Department of Physics, Jadavpur University, 188, Raja S.C. Mallick Rd, Kolkata, 700032, India.
Technical Research Centre, S. N. Bose National Centre for Basic Sciences, Block JD, Sector III, Salt Lake, Kolkata, West Bengal, 700106, India.
Sci Rep. 2023 Feb 9;13(1):2370. doi: 10.1038/s41598-023-29041-w.
The study was aimed to evaluate the performance of a newly developed spectroscopy-based non-invasive and noncontact device (SAMIRA) for the simultaneous measurement of hemoglobin, bilirubin and oxygen saturation as an alternative to the invasive biochemical method of blood sampling. The accuracy of the device was assessed in 4318 neonates having incidences of either anemia, jaundice, or hypoxia. Transcutaneous bilirubin, hemoglobin and blood saturation values were obtained by the newly developed instrument which was corroborated with the biochemical blood tests by expert clinicians. The instrument is trained using Artificial Neural Network Analysis to increase the acceptability of the data. The artificial intelligence incorporated within the instrument determines the disease condition of the neonate. The Pearson's correlation coefficient, r was found to be 0.987 for hemoglobin estimation and 0.988 for bilirubin and blood gas saturation respectively. The bias and the limits of agreement for the measurement of all the three parameters were within the clinically acceptance limit.
本研究旨在评估一种新开发的基于光谱学的无创、非接触式设备(SAMIRA)的性能,该设备可同时测量血红蛋白、胆红素和氧饱和度,作为有创血液采样的生化方法的替代方法。该设备的准确性在 4318 例患有贫血、黄疸或缺氧的新生儿中进行了评估。通过新开发的仪器获得经皮胆红素、血红蛋白和血液饱和度值,由专家临床医生通过生化血液测试进行验证。该仪器使用人工神经网络分析进行训练,以提高数据的可接受性。仪器中包含的人工智能可确定新生儿的疾病状况。血红蛋白估计的 Pearson 相关系数 r 为 0.987,胆红素和血气饱和度分别为 0.988。所有三个参数的测量偏倚和一致性界限均在临床可接受范围内。