of intensity of CO stretch of DPPC and CC stretch of SM as well as CO stretch of DPPC and amide I of SM. Further, we employ a partial least squares regression (PLSR) model to automate the estimation and demonstrate that PLSR method can predict the DPPC and SM ratio with an R2 value of 0.968.

show abstract

“…Analyzing IR and NIR spectra to obtain quantitative information about the substance is a fundamental topic in the field of chemometrics. − Various methods have been proposed for this purpose, mainly using linear regression (classification) models and dimensionality reduction techniques. − For example, Arakawa et al proposed a genetic algorithm-based wavelength selection (GAWLS) in combination with partial least squares regression (PLS). They applied the method to two different objectives: the soluble solid content of apples and the concentrations of moisture, nitrogen, and carbon in soil .…”

Section: Introductionmentioning

confidence: 99%

Chemometrics Approach Based on Wavelet Transforms for the Estimation of Monomer Concentrations from FTIR Spectra

et al. 2023

View full text Add to dashboard Cite

Fourier-transform infrared (FTIR) spectroscopy can detect the presence of functional groups and molecules directly from a mixed solution of organic molecules. Although it is quite useful to monitor chemical reactions, quantitative analysis of FTIR spectra becomes difficult when various peaks of different widths overlap. To overcome this difficulty, we propose a chemometrics approach to accurately predict the concentration of components in chemical reactions, yet interpretable by humans. The proposed method first decomposes a spectrum into peaks with various widths by the wavelet transform. Subsequently, a sparse linear regression model is built using the wavelet coefficients. Models by the method are interpretable using the regression coefficients shown on Gaussian distributions with various widths. The interpretation is expected to reveal the relation of broad regions in spectra to the model prediction. In this study, we conducted the prediction of monomer concentration in copolymerization reactions of five monomers against methyl methacrylate by various chemometric approaches including conventional methods. A rigorous validation scheme revealed that the proposed method overall showed better predictive ability than various linear and non-linear regression methods. The visualization results were consistent with the interpretation obtained by another chemometric approach and qualitative evaluation. The proposed method is found to be useful for calculating the concentrations of monomers in copolymerization reactions and for the interpretation of spectra.

show abstract

Prediction of Neonatal Respiratory Distress Biomarker Concentration by Application of Machine Learning to Mid-Infrared Spectra

Cited by 16 publications

References 44 publications

The Role and Clinical Value of Optimized Fetal Main Pulmonary Artery Doppler Parameters in the Diagnosis and Prognosis Monitoring of Neonatal Respiratory Distress Syndrome

The Role and Clinical Value of Optimized Fetal Main Pulmonary Artery Doppler Parameters in the Diagnosis and Prognosis Monitoring of Neonatal Respiratory Distress Syndrome

Quantification of lung surfactant lipid (dipalmitoylphosphatidylcholine/sphingomyelin) ratio in binary liposomes using Raman spectroscopy

Chemometrics Approach Based on Wavelet Transforms for the Estimation of Monomer Concentrations from FTIR Spectra

Contact Info

Product

Resources

About