Title: Optoelectronic Methods and Devices for Point-of-Care Estimation of Hemoglobin and Hematocrit
Speaker: Ms, Neelam Verma . of Ph.D.(Engg) in Instrumentation and Applied Physics under Instrumentation and Applied Physics
Venue: SV Narasaiah Auditorium, IAP Department
Research Supervisor: Prof. Sai Siva Gorthi
Venue: SV Narasaiah Auditorium, IAP Department
Research Supervisor: Prof. Sai Siva Gorthi
Abstract:
Anemia is a highly prevalent global health condition, affecting approximately 35% of the global population of women and children In India the problem is even more severe according to the National Family Health Survey (NFHS-5, 2019–21), two out of three children aged 6–59 months (67.1%) and more than half of all women aged 15–49 years (57%) are anemic, including 52.2% of pregnant women. Anemia is diagnosed when the hemoglobin (Hb) concentration in blood falls below a reference cut-off that differs by age, sex, and physiological state, such as pregnancy. In young children, untreated anemia impairs cognitive and motor development, whereas in pregnant women it increases the risk of preterm birth, low birth weight, and perinatal death. Timely screening with an accurate hemoglobin measurement allows anemia to be detected before it becomes severe. Hb is conventionally measured with hematology analyzers or laboratory reference methods. However, the bulkiness of such instruments and the requirement for trained personnel limit their use in low-resource and field settings. To address this gap, point-of-care hemoglobinometers have emerged as a practical alternative owing to their portability, ease of use, and ability to deliver results within minutes. Colorimetric optical hemoglobinometers are widely used for anemia screening. They rely either on reagent-based microcuvettes or strips for Hb estimation. These hemoglobinometers have several limitations, including reagent stability issues, lipid and bilirubin interferences, and overestimated Hb in cold weather, all of which compromise their reliability in detecting anemia severity accurately within the CLIA acceptance limit. Apart from Hb Hematocrit (Hct), the fraction of blood volume occupied by red cells is usually read alongside Hb when assessing anemia or managing transfusion and fluid therapy. Most point-of-care devices estimate Hb and Hct separately, using different principles for each. Few devices calculate one parameter from another using a fixed mathematical correlation, such as Hct being 3 times Hb. The constant used in Hct calculation depends on mean corpuscular hemoglobin concentration (MCHC) which is under normal range. Hence, it fails when MCHC values are abnormal and outside the physiological range. Thus, this gap of estimating Hct and Hb accurately, simultaneously and independently is addressed in this thesis. The first part of this thesis presents the development of a handheld photometric device based on the sodium lauryl sulfate–hemoglobin (SLS-Hb) reaction for estimating Hb in point-of-care settings. The device had dual LEDs: a blue LED reads the strong Soret band of the complex, and a green LED reads the weaker Q band, and Hb is calculated from the difference between the two absorbances. The Soret channel is roughly three times as sensitive to Hb as the Q-band channel, while taking their difference removes background common to both wavelengths and tightens the calibration. 6 µL of venous blood is added to the SLS buffer at a 1:250 dilution to measure absorbance at each LED wavelength. The device was calibrated on 54 venous blood samples and validated on 173 independent samples. Developed device agreed with a reference hematology analyzer with negligible bias (+0.004 g/dL) and 95% limits of agreement of ±0.58 g/dL. Overall classification accuracy across four anemia classes was 92.5%, with a sensitivity of 95.7% and a specificity of 95.1% for detecting anemia. Because the SLS reagent lyses red cells, this method cannot measure Hct. Hence a new method is proposed which can estimate Hct as well as Hb from small volume of blood. Point-of-care devices for Hct either use conductometry, which is easily biased by plasma proteins and electrolytes, or use the blood centrifugation principle. We have developed a lysis-free spectrophotometric method that requires no centrifugation. Optical absorbance, in contrast, is specific to hemoglobin, but in intact blood, it is distorted by absorption flattening and scattering, which depend on Hct. This method uses this distortion to estimate Hct and Hb independently from 10 µL of whole blood, a finger-prick volume, diluted 1:151 in isotonic buffer. HCT is obtained by regularized inversion of a forward model that combines absorption flattening and an empirically calibrated scattering term. In contrast, Hb is independently estimated by multiple linear regression, not from Hct. Both Hb and Hct estimation use absorbance at wavelengths 416, 540, 560, 600, and 750 nm. Trained on 23 venous samples and tested on 117 independent samples against a hematology analyzer, the method gave a bias (95% limits of agreement) of −0.18 (−2.65 to 2.28) percentage points for Hct and −0.20 (−0.80 to 0.39) g dL⁻¹ for Hb. The Hb estimated by the regression model and the Hb calculated from Hct were compared independently. In 11 samples with MCHC below 315 g L⁻¹, Hb calculated from Hct with a fixed MCHC was biased by +0.77 g dL⁻¹, against +0.08 g dL⁻¹ estimated by the regression model. Pooled repeatab= ility coefficients of variation from 15 independently prepared duplicates were 1.2% for Hct and 1.4% for Hb. Needing only five wavelengths and no lysis, the method could be implemented in a compact LED-based reader for anemia screening, obstetric monitoring, and triage. In the final part of this thesis, a battery-powered handheld prototype was developed to investigate the translation of the lysis-free optical approach from a benchtop spectrophotometer to compact hardware. The prototype uses two LEDs and a laser diode arranged along two optical paths, each with a dedicated photodetector. A pilot comparison using clinical venous samples examined the optical response and calibration associated with replacing the spectrophotometer with discrete light sources. Further evaluation is required with capillary blood, intended users and larger clinical cohorts, including testing under field conditions. Collectively, the thesis demonstrates compact photometric Hb measurement, establishes a lysis-free method for estimating Hb and Hct from the same small blood sample, and provides a handheld prototype for further development and validation.
Anemia is a highly prevalent global health condition, affecting approximately 35% of the global population of women and children In India the problem is even more severe according to the National Family Health Survey (NFHS-5, 2019–21), two out of three children aged 6–59 months (67.1%) and more than half of all women aged 15–49 years (57%) are anemic, including 52.2% of pregnant women. Anemia is diagnosed when the hemoglobin (Hb) concentration in blood falls below a reference cut-off that differs by age, sex, and physiological state, such as pregnancy. In young children, untreated anemia impairs cognitive and motor development, whereas in pregnant women it increases the risk of preterm birth, low birth weight, and perinatal death. Timely screening with an accurate hemoglobin measurement allows anemia to be detected before it becomes severe. Hb is conventionally measured with hematology analyzers or laboratory reference methods. However, the bulkiness of such instruments and the requirement for trained personnel limit their use in low-resource and field settings. To address this gap, point-of-care hemoglobinometers have emerged as a practical alternative owing to their portability, ease of use, and ability to deliver results within minutes. Colorimetric optical hemoglobinometers are widely used for anemia screening. They rely either on reagent-based microcuvettes or strips for Hb estimation. These hemoglobinometers have several limitations, including reagent stability issues, lipid and bilirubin interferences, and overestimated Hb in cold weather, all of which compromise their reliability in detecting anemia severity accurately within the CLIA acceptance limit. Apart from Hb Hematocrit (Hct), the fraction of blood volume occupied by red cells is usually read alongside Hb when assessing anemia or managing transfusion and fluid therapy. Most point-of-care devices estimate Hb and Hct separately, using different principles for each. Few devices calculate one parameter from another using a fixed mathematical correlation, such as Hct being 3 times Hb. The constant used in Hct calculation depends on mean corpuscular hemoglobin concentration (MCHC) which is under normal range. Hence, it fails when MCHC values are abnormal and outside the physiological range. Thus, this gap of estimating Hct and Hb accurately, simultaneously and independently is addressed in this thesis. The first part of this thesis presents the development of a handheld photometric device based on the sodium lauryl sulfate–hemoglobin (SLS-Hb) reaction for estimating Hb in point-of-care settings. The device had dual LEDs: a blue LED reads the strong Soret band of the complex, and a green LED reads the weaker Q band, and Hb is calculated from the difference between the two absorbances. The Soret channel is roughly three times as sensitive to Hb as the Q-band channel, while taking their difference removes background common to both wavelengths and tightens the calibration. 6 µL of venous blood is added to the SLS buffer at a 1:250 dilution to measure absorbance at each LED wavelength. The device was calibrated on 54 venous blood samples and validated on 173 independent samples. Developed device agreed with a reference hematology analyzer with negligible bias (+0.004 g/dL) and 95% limits of agreement of ±0.58 g/dL. Overall classification accuracy across four anemia classes was 92.5%, with a sensitivity of 95.7% and a specificity of 95.1% for detecting anemia. Because the SLS reagent lyses red cells, this method cannot measure Hct. Hence a new method is proposed which can estimate Hct as well as Hb from small volume of blood. Point-of-care devices for Hct either use conductometry, which is easily biased by plasma proteins and electrolytes, or use the blood centrifugation principle. We have developed a lysis-free spectrophotometric method that requires no centrifugation. Optical absorbance, in contrast, is specific to hemoglobin, but in intact blood, it is distorted by absorption flattening and scattering, which depend on Hct. This method uses this distortion to estimate Hct and Hb independently from 10 µL of whole blood, a finger-prick volume, diluted 1:151 in isotonic buffer. HCT is obtained by regularized inversion of a forward model that combines absorption flattening and an empirically calibrated scattering term. In contrast, Hb is independently estimated by multiple linear regression, not from Hct. Both Hb and Hct estimation use absorbance at wavelengths 416, 540, 560, 600, and 750 nm. Trained on 23 venous samples and tested on 117 independent samples against a hematology analyzer, the method gave a bias (95% limits of agreement) of −0.18 (−2.65 to 2.28) percentage points for Hct and −0.20 (−0.80 to 0.39) g dL⁻¹ for Hb. The Hb estimated by the regression model and the Hb calculated from Hct were compared independently. In 11 samples with MCHC below 315 g L⁻¹, Hb calculated from Hct with a fixed MCHC was biased by +0.77 g dL⁻¹, against +0.08 g dL⁻¹ estimated by the regression model. Pooled repeatab= ility coefficients of variation from 15 independently prepared duplicates were 1.2% for Hct and 1.4% for Hb. Needing only five wavelengths and no lysis, the method could be implemented in a compact LED-based reader for anemia screening, obstetric monitoring, and triage. In the final part of this thesis, a battery-powered handheld prototype was developed to investigate the translation of the lysis-free optical approach from a benchtop spectrophotometer to compact hardware. The prototype uses two LEDs and a laser diode arranged along two optical paths, each with a dedicated photodetector. A pilot comparison using clinical venous samples examined the optical response and calibration associated with replacing the spectrophotometer with discrete light sources. Further evaluation is required with capillary blood, intended users and larger clinical cohorts, including testing under field conditions. Collectively, the thesis demonstrates compact photometric Hb measurement, establishes a lysis-free method for estimating Hb and Hct from the same small blood sample, and provides a handheld prototype for further development and validation.
Refreshments will be served after the colloquium.