SMART-NeuroDx: A Reagent-Free Multianalyte Biosensor Platform with Machine-Learning Readout for Point-of-Care Dementia Screening
Deshpande, S.; Pawlak, K.; Prathap, M. B.; Evans, C.; Al-Alam, T.; Joy, A. M.; Arjun, A. M.; Sharma, S.
Show abstract
Differentiating overlapping dementia pathologies, such as frontotemporal dementia and Alzheimers disease, calls for the simultaneous measurement of several blood biomarkers, yet electrochemical sensors remain predominantly single-target and dependent on labels and sample preparation. Here we report SMART-NeuroDx, a reagent-free electrochemical array that quantifies pTau217, GFAP, pTau181, and NfL directly from unprocessed plasma and serum in under 35 minutes. Four surface-confined redox-active molecularly imprinted polymer and aptamer-MIP recognition matrices are electropolymerized onto a four-working-electrode porous-gold printed-circuit array, using potential-assisted electrostatic gating keyed to each targets isoelectric point to prevent cross-channel template contamination during synthesis. Label-free Faradaic responses from the redox-active polymer backbone are acquired on a custom battery-powered STM32G4 handheld potentiostat and converted to concentration from five voltammetric features using cross-validated Random Forest and XGBoost regressors. The handheld unit reproduced the baseline fidelity of a commercial benchtop workstation and resolved sub-picogram pTau217 (limit of detection 0.087 pg mL-{superscript 1}) across marker-appropriate dynamic ranges, with inter-chip relative standard deviation at or below 5.34% and coefficients of determination of 0.94 to 0.99 against reference concentrations. Each channel retained selectivity in plasma and serum against competing neurological and inflammatory proteins, with non-specific signal deviation held below 10% by the hydrated PyPEG interfacial shell. The platform establishes reagent-free, simultaneous four-analyte neurodegeneration sensing on a single point-of-care device.
Matching journals
The top 5 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Affordable plasmonic biosensing: democratizing SERS with scalable, field-compatible substrate fabrication 97%
- Synthetic Nanobody-Functionalized Nanoparticles for Accelerated Development of Rapid, Accessible Detection of Viral Antigens 96%
- Nanoparticle-Supported, Rapid, Digital Quantification of Neutralizing Antibodies Against SARS-CoV-2 Variants 96%
Similar papers in this journal
Similar papers in this journal
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.