A Paper-Based Multiplexed Serological Test to Monitor Immunity against SARS-COV-2 Using Machine Learning.

The rapid spread of SARS-CoV-2 caused the COVID-19 pandemic and accelerated vaccine development to prevent the spread of the virus and control the disease. Given the sustained high infectivity and evolution of SARS-CoV-2, there is an ongoing interest in developing COVID-19 serology tests to monitor population-level immunity. To address this critical need, we designed a paper-based multiplexed vertical flow assay (xVFA) using five structural proteins of SARS-CoV-2, detecting IgG and IgM antibodies to monitor changes in COVID-19 immunity levels. Our platform not only tracked longitudinal immunity levels but also categorized COVID-19 immunity into three groups: protected, unprotected, and infected, based on the levels of IgG and IgM antibodies. We operated two xVFAs in parallel to detect IgG and IgM antibodies using a total of 40 µL of human serum sample in <20 min per test. After the assay, images of the paper-based sensor panel were captured using a mobile phone-based custom-designed optical reader and then processed by a neural network-based serodiagnostic algorithm. The serodiagnostic algorithm was trained with 120 measurements/tests and 30 serum samples from 7 randomly selected individuals and was blindly tested with 31 serum samples from 8 different individuals, collected before vaccination as well as after vaccination or infection, achieving an accuracy of 89.5%. The competitive performance of the xVFA, along with its portability, cost-effectiveness, and rapid operation, makes it a promising computational point-of-care (POC) serology test for monitoring COVID-19 immunity, aiding in timely decisions on the administration of booster vaccines and general public health policies to protect vulnerable populations.

Single-tier point-of-care serodiagnosis of Lyme disease.

Point-of-care (POC) serological testing provides actionable information for several difficult to diagnose illnesses, empowering distributed health systems. Accessible and adaptable diagnostic platforms that can assay the repertoire of antibodies formed against pathogens are essential to drive early detection and improve patient outcomes. Here, we report a POC serologic test for Lyme disease (LD), leveraging synthetic peptides tuned to be highly specific to the LD antibody repertoire across patients and compatible with a paper-based platform for rapid, reliable, and cost-effective diagnosis. A subset of antigenic epitopes conserved across Borrelia burgdorferi genospecies and targeted by IgG and IgM antibodies, were selected based on their seroreactivity to develop a multiplexed panel for a single-step measurement of combined IgM and IgG antibodies from LD patient sera. Multiple peptide epitopes, when combined synergistically using a machine learning-based diagnostic model, yielded a high sensitivity without any loss in specificity. We blindly tested the platform with samples from the U.S. Centers for Disease Control & Prevention (CDC) LD repository and achieved a sensitivity and specificity matching the lab-based two-tier results with a single POC test, correctly discriminating cross-reactive look-alike diseases. This computational LD diagnostic test can potentially replace the cumbersome two-tier testing paradigm, improving diagnosis and enabling earlier effective treatment of LD patients while also facilitating immune monitoring and surveillance of the disease in the community.

Deep Learning-Enabled Multiplexed Point-of-Care Sensor using a Paper-Based Fluorescence Vertical Flow Assay.

Multiplexed computational sensing with a point-of-care serodiagnosis assay to simultaneously quantify three biomarkers of acute cardiac injury is demonstrated. This point-of-care sensor includes a paper-based fluorescence vertical flow assay (fxVFA) processed by a low-cost mobile reader, which quantifies the target biomarkers through trained neural networks, all within <15 min of test time using 50 µL of serum sample per patient. This fxVFA platform is validated using human serum samples to quantify three cardiac biomarkers, i.e., myoglobin, creatine kinase-MB, and heart-type fatty acid binding protein, achieving less than 0.52 ng mL-1 limit-of-detection for all three biomarkers with minimal cross-reactivity. Biomarker concentration quantification using the fxVFA that is coupled to neural network-based inference is blindly tested using 46 individually activated cartridges, which shows a high correlation with the ground truth concentrations for all three biomarkers achieving >0.9 linearity and <15% coefficient of variation. The competitive performance of this multiplexed computational fxVFA along with its inexpensive paper-based design and handheld footprint makes it a promising point-of-care sensor platform that can expand access to diagnostics in resource-limited settings.

Proteasomes in Patient Rectal Cancer and Different Intestine Locations: Where Does Proteasome Pool Change?

A special problem in the surgery of rectal cancer is connected with a need for appropriate removal of intestine parts, along with the tumor, including the fragment close to the sphincter. To determine the length of fragments to remove, it is necessary to reveal areas without changes in molecule functioning, specific for tumor. The purpose of the present study was to investigate functioning the proteasomes, the main actors in protein hydrolysis, in patient rectal adenocarcinoma and different intestine locations. Chymotrypsin-like and caspase-like activities, open to complex influence of different factors, were analyzed in 43-54 samples by Suc-LLVY-AMC- and Z-LLE-AMC-hydrolysis correspondingly. Both activities may be arranged by the decrease in the location row: cancer→adjacent tissue→proximal (8-20 cm from tumor) and distal (2 and 4 cm from tumor) sides. These activities did not differ noticeably in proximal and distal locations. Similar patterns were detected for the activities and expression of immune subunits LMP2 and LMP7 and expression of 19S and PA28αß activators. The largest changes in tumor were related to proteasome subtype containing LMP2 and PA28αß that was demonstrated by native electrophoresis. Thus, the results indicate a significance of subtype LMP2-PA28αß for tumor and absence of changes in proteasome pool in distal fragments of 2-4 cm from tumor.

Neural Network-Based On-Chip Spectroscopy Using a Scalable Plasmonic Encoder.

Conventional spectrometers are limited by trade-offs set by size, cost, signal-to-noise ratio (SNR), and spectral resolution. Here, we demonstrate a deep learning-based spectral reconstruction framework using a compact and low-cost on-chip sensing scheme that is not constrained by many of the design trade-offs inherent to grating-based spectroscopy. The system employs a plasmonic spectral encoder chip containing 252 different tiles of nanohole arrays fabricated using a scalable and low-cost imprint lithography method, where each tile has a specific geometry and thus a specific optical transmission spectrum. The illumination spectrum of interest directly impinges upon the plasmonic encoder, and a CMOS image sensor captures the transmitted light without any lenses, gratings, or other optical components in between, making the entire hardware highly compact, lightweight, and field-portable. A trained neural network then reconstructs the unknown spectrum using the transmitted intensity information from the spectral encoder in a feed-forward and noniterative manner. Benefiting from the parallelization of neural networks, the average inference time per spectrum is ∼28 µs, which is much faster compared to other computational spectroscopy approaches. When blindly tested on 14 648 unseen spectra with varying complexity, our deep-learning based system identified 96.86% of the spectral peaks with an average peak localization error, bandwidth error, and height error of 0.19 nm, 0.18 nm, and 7.60%, respectively. This system is also highly tolerant to fabrication defects that may arise during the imprint lithography process, which further makes it ideal for applications that demand cost-effective, field-portable, and sensitive high-resolution spectroscopy tools.

Deep learning-enabled point-of-care sensing using multiplexed paper-based sensors.

We present a deep learning-based framework to design and quantify point-of-care sensors. As a use-case, we demonstrated a low-cost and rapid paper-based vertical flow assay (VFA) for high sensitivity C-Reactive Protein (hsCRP) testing, commonly used for assessing risk of cardio-vascular disease (CVD). A machine learning-based framework was developed to (1) determine an optimal configuration of immunoreaction spots and conditions, spatially-multiplexed on a sensing membrane, and (2) to accurately infer target analyte concentration. Using a custom-designed handheld VFA reader, a clinical study with 85 human samples showed a competitive coefficient-of-variation of 11.2% and linearity of R 2 = 0.95 among blindly-tested VFAs in the hsCRP range (i.e., 0-10 mg/L). We also demonstrated a mitigation of the hook-effect due to the multiplexed immunoreactions on the sensing membrane. This paper-based computational VFA could expand access to CVD testing, and the presented framework can be broadly used to design cost-effective and mobile point-of-care sensors.