Polymer-grafted SWCNTs as Synthetic Receptors

Overview

Single-walled carbon nanotubes (SWCNTs) are cylindrical carbon tubes capable of emitting fluorescence in the near -infrared region (NIR), which overlaps with the transparent window for imaging biological tissue. SWCNT fluorescence is environmentally sensiti ve, with peak wavelengths and intensity changing upon interaction with different molecules such as proteins, peptides, or lipids. To utilize SWCNTs as synthetic receptors, they must be water soluble and fluorescent; however, their hydrophobic properties qu ench fluorescence in aqueous solutions. A promising alternative is grafting polymers onto SWCNTs by optimizing the monomer concentrations. In this study, we successfully grafted Bis[2-(methacryloyloxy)ethyl] phosphate (PC), a polymer with a similar molecular structure to DNA, onto SWCNTs, achieving dispersion in aqueous solutions without quenching fluorescence. We found that polymer -grafted SWCNTs exhibit stability under extreme pH conditions, making them potentially useful for imaging and examining biological tissue. We also began to develop a method to graft molecularly imprinted polymers (MIPs) onto SWCNTs to create highly specific and sensitive receptors. MIPs are synthetic antibodies with predetermined selectivity and specificity for a template molecule. A deeper understanding of polymer -grafted and MIP-SWCNTs is critical for their use as synthetic receptors, enabling imprinting with disease bio markers. This advancement would enable early detection of diseases and significantly improve patient survival rates and clinical treatments. New York-Upstate Neural Biomarker-Based Diagnosis of Alzheimer’s Disease: AI Models Sensitivity and Accuracy Results on Multiple Electroencephalography Data Sets Shrey Kumar Horace Greeley High School, Chappaqua, NY Millions of people in the United States suffer from Alzheimer’s Disease (AD), an incurable form of dementia that continues to increase in prevalence. Current methods of AD diagnosis are limited to a late stage by which time the treatment options are limite d, quality of life is poor, and cost of treatment is exponentially high. Early medical diagnosis of AD is difficult since non -invasive techniques require extensive tests and still generate false positives and negatives, leading to misdiagnosis. This study proposes supervised machine learning models trained on readily available Electroencephalography (EEG) patient data to diagnose potential AD patients. Relevant features were extracted and analyzed from two opensource EEG databases, collected from 180 patients and used in four experiments to train and test the machine learning model of best fit. Our artificial intelligence (AI) model is an alternative to current late -stage detection methods which require complex and risky procedures that can lead to inaccuracies. In addition, current algorithms require feature manipulation and sort through thousands of raw EEG data points to obtain unreliable results. The results demonstrate that, given EEG data of 180 close -eyed patients, the trained logistic regression model - the machine learning model of best fit – achieved 100% sensitivity and 94% overall accuracy for the C4 electrode, using data recordings of only eight second segments for each patient. This novel AD screening tool, with a cloud -based AI model, can be easily deployed at primary health care clinics to screen patients for AD during their yearly clinical visits to increase early diagnosis.

Competition history

• JSHS 2025 Category not listed

Resources

• National JSHS abstract book ↗

Related projects

Closest projects by meaning, across every fair and year in the corpus.