skip to main content
US FlagAn official website of the United States government
dot gov icon
Official websites use .gov
A .gov website belongs to an official government organization in the United States.
https lock icon
Secure .gov websites use HTTPS
A lock ( lock ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

Explore Research Products in the PAR
It may take a few hours for recently added research products to appear in PAR search results.


Title: Data-Driven Modeling of Smartphone-Based Electrochemiluminescence Sensor Data Using Artificial Intelligence
Understanding relationships among multimodal data extracted from a smartphone-based electrochemiluminescence (ECL) sensor is crucial for the development of low-cost point-of-care diagnostic devices. In this work, artificial intelligence (AI) algorithms such as random forest (RF) and feedforward neural network (FNN) are used to quantitatively investigate the relationships between the concentration of   Ru ( bpy ) 3 2 + luminophore and its experimentally measured ECL and electrochemical data. A smartphone-based ECL sensor with   Ru ( bpy ) 3 2 + /TPrA was developed using disposable screen-printed carbon electrodes. ECL images and amperograms were simultaneously obtained following 1.2-V voltage application. These multimodal data were analyzed by RF and FNN algorithms, which allowed the prediction of   Ru ( bpy ) 3 2 + concentration using multiple key features. High correlation (0.99 and 0.96 for RF and FNN, respectively) between actual and predicted values was achieved in the detection range between 0.02 µM and 2.5 µM. The AI approaches using RF and FNN were capable of directly inferring the concentration of   Ru ( bpy ) 3 2 + using easily observable key features. The results demonstrate that data-driven AI algorithms are effective in analyzing the multimodal ECL sensor data. Therefore, these AI algorithms can be an essential part of the modeling arsenal with successful application in ECL sensor data modeling.  more » « less
Award ID(s):
1706597
PAR ID:
10167174
Author(s) / Creator(s):
; ; ; ; ; ;
Date Published:
Journal Name:
Sensors
Volume:
20
Issue:
3
ISSN:
1424-8220
Page Range / eLocation ID:
625
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. ; ; ; ; ; (, Sensors)
    Machine learning (ML) can be an appropriate approach to overcoming common problems associated with sensors for low-cost, point-of-care diagnostics, such as non-linearity, multidimensionality, sensor-to-sensor variations, presence of anomalies, and ambiguity in key features. This study proposes a novel approach based on ML algorithms (neural nets, Gaussian Process Regression, among others) to model the electrochemiluminescence (ECL) quenching mechanism of the [Ru(bpy)3]2+/TPrA system by phenolic compounds, thus allowing their detection and quantification. The relationships between the concentration of phenolic compounds and their effect on the ECL intensity and current data measured using a mobile phone-based ECL sensor is investigated. The ML regression tasks with a tri-layer neural net using minimally processed time series data showed better or comparable detection performance compared to the performance using extracted key features without extra preprocessing. Combined multimodal characteristics produced an 80% more enhanced performance with multilayer neural net algorithms than a single feature based-regression analysis. The results demonstrated that the ML could provide a robust analysis framework for sensor data with noises and variability. It demonstrates that ML strategies can play a crucial role in chemical or biosensor data analysis, providing a robust model by maximizing all the obtained information and integrating nonlinearity and sensor-to-sensor variations. 
    more » « less
  2. ; ; ; ; ; (, Dalton Transactions)
    The synthesis, photophysics, and electrochemiluminescence (ECL) of four water-soluble dinuclear Ir( iii ) and Ru( ii ) complexes (1–4) terminally-capped by 4′-phenyl-2,2′:6′,2′′-terpyridine (tpy) or 1,3-di(pyrid-2-yl)-4,6-dimethylbenzene (N^C^N) ligands and linked by a 2,7-bis(2,2′:6′,2′′-terpyridyl)fluorene with oligoether chains on C9 are reported. The impact of the tpy or N^C^N ligands and metal centers on the photophysical properties of 1–4 was assessed by spectroscopic methods including UV-vis absorption, emission, and transient absorption, and by time-dependent density functional theory (TDDFT) calculations. These complexes exhibited distinct singlet and triplet excited-state properties upon variation of the terminal-capping terdentate ligands and the metal centers. The ECL properties of complexes 1–3 with better water solubility were investigated in neutral phosphate buffer solutions (PBS) by adding tripropylamine (TPA) as a co-reactant, and the observed ECL intensity followed the descending order of 3 > 1 > 2. Complex 3 bearing the [Ru(tpy) 2 ] 2+ units displayed more pronounced ECL signals, giving its analogues great potential for further ECL study. 
    more » « less
  3. ; ; ; ; ; ; (, Chemical Science)
    We report the excited-state behavior of a structurally simple bis -sulfoxide complex, cis -S,S-[Ru(bpy) 2 (dmso) 2 ] 2+ , as investigated by femtosecond pump–probe spectroscopy. The results reveal that a single photon prompts phototriggered isomerization of one or both dmso ligands to yield a mixture of cis -S,O-[Ru(bpy) 2 (dmso) 2 ] 2+ and cis -O,O-[Ru(bpy) 2 (dmso) 2 ] 2+ . The quantum yields of isomerization of each product and relative product distribution are dependent upon the excitation wavelength, with longer wavelengths favoring the double isomerization product, cis -O,O-[Ru(bpy) 2 (dmso) 2 ] 2+ . Transient absorption measurements on cis -O,O-[Ru(bpy) 2 (dmso) 2 ] 2+ do not reveal an excited-state isomerization pathway to produce either the S,O or S,S isomers. Femtosecond pulse shaping experiments reveal no change in the product distribution. Pump–repump–probe transient absorption spectroscopy of cis -S,S-[Ru(bpy) 2 (dmso) 2 ] 2+ shows that a pump–repump time delay of 3 ps dramatically alters the S,O : O,O product ratio; pump–repump–probe transient absorption spectroscopy of cis -O,O-[Ru(bpy) 2 (dmso) 2 ] 2+ with a time delay of 3 ps uncovers an excited-state isomerization pathway to produce the S,O isomer. In conjunction with low-temperature steady-state emission spectroscopy, these results are interpreted in the context of an excited-state bifurcating pathway, in which the isomerization product distribution is determined not by thermodynamics, but rather as a dynamics driven reaction. 
    more » « less
  4. ; ; ; ; ; (, Molecules)
    null (Ed.)
    4,5-diazafluorene (daf) and 9,9’-dimethyl-4,5-diazafluorene (Me2daf) are structurally similar to the important ligand 2,2’-bipyridine (bpy), but significantly less is known about the redox and spectroscopic properties of metal complexes containing Me2daf as a ligand than those containing bpy. New complexes Mn(CO)3Br(daf) (2), Mn(CO)3Br(Me2daf) (3), and [Ru(Me2daf)3](PF6)2 (5) have been prepared and fully characterized to understand the influence of the Me2daf framework on their chemical and electrochemical properties. Structural data for 2, 3, and 5 from single-crystal X-ray diffraction analysis reveal a distinctive widening of the daf and Me2daf chelate angles in comparison to the analogous Mn(CO)3(bpy)Br (1) and [Ru(bpy)3]2+ (4) complexes. Electronic absorption data for these complexes confirm the electronic similarity of daf, Me2daf, and bpy, as spectra are dominated in each case by metal-to-ligand charge transfer bands in the visible region. However, the electrochemical properties of 2, 3, and 5 reveal that the redox-active Me2daf framework in 3 and 5 undergoes reduction at a slightly more negative potential than that of bpy in 1 and 4. Taken together, the results indicate that Me2daf could be useful for preparation of a variety of new redox-active compounds, as it retains the useful redox-active nature of bpy but lacks the acidic, benzylic C–H bonds that can induce secondary reactivity in complexes bearing daf. 
    more » « less
  5. ; ; ; ; ; ; (, Dalton Transactions)
    null (Ed.)
    Combining strain-promoted azide–alkyne cycloaddition (SPAAC) and inorganic click (iClick) reactivity provides access to metal 1,2,3-triazolates. Experimental and computational insights demonstrate that iClick reactivity of the tested metal azides (LM-N 3 , M = Au, W, Re, Ru and Pt) depends on the accessibility of the azide functionality rather than electronic effects imparted by the metal. SPAAC iClick reactivity with cyclooctyne is observed when the azide functionality is sterically unencumbered, e.g. [Au(N 3 )(PPh 3 )] (Au–N3), [W(η 3 -allyl)(N 3 )(bpy)(CO) 2 ] (W–N3), and [Re(N 3 )(bpy)(CO) 3 ] [bpy = 2,2′-bipyridine] (Re–N3). Increased steric bulk and/or preequilibria with high activation barriers prevent SPAAC iClick reactivity for the complexes [Ru(N 3 )(Tp)(PPh 3 ) 2 ] [Tp = tris(pyrazolyl)borate] (Ru–N3), [Pt(N 3 )(CH 3 )(P i Pr 3 ) 2 ] [ i Pr = isopropyl] (Pt(II)–N3), and [Pt(N 3 )(CH 3 ) 3 ] 4 ((PtN3)4). Based on these computational insights, the SPAAC iClick reactivity of [Pt(N 3 )(CH 3 ) 3 (P(CH 3 ) 3 ) 2 ] (Pt(IV)–N3) was successfully predicted. 
    more » « less
  • Citation Formats
  • MLA
  • APA
  • Chicago
  • BibTeX
  • Save / Share this Record
  • Send to Email