Nucleic acid based point-of-care diagnostic technology for infectious disease detection using machine learning empowered smartphone-interfaced quantitative colorimetry.

We report a nucleic acid-based point of care testing technology for infectious disease detection at resource limited settings by integrating a low-cost portable device with machine learning-empowered quantitative colorimetric analytics that can be interfaced via a smartphone application. We substantiate our proposition by demonstrating the efficacy of this technology in detecting COVID-19 infection from human swab samples, using the RT-LAMP protocol. Comparison with gold standard results from real-time PCR evidences high sensitivity and specificity, ensuring simplicity, portability, and user-friendliness of the technology at the same time. Colorimetric analytics of the reaction output without necessitating the opening of the reaction microchambers enables execution of the complete test workflow without any laboratory control that may otherwise be required stringently for safeguarding against carryover contamination. Seamless sample-to-answer workflow and machine learning-based readout further assures minimal human intervention for the test readout, thus eliminating inevitable inaccuracies stemming from erroneous execution of the test as well as subjectivity in interpreting the outcome. Our results further indicate the possibilities of upgrading the technology to predict the pathogenic load on the infected patients akin to the cyclic threshold value of the real-time PCR, when calibrated with reference to a wide range of 'training' data for the machine learner, thereby putting forward the same as viable alternative to the resource-intensive PCR tests that cannot be made readily accessible at underserved community settings.

Instrument-free single-step direct estimation of the plasma glucose level from one drop of blood using smartphone-interfaced analytics on a paper strip.

We demonstrated an instrument-free miniaturized adaptation of the laboratory gold standard methodology for the direct estimation of plasma glucose from a drop of whole blood using a low-cost single-user-step paper-strip sensor interfaced with a smartphone. Unlike a majority of the existing glucose meters that use whole blood-based indirect sensing technologies, our direct adaptation of the gold-standard laboratory benchmark could eliminate the possibilities of cross interference with other analytes present in the whole blood by facilitating an in situ plasma separation, capillary flow and colorimetric reaction occurring concomitantly, without incurring additional device complexity or embodiment. The test reagents were dispensed in lyophilized form, and the resulting paper strips were found to be stable over three months stored in a normal freezer, rendering easy adaptability commensurate with the constrained supply chains in extreme resource-poor settings. Quantitative results could be arrived at via a completely-automated mobile-app-based image analytics interface developed using dynamic machine learning, obviating manual interpretation. The tests were demonstrated to be of high efficacy, even when executed by minimally trained frontline personnel having no special skill of drawing precise volume of blood, on deployment at under-resourced community centres having no in-built or accessible healthcare infrastructure. Clinical validation using 220 numbers of human blood samples in a double-blinded manner evidenced sensitivity and specificity of 98.11% and 96.7%, respectively, as compared to the results obtained from a laboratory-benchmarked biochemistry analyser, establishing its efficacy for public health and community disease management in resource-limited settings without any quality compromise of the test outcome.

Smartphone-Enabled Paper-Based Hemoglobin Sensor for Extreme Point-of-Care Diagnostics.

We report a simple, affordable (∼0.02 US $/test), rapid (within 5 min), and quantitative paper-based sensor integrated with smartphone application for on-spot detection of hemoglobin (Hgb) concentration using approximately 10 µL of finger-pricked blood. Quantitative analytical colorimetry is achieved via an Android-based application (Sens-Hb), integrating key operational steps of image acquisition, real-time analysis, and result dissemination. Further, feedback from the machine learning algorithm for adaptation of calibration data offers consistent dynamic improvement for precise predictions of the test results. Our study reveals a successful deployment of the extreme point-of-care test in rural settings where no infrastructural facilities for diagnostics are available. The Hgb test device is validated both in the controlled laboratory environment (n = 200) and on the field experiments (n = 142) executed in four different Indian villages. Validation results are well correlated with the pathological gold standard results (r = 0.9583) with high sensitivity and specificity for the healthy (n = 136) (>11 g/dL) (specificity: 97.2%), mildly anemic (n = 55) (<11 g/dL) (sensitivity: 87.5%, specificity: 100%), and severely anemic (n = 9) (<7 g/dL) (sensitivity: 100%, specificity: 100%) samples. Results from field trials reveal that only below 5% cases of the results are interpreted erroneously by classifying mildly anemic patients as healthy ones. On-field deployment has unveiled the test kit to be extremely user friendly that can be handled by minimally trained frontline workers for catering the needs of the underserved communities.