A Low-Cost Paper-Based Device for the Colorimetric Quantification of Bilirubin in Serum Using Smartphone Technology

Brittany AuYoung, Akshay Gutha Ravichandran, Divykumar Patel, Nisarg Dave, Achal Shah, Brianna Wronko-Stevens, Franklin Bettencourt, Reshma Rajan, Nidhi Menon, Brittany AuYoung, Akshay Gutha Ravichandran, Divykumar Patel, Nisarg Dave, Achal Shah, Brianna Wronko-Stevens, Franklin Bettencourt, Reshma Rajan, Nidhi Menon

Abstract

Total bilirubin values have been used as a potential marker to pre-screen and diagnose various liver-based diseases such as jaundice, bile obstruction, liver cancer, etc. A device known as KromaHealth Kit, composed of paper and an acrylic backbone, is developed to quantify total bilirubin in human serum using image processing and machine learning technology. The biochemical assays are deposited on absorbent paper pads that act as reaction zones when serum is added. A dedicated smartphone app captures images of the colorimetric changes on the pad and converts them into quantitative values of bilirubin. The range of bilirubin concentration that can be quantified using the device ranges from 0.5 mg/dl to 7.0 mg/dl. The precision, limit of detection, interference analysis, linearity, stability, and comparison with a predicate are studied in this paper in accordance with clinical and laboratory standards institute. The results indicate that the KromaHealth Kit can be used as an inexpensive alternative to conventional bilirubin testing in clinical settings. With its level of precision, ease-of-use, long shelf-life, and short turnaround time, it will prove to be invaluable in limited-resource settings.

Keywords: bilirubin; diagnostics; image processing; object detection; paper microfluidics.

Conflict of interest statement

Financial support for the development of this product was provided by Group K Diagnostics, an early-stage venture-capital-backed company. All authors are past or present employees of the company and used to or currently hold stock options. Author BW-S is the CEO and holds a position on the board of Group K Diagnostics. Patent applications filed pertaining to the technology (patent pending) are US Provisional Patent Appl. No. 63/140,035 and US Provisional Patent Appl. No. 63/277,796. KromaHealth Kit is currently under review with the FDA.

Copyright © 2022 AuYoung, Gutha Ravichandran, Patel, Dave, Shah, Wronko-Stevens, Bettencourt, Rajan and Menon.

Figures

FIGURE 1
FIGURE 1
Device and box design parameters. (A) The top and side views of the device, with three reaction panels consisting of the alpha cotton linter cellulose absorbent pads. (B) The top cover of the box has a pinhole for the phone camera, the inside of the box consists of an indent for the device to be placed for capturing the image, and the lighting within the box standardizes the lighting for the image process analysis.
FIGURE 2
FIGURE 2
Quantification of bilirubin using kromaHealth Kit. (A) The color gradient observed in the panel with increasing concentrations of Bilirubin as it reacts with the stabilized reagents on the pads. (B) The process followed to quantify these colorimetric changes start with the pipetting of serum on the pad. After 50 min of waiting for the reaction that produces the color change, the device is placed inside the box and the camera is aligned with the pinhole for the image to be captured and processed for quantification.
FIGURE 3
FIGURE 3
GKD iOS aplication workflow. The GKD application allows patients to be registered based on unique QR codes that are attached to an extension of the device. Once the patient is registered and the reaction process is complete, the application allows you to take a photo after scanning the QR code. The image is captured, and the selected ROI runs through the algorithm to predict a quantitative value for bilirubin. This is saved in the database and can be viewed again by scanning the patient QR code.
FIGURE 4
FIGURE 4
Relative frequency distribution. The relative frequency distribution for the limit of blank studies and limit of detection studies, as expected, shows minimal overlap in the concentration detected by the KromaHealth Kit.
FIGURE 5
FIGURE 5
Predicted 1st and 2nd Order Models. The predicted first and second order models had no non-linear components, and the plot shows the predicted concentration for the two models.
FIGURE 6
FIGURE 6
Performance compared to predicate device. Various distributions of the comparison studies can be seen. (A) The individual read-outs are plotted for each concentration tested in the predicate device and KromaHealth kit. Identical concentration readouts overlap when plotted and appear as darker points with increasing overlaps. The mean and standard error of the device is plotted in blue for each concentration tested. (B) Each point here shows the difference between a single read-out and its corresponding known concentration from the predicate device. (C) The skewed frequency distribution of the differences is seen here, and this is used to calculate the bias estimate.
FIGURE 7
FIGURE 7
Device stability over a 6-month period. The device performance was measured over a period of 6 months and found to be satisfactory, without significant deviation from the original known concentrations of the samples studied.

References

    1. Alawsi T., Mattia G. P., Al-Bawi Z., Beraldi R. (2021). Smartphone-based Colorimetric Sensor Application for Measuring Biochemical Material Concentration. Sens. Bio-Sensing Res. 32, 100404. 10.1016/j.sbsr.2021.100404
    1. Bradshi G. (2000). Open CV Library. Dr. Dobb’s J. Softw. Tools 120, 122–125.
    1. Budd J. R. (2018). EP09 Measurement Procedure Comparison And Bias Estimation Using Patient Samples. Pennsylvania, United States: Clinical and Laboratory.
    1. Chu S., Wang H., Du Y., Yang F., Yang L., Jiang C. (2020). Portable Smartphone Platform Integrated with a Nanoprobe-Based Fluorescent Paper Strip: Visual Monitoring of Glutathione in Human Serum for Health Prognosis. ACS Sustain. Chem. Eng. 8, 8175–8183. 10.1021/acssuschemeng.0c00690
    1. Clarke W. (2011). American Association for Clinical Chemistry Eds. Contemporary Practice in Clinical Chemistry . 2nd ed. Washington, DC: AACC Press, 652.
    1. Dahl R. (2009). Node.js. San Francisco: OpenJS Foundation.
    1. Dasgupta A., Wahed A. (2014). “Immunoassay Platform and Designs,” in Clinical Chemistry, Immunology and Laboratory Quality Control (Amsterdam, Netherlands: Elsevier; ), 19–34. 10.1016/B978-0-12-407821-5.00002-4
    1. Ellairaja S., Shenbagavalli K., Ponmariappan S., Vasantha V. S. (2017). A Green and Facile Approach for Synthesizing Imine to Develop Optical Biosensor for Wide Range Detection of Bilirubin in Human Biofluids. Biosens. Bioelectron. 91, 82–88. 10.1016/j.bios.2016.12.026
    1. Fossati P., Ponti M., Prencipe L., Tarenghi G. (1989). One-step Protocol for Assays of Total and Direct Bilirubin with Stable Combined Reagents. Clin. Chem. 35, 173–176. 10.1093/clinchem/35.1.173
    1. Garber C. C. (1981). Jendrassik--Grof Analysis for Total and Direct Bilirubin in Serum with a Centrifugal Analyzer. Clin. Chem. 27, 1410–1416. 10.1093/clinchem/27.8.1410
    1. Gourley G. R. (1997). Bilirubin Metabolism and Kernicterus. Adv. Pediatr. 44, 173–229.
    1. Hong J. I., Chang B.-Y. (2014). Development of the Smartphone-Based Colorimetry for Multi-Analyte Sensing Arrays. Lab. Chip 14, 1725–1732. 10.1039/C3LC51451J
    1. Hooda V., Gahlaut A., Gothwal A., Hooda V. (2017). Bilirubin Enzyme Biosensor: Potentiality and Recent Advances towards Clinical Bioanalysis. Biotechnol. Lett. 39, 1453–1462. 10.1007/s10529-017-2396-0
    1. Jendrassik L., Gróf P. (1938). Vereinfachte photometrische Methoden zur Bestimmung des Blutbilirubins. Biochemische Zeitschrift. Biochem. Z. 297, 82–89.
    1. Kamath P. S. (1996). Clinical Approach to the Patient with Abnormal Liver Test Results. Mayo Clin. Proc. 71, 1089–1095. quiz 1094–1095. 10.4065/71.11.1089
    1. Keahey P. A., Simeral M. L., Schroder K. J., Bond M. M., Mtenthaonnga P. J., Miros R. H., et al. (2017). Point-of-care Device to Diagnose and Monitor Neonatal Jaundice in Low-Resource Settings. Proc. Natl. Acad. Sci. U.S.A. 114, E10965–E10971. 10.1073/pnas.1714020114
    1. Mano N., Edembe L. (2013). Bilirubin Oxidases in Bioelectrochemistry: Features and Recent Findings. Biosens. Bioelectron. 50, 478–485. 10.1016/j.bios.2013.07.014
    1. McEnroe R. J. (2020). EP06 Evaluation of Linearity of Quantitative Measurement Procedures. Pennsylvania, United States: Clinical And Laboratory.
    1. McEnroe R. J., Durham D. G., Paul A., Miller J. J., Petrides W. G., Victoria M., et al. (2018). Interference Testing in Clinical Chemistry. Wayne, PA: Clinical and Laboratory Standards Institute.
    1. McEnroe R. J. (2014). Evaluation of Precision of Quantitative Measurement Procedures: Approved Guideline. Pennsylvania, United States: Clinical and Laboratory Standards Institute.
    1. Mutlu A. Y., Kılıç V., Özdemir G. K., Bayram A., Horzum N., Solmaz M. E. (2017). Smartphone-based Colorimetric Detection via Machine Learning. Analyst 142, 2434–2441. 10.1039/C7AN00741H
    1. Ngashangva L., Bachu V., Goswami P. (2019). Development of New Methods for Determination of Bilirubin. J. Pharm. Biomed. Analysis 162, 272–285. 10.1016/j.jpba.2018.09.034
    1. Pedregosa F., Varoquaux G., Gramfort A., Michel V., Thirion B., Grisel O., et al. (2011). Scikit-Learn: Machine Learning in Python. Journal of Machine Learning Research 12, 2825–2830.
    1. Pierson-Perry J. F. (2012). Evaluation of Detection Capability for Clinical Laboratory Measurement Procedures: Approved Guideline. Pennsylvania, United States: Clinical and Laboratory Standards Institute.
    1. Pierson-Perry J. F. (2009). Evaluation of Stability of in Vitro Diagnostic Reagents; Approved Guideline. Pennsylvania, United States: Clinical and Laboratory Standards Insitute.
    1. Rand R. N., Pasqua A. d. (1962). A New Diazo Method for the Determination of Bilirubin. Clin. Chem. 8, 570–578. 10.1093/clinchem/8.6.570
    1. Rawal R., Kharangarh P. R., Dawra S., Tomar M., Gupta V., Pundir C. S. (2020). A Comprehensive Review of Bilirubin Determination Methods with Special Emphasis on Biosensors. Process Biochem. 89, 165–174. 10.1016/j.procbio.2019.10.034
    1. Rutkowski R. B., deBaare L. (1966). An Ultramicro Colorimetric Method for Determination of Total and Direct Serum Bilirubin. Clin. Chem. 12, 432–438. 10.1093/clinchem/12.7.432
    1. Shoham B. (1995). A Bilirubin Biosensor Based on a Multilayer Network Enzyme Electrode. Biosens. Bioelectron. 10, 341–352. 10.1016/0956-5663(95)96852-P
    1. Swift. Apple Inc (2015). Available at: .
    1. Tan W., Zhang L., Doery J. C. G., Shen W. (2020). Study of Paper-Based Assaying System for Diagnosis of Total Serum Bilirubin by Colorimetric Diazotization Method. Sensors Actuators B Chem. 305, 127448. 10.1016/j.snb.2019.127448
    1. Tan W., Zhang L., Doery J. C. G., Shen W. (2020). Three-dimensional Microfluidic Tape-Paper-Based Sensing Device for Blood Total Bilirubin Measurement in Jaundiced Neonates. Lab. Chip 20, 394–404. 10.1039/C9LC00939F

Source: PubMed

Szponzorok és közreműködők

Egészségi állapot

Kábítószer-beavatkozások

3
Iratkozz fel