Smartphone-based urinary biomarker detection: an application-oriented device and algorithm
Abstract
Persons of advanced age have a higher risk of contracting urinary tract infection than the general population. Functional impairments, cognitive impairments, urine incontinence and advanced age encumbers the management of urinary tract infection. Persons in nursing homes or who receive other types of care, such as home care services often present with symptoms interpreted as urinary incontinence, and hence a large portion often use adult diapers. Impairments, incontinence and diaper usage can make collection of urine samples for analysis difficult.
A non-invasive rapid screening point-of-care system was developed to avoid unnecessary catheterization that may inadvertently cause infections, avoid overuse of antibiotics that may increase antibiotic resistance of bacteria, and to help nurses in the home care service. The system consists of a paper-based microfluidic colorimetric assay compatible with commercially available diapers, and a smartphone application for analysis of colorimetric results. The device is capable of automatic urine sampling from a diaper by utilizing absorbent media, and transports the urine internally with capillary flow in porous media to a set of colorimetric reaction pads. A superabsorbent polymer-swelling valve closes the inlet after a sufficient amount of urine have been absorbed and isolates the inside of the device from environmental influence and evaporation. The device is specifically made to provide a non-invasive solution for automatic sampling and testing of a urine sample, where, in the case of home care nurses, a device can be inserted during one diaper change, then removed and read during a consecutive diaper change.
The fabricated prototype devices exhibited consistent and repeatable behaviour. Saturation of reaction pads, and valve swelling occurred in approximately 7 minutes. All colorimetric reactions had run to completion within 30 minutes, and reaction colors were preserved and distinguishable for 8 hours after saturation.
The smartphone application attempts to provide an objective measurement independent of illumination and device conditions, and presents measurements in a report format or as a historical plot that includes previous tests to visualize biomarker variation over time. Illumination and device independence is crucial. If test and calibration is performed under different conditions, the calibration may not be a representative basis of comparison for the test. Other published methods in the scientific literature tended to have impractical constraints, such as putting limitations on the permissible color change. To provide Illumination and device independence for general colorimetric reactions without requiring additional equipment such as smartphone attachments, illumination sources, setups etc. a set of printed reference colors were provided for each biomarker. When photographing test and reference simultaneously, the variation in conditions apply approximately equally to both, thereby minimizing the impact of illumination and device conditions. Multiple references for multiple biomarkers require data extraction from multiple locations on the device surface. For the purpose of user-friendliness and simplicity, a pattern detection algorithm was included to detect device corners, and a homographic transform algorithm was implemented to establish a map between the known data extraction locations in the design and the photographed image. The algorithm was capable of correct classification for different illuminations on different smartphone models.
Has parts
Article 1: Karlsen, H. and T. Dong, A smart phone-based robust correction algorithm for the colorimetric detection of Urinary Tract Infection. 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), publisher: IEEE, 2015: p. 1251-1254. DOI: 10.1109/EMBC.2015.7318594Article 2: Karlsen, H. and T. Dong, Illumination and device independence for colorimetric detection of urinary biomarkers with smartphone. 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), publisher: IEEE, 2016: p. 5184-5187. DOI: 10.1109/EMBC.2016.7591895.
Article 3: Karlsen, H. and T. Dong, Smartphone-Based Rapid Screening of Urinary Biomarkers. IEEE Transactions on Biomedical Circuits and Systems, publisher: IEEE, 2017. 11(2): p. 455-463. DOI: 10.1109/TBCAS.2016.2633508
Article 4: Karlsen, H. and T. Dong, Biomarkers of urinary tract infections: state of the art, and promising applications for rapid strip-based chemical sensors. Analytical Methods, publisher: Royal Society of Chemistry, 2015. 7(19): p. 7961-7975. DOI:10.1039/C5AY01678A
Article 5: Karlsen, H. and T. Dong, A Compact Device for Urine Collection and Transport in Porous Media. Mechatronics 2017: Recent Technological and Scientific Advances, publisher Springer, 2017. 644: p. 3. DOI: 10.1007/978-3-319-65960-2_1
Article 6: Karlsen, H. and T. Dong, A diaper pad for diaper-based urine collection and colorimetric screening of urinary biomarkers. Accepted. Annals of Biomedical Engineering, publisher: Springer