Towards real-time additive-free dopamine detection at $10^{-8}$ mM with hardware accelerated platform integrated on camera

Andrea Fratalocchi; Arturo Burguete-Lopez; Fei Xiang; Ning Li; Qizhou Wang; Zhao He

REVIEW 3 major objections 6 minor 75 references

Towards real-time additive-free dopamine detection at $10^{-8}$ mM with hardware accelerated platform integrated on camera

T0 review · 3 major / 6 minor · reviewed 2026-08-15 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read This paper claims that a camera-integrated optical platform using a light-scattering membrane and trained metasurface filters can detect dopamine at $10^{-8}$ mM in real time without added enzymes, two orders of magnitude below the best…

desk verdict Clever hardware demo, but the 10^-8 mM dopamine claim needs blank controls and proper LOD statistics before it can be taken at face value. read the letter →

arxiv 2506.13447 v1 pith:RB6JLVMS submitted 2025-06-16 physics.ins-det physics.med-ph

Ning Li , Qizhou Wang , Zhao He , Arturo Burguete-Lopez , Fei Xiang , Andrea Fratalocchi This is my paper · ORCID

classification physics.ins-detphysics.med-ph

keywords dopaminedetectionhardware-acceleratedsensingmetasurfaceencoderprincipalcomponentanalysisexplainableAIscatteringspectroscopypoint-of-carediagnosticsreal-timebiosensor

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper tries to establish that an optical, additive-free sensor built onto a commercial camera can detect dopamine at $10^{-8}$ mM in real time, even in the presence of uric acid, ascorbic acid, and phosphate buffer. If true, that would put the detection limit two orders of magnitude below the best cited additive-free electrochemical sensor and inside the concentration range linked to neurological disorders. The authors report a dynamic range from $10^{-8}$ to $10^{-1}$ mM in PBS and from $10^{-8}$ to $10^{-3}$ mM with uric and ascorbic acid interferents, with test-set $R^2$ scores above 0.99. The appeal is practical: the camera form factor and roughly $1000 USD cost could replace bulky electrochemical workstations and enable continuous point-of-care monitoring.

What carries the argument

The load-bearing object is the optical hardware encoder: a set of subpixel metasurface filters with transmission functions $\Lambda_i(\omega)$, inverse-designed and placed directly over camera sensor pixels. Each latent feature is computed as $z_i = \sigma\left[\int_{\omega_0}^{\omega_1} I_o(\omega,y)\Lambda_i(\omega)\,d\omega\right]$, a multiply-and-accumulate operation with the photodetector response as the nonlinearity. The filters are trained from principal component analysis of 240 reflection spectra for DA in PBS and 180 spectra with AA and UA, and the four most informative components are chosen by integrated-gradient explainable-AI scores rather than by explained variance alone. The physical transducer is a microfluidic cell whose bottom surface is a nanostructured silicon wafer decorated with gold nanoparticles, engineered per analyte to maximize latent-space separation, plus a hydrophobic top layer that keeps samples from contaminating the scatterer and makes the cell reusable.

What would settle it

Prepare several blank samples (PBS only, and PBS with uric and ascorbic acid but no dopamine) and several independently prepared $10^{-8}$ mM dopamine samples, read them on the same scatterer and camera over repeated runs, and check whether the regression outputs form two non-overlapping, time-stable groups. If blank samples are predicted near $10^{-8}$ mM, or if repeated readings of the same sample drift across the claimed detection limit, the central claim fails.

Watch

Extended reading notes

Core claim

The central claim is that a trained optical encoder placed over the pixels of a standard monochrome camera can translate the scattering spectrum of a dopamine solution directly into a concentration readout, requiring no additives. Experimental results report dopamine detection at $10^{-8}$ mM in phosphate-buffered saline and in mixtures with uric and ascorbic acid, spanning a dynamic range of eight orders of magnitude in the first case and five orders in the second, with test-set $R^2$ scores of 0.9898 and 0.9926. The authors present this as a two-order improvement over the best additive-free electrochemical sensor, and six orders over a prior AI-assisted biosensor. The device operates at video rates and fits in a camera-sized footprint.

Load-bearing premise

The load-bearing premise is that the measured scattering spectra are a stable and reproducible function of dopamine concentration down to $10^{-8}$ mM, with no drift, no batch-to-batch variation, and no contribution from the buffer or interferent matrix that correlates with concentration.

Editorial extensions

If this is right

The claimed sensitivity reaches the $10^{-8}$\,$10^{-6}$ mM window linked to dopamine fluctuations in Parkinson's disease, schizophrenia, and related conditions, making the threshold relevant to disease monitoring.
Detection is claimed at video rates, so the platform could track dopamine dynamics in time rather than requiring tens of seconds to minutes per measurement.
Because no additives are used, the readout avoids the impurity and instability the authors attribute to enzyme- or polymer-based sensors, and the reusable hydrophobic cell removes the need for frequent sensor replacement.
The same trained-encoder architecture is claimed to be universal, with retraining of the optical filters expected to extend it to serotonin, norepinephrine, glutamate, GABA, and cytokine analytes.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

An immediate testable extension is to collect blank PBS and interferent-only spectra alongside repeated independent preparations at $10^{-8}$ mM; that would quantify how cleanly the claimed low-concentration readout separates from the matrix baseline.
A testable corollary is that the four selected principal components fully carry the concentration information, so the same subpixel filters should transfer to a second camera with only recalibration of the regression head; if cross-device transfer fails, the bottleneck is likely fabrication reproducibility rather than sensitivity.
If the low-concentration signal really arises from broadband scattering by the gold-decorated nanopillars, then variations in pillar height, gold decoration, or liquid contact angle should shift the calibration, suggesting batch-release testing of the scattering substrate as a quality-control step.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Dear X,

Quick take: this is a clever hardware demo with a real fabricated device, but the headline detection limit is not yet warranted by the experiments as reported. The gap between the claim and the evidence is specific and fixable with blank controls, replicates, and a proper LOD definition.

What is actually new: applying the group's trained-metasurface-encoder camera platform to dopamine in PBS and in UA/AA mixtures, with all the engineering that involves. The device is fabricated, characterized with SEM/EELS/XPS, and the authors provide source data and a public code repo. That is more reproducible than a lot of what crosses my desk. The comparison to electrochemical SOTA is useful context, and the idea of a camera-based, additive-free point-of-care dopamine sensor is genuinely attractive.

Where it gets soft: the 10^-8 mM claim is built on a regression R2 of ~0.99 over a pooled test set spanning eight concentration decades. That does not tell you the sensor can distinguish 10^-8 mM from blank. There are no PBS-only or AA/UA-only spectra, no repeated independent sample preparations per concentration, and no day-to-day or chip-to-chip consistency data. At 10 pM, any drift, buffer variation, or carryover can masquerade as signal. The PCA basis and IG feature selection are computed on the full dataset before the 80/20 split, so the test set partially leaks into the feature extractor. That makes the reported test R2 optimistic, probably by an amount that matters at the low end. The 'real-time video rate' claim also has no time-series experiment behind it, so it currently reads as a capability of the camera rather than a demonstrated performance.

None of this sinks the hardware concept. It means the paper needs a serious revision: show blanks, show replicates with error bars, define LOD from the blank distribution, and either redo feature selection inside the training folds or use a nested CV. Also show at least one independent measurement run.

Who is it for: people building ML-based optical sensors and anyone comparing dopamine LODs across platforms. I would bring it to reading group as a case study in how to, and how not to, report LODs from regression models.

Recommendation: send to peer review. A good referee will ask for the right controls, and the underlying engineering deserves a proper venue. But if I were the referee, I would be asking for the blank distribution before signing off on any number below 10^-6 mM.

Best, [Your name]

Referee Report

3 major / 6 minor

Summary. The manuscript reports a hardware-accelerated optical sensing platform for dopamine (DA) detection, combining a nanostructured scattering surface (Si pillars decorated with Au nanoparticles) with metasurface encoders integrated onto a commercial monochrome camera. The system is claimed to achieve real-time, additive-free DA detection at 10^-8 mM in phosphate buffer solution (PBS) and in the presence of ascorbic acid (AA) and uric acid (UA), with dynamic ranges spanning eight orders of magnitude and test-set R² values of 0.9898 and 0.9926. The data pipeline consists of collecting reflection spectra, applying principal component analysis (PCA) for dimensionality reduction, using integrated-gradients (IG) feature selection to identify the most informative components, fabricating hardware encoders that implement the selected projections, and training a regression network to map encoder outputs to concentration. The paper claims a two-order-of-magnitude improvement over the cited additive-free state of the art (Mn–MoS2/PGS, 5e-6 mM) and emphasizes the platform's universality and low cost.

Significance. If the central detection claim is properly substantiated, this work would represent a substantial advance in label-free, real-time neurotransmitter sensing, with a claimed two-order improvement in detection limit over the current additive-free electrochemical state of the art, a compact camera form factor, and video-rate operation. The hardware-encoder approach, in which trained spectral projections are physically implemented as nanostructured filters on a standard camera sensor, is a creative and potentially generalizable idea. The manuscript provides open-source code, describes fabrication in detail, and includes SEM/STEM/XPS characterization of the scattering surface. These strengths are real. However, the load-bearing experimental claim of detection at 10^-8 mM is not yet supported by the evidence presented: the absence of blank controls, replicate statistics, and proper validation of the feature-selection procedure leaves the result vulnerable to artifacts and optimistic bias. The paper's significance therefore rests on validation work that is currently missing rather than on the data as presented.

major comments (3)

[Results, Eq. (1) and Figs. 2a, 5g, 6c] The central claim that the platform detects DA at 10^-8 mM is not supported by adequate control experiments. In Eq. (1), the scattering response Io(ω, y) is treated as a function of analyte concentration y alone, but the manuscript shows no PBS-only or AA/UA-only blank spectra, no repeated independent sample preparations at each concentration, and no day-to-day or device-to-device consistency data. At 10^-8 mM (10 pM), any concentration-correlated artifact from buffer batch variation, drift during the measurement session, or matrix effects from AA/UA could masquerade as DA signal. To support the claimed LOD, the authors should provide blank spectra, replicate measurements with error bars, and ideally cross-device or cross-session validation.
[Results, Fig. 2a and Methods (feature selection)] The PCA and IG feature-selection procedures appear to be applied to the full dataset X before the 80/20 training/test split, as described in Fig. 2a and the surrounding text. If the test set contributes to the computation of the PCA basis and to the IG-selected features, the reported test-set R² values (0.9898 and 0.9926) are optimistically biased, because the test spectra have already influenced the encoder design. The RIE process is also described as separately optimized for each analyte to maximize feature separation, presumably using the same data. The authors should restrict all feature extraction and device optimization to the training portion only (e.g., via nested cross-validation) or otherwise demonstrate that the evaluation is not contaminated by this leakage.
[Results, Figs. 5g and 6c] The claimed limit of detection (LOD) of 10^-8 mM is not established by the reported pooled R² values. R² computed across eight decades of concentration does not quantify detection ability at the low-concentration end; LOD requires either the blank distribution (e.g., mean + 3σ of blank response) or per-concentration residuals and replicate statistics. The figures show no error bars and no per-concentration scatter. The text also contains an unclear statement about an 'enhancement of LOD of 80.0%' relative to the best integrated state of the art. The authors should define LOD operationally, provide per-concentration residuals, and show that the lowest concentration (10^-8 mM) is statistically distinguishable from the blank.

minor comments (6)

[Figure 2 caption] The caption contains a typo: 'Exaplainable' should be 'Explainable'.
[Results, Fig. 2d] The text says the four most informative components are selected, and Fig. 2d indicates components 1, 2, 3, and 8; the figure caption should state this explicitly for clarity.
[Results, Fig. 5g] The phrase 'enhancement of the LOD of 80.0%' is ambiguous; please specify the quantitative comparison used to derive this percentage.
[Methods, dataset description] The manuscript reports N=240 spectra across 8 concentrations (Fig. 2a) and N=180 spectra for the AA/UA case, but does not state how many independent sample preparations and how many replicate spectra per concentration were acquired; this information is needed to assess statistical power.
[Results, Eq. for Lrec] The loss is written as ||Io(ω,y) - I'_o(ω,y)|| without specifying the norm; if it is the mean squared error, the squared norm should be used. Please clarify.
[Data availability] The paper states 'Source data are provided with this paper' but does not describe the format or location; please specify where the spectral datasets and processed data can be accessed.

Circularity Check

2 steps flagged · score 6.0 of 10

Claimed LOD is the lowest calibration concentration; test R^2 is biased by pre-split feature selection.

self definitional [Results, Fig. 2a, Fig. 5g, Fig. 6c; Eq. (2) regression loss]
"The dataset X comprises N = 240 reflection spectra equally measured in 8 concentrations varying from 10−1 mM to 10−8 mM (Fig. 2a). ... The dynamic range achievable with this technique spans eight orders of magnitudes, from 10−8 mM to 10−1 mM, while the R2 score estimated from the test dataset reaches 0.9898. ... We here obtain LOD of 10−8 mM and a dynamic range spanning from 10−8 to 10−3 mM."

The claimed LOD is not derived from blank measurements, noise statistics, or extrapolation; it is simply the lowest concentration y included in the supervised training set. The regressor F is trained to minimize L_reg = ||ŷ − y|| with ground-truth labels y from the eight prepared concentrations, so the 'detection' at 10−8 mM restates the calibration range endpoint. The test-set R^2 reports interpolation accuracy within the trained range, not a detection limit, making the central LOD claim equivalent by construction to the experimental choice to measure down to 10−8 mM.
fitted input called prediction [Results, Fig. 2b-c; 'The training and testing portion of the dataset comprises 80% and 20%...']
"The training and testing portion of the dataset comprises 80% and 20% of the total samples, respectively. In this work, we perform dimensionality reduction by principal component analysis (PCA) (Fig. 2b). ... We identify the most useful principal components for analyte concentration prediction using the XAI integrated gradient (IG) method 50."

The paper describes PCA and IG feature selection on the full dataset X before stating the 80/20 split, so test-set spectra contribute to the learned projection basis and to the selected features. The reported test-set R^2 values (0.9898 and 0.9926) are therefore not fully independent held-out predictions: the feature extractor has already been fit using information from the test samples, which optimistically biases the evaluation of prediction performance.

full rationale

The central detection claim (LOD 10^-8 mM) is a calibration range endpoint rather than an independently derived detection limit: the regression network is trained on spectra labeled with the eight prepared concentrations, so reporting 'detection' at the lowest trained concentration is a restatement of the experimental design. This is partially circular for the LOD claim, although the held-out R^2 scores do provide some evidence of interpolation within the calibrated range. A second issue is that PCA and IG feature selection are performed on the full dataset before the 80/20 split, so the test-set R^2 is optimistically biased by information from test spectra in the feature extractor. The self-citations (refs 45 and 46) are prior work by the same group on metasurface encoders and inverse design; they support the hardware implementation but are not load-bearing for the dopamine result itself, and no uniqueness argument is imported from them. The absence of blank controls and replicate sample preparations is a support issue for the low-concentration claim but is not itself a circularity step.

Assumptions & free parameters 4 free parameters · 5 assumptions · 0 invented entities

The central claim rests primarily on data-driven feature selection and a supervised regressor; the paper contributes no first-principles derivation. The main assumptions are the stability and specificity of the scattering response to dopamine, neither of which is backed by blank controls or independent validation.

free parameters (4)

M = 8 PCA components = 8 (99.8% explained variance)
Truncation of the spectral dataset to 8 principal components is a modeling choice selected by explained variance from the same data.
Number of IG-selected encoder components = 4
The four PCA components with highest integrated-gradient importance are chosen for hardware; the cutoff is data-driven from the training set.
RIE process parameters per analyte = not specified numerically
SF6/CHF3/O2 flows, pressure, ICP/RF power and etch time are separately optimized for each analyte to maximize latent-space separation; these are hand-tuned experimental knobs.
Training/test split 80/20 = 0.8/0.2
Standard split choice; affects holdout estimates of R2.

assumptions (5)

domain assumption PCA captures the concentration-relevant spectral variance with M=8 components.
The encoder rests on linear dimensionality reduction of 240 reflection spectra; if the low-concentration contrast is nonlinear or noise-dominated, PCA features may not generalize.
domain assumption Integrated-gradient importance scores identify the PCA components that best predict concentration.
IG is applied to the auxiliary regressor F trained on the same spectra; its attributions are taken as ground truth for which optical filters to fabricate.
domain assumption Scattering spectra are stable across repeated measurements and device reuse.
No repeated independent spectra or inter-device reproducibility data are reported; PFDT hydrophobicity is offered as evidence of reusability.
domain assumption No analyte-independent drift or matrix effect contributes to class separation.
Blank PBS and interferent-only controls are not shown, so the concentration response is assumed to be caused by dopamine.
standard math Standard linear algebra and backpropagation for PCA, IG, and regression.
Uncontroversial background, but not independently verified in this manuscript.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Towards real-time additive-free dopamine detection at $10^{-8}$ mM with hardware accelerated platform integrated on camera." pith.science (2026). https://pith.science/paper/RB6JLVMS

@misc{pith2026250613447,
  author       = {Pith},
  title        = {Pith review of: Towards real-time additive-free dopamine detection at $10^-8$ mM with hardware accelerated platform integrated on camera},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/RB6JLVMS}},
  note         = {Machine review of arXiv:2506.13447}
}

abstract

Tracing physiological neurotransmitters such as dopamine (DA) with detection limits down to $\mathrm{1\times10^{-8}}$ mM is a critical goal in neuroscience for studying brain functions and progressing the understanding of cerebral disease. Addressing this problem requires enhancing the current state-of-the-art additive-free electrochemical workstation methods by over two orders of magnitude. In this work, we implement an ultra-sensitive, additive-free platform exploiting suitably engineered light-scattering membranes and optical accelerators integrated into commercial vision cameras, reporting real-time detection of DA in uric and ascorbic acid below the concentration of $\mathrm{10^{-8}}$ mM. These performances improve the current best technology by over two orders of magnitude in resolution while providing continuous, real-time detection at video rates. This technology also upgrades the bulk form factor of an electrochemical workstation with an imaging camera's compact and portable footprint. The optical accelerator implemented in this work is universal and trainable to detect a wide range of biological analytes. This technology's wide adoption could help enable early disease detection and personalized treatment adjustments while improving the management of neurological, mental, and immune-related conditions.

Figures

Figures reproduced from arXiv: 2506.13447 by the authors.

**Figure 1.** Schematics of the general sensing platform (a) Spectral dataset generation using a scattering surface with broadband illumination. (b) Example of a generated spectral dataset for a varying concentration of analyte. (c) Encoder-decoder unsupervised training model with explainable-AI (XAI) interface for selecting latent features. (d) Implementation design of hardware encoder with sub-pixels of inverse-designed nanores… view at source ↗

**Figure 2.** Design workflow of the optical hardware encoder. (a) Example of a dataset comprising scattered spectra from DA in PBS. (b) The dataset projected along the first three principal components extracted from the data. (c) Exaplainable-AI integrated gradient method for extracting the essential features in the latent space for predicting the analyte concentration y. (d) Principal component’s importance score vs. explained… view at source ↗

**Figure 3.** Hardware implementation on the camera chip of XAI-selected encoders (a) Integration of the hardware encoders on a CMOS camera sensor. (b) SEM image of the fabricated encoder nanostructures. 27 [PITH_FULL_IMAGE:figures/full_fig_p027_3.png] view at source ↗

Figures from the paper (7 more)

**Figure 4.** Figure 4: Implementation of the scattering device. (a) Optical image of an experimentally assembled scattering device comprising a reusable microfluidic cell with input and output channels. (b) Schematic assembly of the microfluidic device composed of the scatterer, PFDT, PDMS, …

**Figure 5.** Figure 5: Experimental results on DA interfered with PBS. (a) Schematic of the setup for measuring DA in PBS. (b) Tilted SEM and (c) EELS-elemental mapping image of the fabricated optical scattering device. (d) CA image of the scatterer treated by PFDT. (e) SEM image of the firs…

**Figure 6.** Figure 6: Detection of DA interfered with UA and AA. (a) SEM image of first three hardware encoders with corresponding (b) measured trained spectral weights. (c) LOD and dynamic range of DA detection compared against the state-of-the-art. The green rectangle region linked to the…

**Figure 1.** Figure 1: Cross-sectional HAADF-STEM image and EDS elemental mapping of the Si, Au, S and [PITH_FULL_IMAGE:figures/full_fig_p032_1.png]

**Figure 2.** Figure 2: XPS of scatterer (orange curve) and PFDT decorated scatterer (blue curve) : (a) S 2p, and [PITH_FULL_IMAGE:figures/full_fig_p032_2.png]

**Figure 3.** Figure 3: (a) Schematic of the measuring CA on the PFDT decorated scatterer at the same point for [PITH_FULL_IMAGE:figures/full_fig_p033_3.png]

**Figure 4.** Figure 4: CA image of the cover glass treated by FDTS. [PITH_FULL_IMAGE:figures/full_fig_p033_4.png]

Review history (2 revisions) →

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

75 extracted references · 75 canonical work pages

[1]

Tümay, S. O. & Y e¸ silot, S. Small molecule based water-soluble ﬂuorescence material for highly selective and ultra-sensitive detection of tnt: Design and spectroﬂuorimetric determination in real samples. Sensors and Actuators B: Chemical 343, 130088 (2021)

work page 2021
[2]

C., Ben-Jaber, S

To, K. C., Ben-Jaber, S. & Parkin, I. P . Recent Developments in the Field of Explosive Trace Detection. ACS Nano 14, 10804–10833 (2020)

work page 2020
[3]

Dai, L. et al. Ultrasensitive Physical, Bio, and Chemical Sensors Derived from 1-, 2-, and 3-D Nanocellulosic Materials. Small 16, 1906567 (2020)

work page 2020
[4]

& Bischof, J

Liu, Y ., Zhan, L., Qin, Z., Sackrison, J. & Bischof, J. C. Ultrasensitive and Highly Speciﬁc Lateral Flow Assays for Point-of-Care Diagnosis. ACS Nano 15, 3593–3611 (2021)

work page 2021
[5]

Kang, H. et al. Ultrasensitive Detection of SARS-CoV -2 Antibody by Graphene Field-Effect Transistors. Nano Letters 21, 7897–7904 (2021)

work page 2021
[6]

Mejía-Salazar, J. R. & Oliveira, O. N. J. Plasmonic Biosensing. Chemical Reviews 118, 10617–10625 (2018)

work page 2018
[7]

Sun, Z. et al. Skin-like ultrasensitive strain sensor for full-range detection of human health monitoring. ACS applied materials & interfaces 12, 13287–13295 (2020)

work page 2020
[8]

Feng, L. et al. Ultrasensitive and highly selective detection of strontium ions. Nature Sustainability 1–8 (2023). 14

work page 2023

Show all 75 references

[9]

& Y oon, M.-Y

Ha, N.-R., Jung, I.-P ., La, I.-J., Jung, H.-S. & Y oon, M.-Y . Ultra-sensitive detection of kanamycin for food safety using a reduced graphene oxide-based ﬂuorescent aptasensor. Scientiﬁc reports 7, 40305 (2017)

2017
[10]

Zhang, X. et al. Ultrasensitive Field-Effect Biosensors Enabled by the Unique Electronic Properties of Graphene. Small 16, 1902820 (2020)

2020
[11]

Wan, X. et al. Cascaded amplifying circuits enable ultrasensitive cellular sensors for toxic metals. Nature Chemical Biology 15, 540–548 (2019)

2019
[12]

& Vécsei, L

Sandi, D., Fricska-Nagy, Z., Bencsik, K. & Vécsei, L. Neurodegeneration in multiple sclerosis: Symptoms of silent progression, biomarkers and neuroprotective therapy—kynurenines are important players. Molecules 26, 3423 (2021)

2021
[13]

Sharma, N. S. et al. Dopamine levels in human tear ﬂuid. Indian Journal of Ophthalmology 67, 38–41 (2019)

2019
[14]

& Mysliwiec, M

Zbroch, E., Koc-Zorawska, E., Malyszko, J., Malyszko, J. & Mysliwiec, M. Circulating levels of renalase, norepinephrine, and dopamine in dialysis patients. Renal failure 35, 673–679 (2013)

2013
[15]

Shine, J. M. et al. Dopamine depletion alters macroscopic network dynamics in parkinson’s disease. Brain 142, 1024–1034 (2019)

2019
[16]

D., Bukala, B

Howes, O. D., Bukala, B. R. & Beck, K. Schizophrenia: from neurochemistry to circuits, symptoms and treatments. Nature Reviews Neurology 20, 22–35 (2024). 15

2024
[17]

Xu, J. et al. Dopamine-dependent neurotoxicity of α -synuclein: a mechanism for selective neurodegeneration in parkinson disease. Nature medicine 8, 600–606 (2002)

2002
[18]

& Brundin, P

Lotharius, J. & Brundin, P . Pathogenesis of parkinson’s disease: dopamine, vesicles and α -synuclein. Nature Reviews Neuroscience 3, 932–942 (2002)

2002
[19]

Martorana, A. et al. Dopamine modulates cholinergic cortical excitability in alzheimer’s disease patients. Neuropsychopharmacology 34, 2323–2328 (2009)

2009
[20]

Prosaposin maintains lipid homeostasis in dopamine neurons and counteracts experimental parkinsonism in rodents

He, Y .et al. Prosaposin maintains lipid homeostasis in dopamine neurons and counteracts experimental parkinsonism in rodents. Nature Communications 14, 5804 (2023)

2023
[21]

Decreased prefrontal dopamine d1 receptors in schizophrenia revealed by pet

Okubo, Y .et al. Decreased prefrontal dopamine d1 receptors in schizophrenia revealed by pet. Nature 385, 634–636 (1997)

1997
[22]

T., Loh, M., Deco, G

Rolls, E. T., Loh, M., Deco, G. & Winterer, G. Computational models of schizophrenia and dopamine modulation in the prefrontal cortex. Nature Reviews Neuroscience 9, 696–709 (2008)

2008
[23]

Demontis, D. et al. Genome-wide analyses of attention deﬁcit hyperactivity disorder iden- tify 27 risk loci, reﬁne the genetic architecture, and implicate several cognitive domains. Nature genetics 55, 198 (2023)

2023
[24]

Wang, S. et al. Dopamine detection and integration in neuromorphic devices for applications in artiﬁcial intelligence. Device 2 (2024)

2024
[25]

& Liu, J

Liu, X. & Liu, J. Biosensors and sensors for dopamine detection. View 2, 20200102 (2021). 16

2021
[26]

Single-atom doping of mos2 with manganese enables ultrasensitive detection of dopamine: Experimental and computational approach

Lei, Y .et al. Single-atom doping of mos2 with manganese enables ultrasensitive detection of dopamine: Experimental and computational approach. Science advances 6, eabc4250 (2020)

2020
[27]

A., Bukhari, N., Wabaidur, S

Alothman, Z. A., Bukhari, N., Wabaidur, S. M. & Haider, S. Simultaneous electrochemical determination of dopamine and acetaminophen using multiwall carbon nanotubes modiﬁed glassy carbon electrode. Sensors and Actuators B: Chemical 146, 314–320 (2010)

2010
[28]

Liu, Z. et al. Electrochemical sensor integrated microﬂuidic device for sensitive and simultane- ous quantiﬁcation of dopamine and 5-hydroxytryptamine. Sensors and Actuators B: Chemical 273, 873–883 (2018)

2018
[29]

Ly, S. Y . Detection of dopamine in the pharmacy with a carbon nanotube paste electrode using voltammetry. Bioelectrochemistry 68, 227–231 (2006)

2006
[30]

Orhan, F. et al. Csf dopamine is elevated in ﬁrst-episode psychosis and associates to symptom severity and cognitive performance. Schizophrenia Research 257, 34–40 (2023)

2023
[31]

S., Holmes, C

Goldstein, D. S., Holmes, C. & Sharabi, Y . Cerebrospinal ﬂuid biomarkers of central cate- cholamine deﬁciency in parkinson’s disease and other synucleinopathies. Brain 135, 1900–1913 (2012)

2012
[32]

Cui, M. et al. Highly anti-interference electrocatalytic sensing for dopamine with nitrogen- doped graphdiyne directly in bioﬂuids. Chemical Engineering Journal 464, 142629 (2023)

2023
[33]

N., Vij, V ., Kemp, K

Tiwari, J. N., Vij, V ., Kemp, K. C. & Kim, K. S. Engineered carbon-nanomaterial-based electrochemical sensors for biomolecules. ACS nano 10, 46–80 (2016). 17

2016
[34]

Zhang, S. et al. Discrimination of disaccharide isomers of different glycosidic linkages using a modiﬁed mspa nanopore. Angewandte Chemie e202316766 (2024)

2024
[35]

Masson, J.-F., Biggins, J. S. & Ringe, E. Machine learning for nanoplasmonics. Nature Nanotechnology 18, 111–123 (2023)

2023
[36]

I., Faleiros, M

de Oliveira Filho, J. I., Faleiros, M. C., Ferreira, D. C., Mani, V . & Salama, K. N. Empowering electrochemical biosensors with ai: Overcoming interference for precise dopamine detection in complex samples. Advanced Intelligent Systems 5, 2300227 (2023)

2023
[37]

Rao, Z. et al. Construction of an ecl-dpv dual model biosensor for dopamine detection based on pso-ann algorithm. IEEE Sensors Journal (2024)

2024
[38]

& Das, S

Pal, A., Biswas, S., Chaudhury, K. & Das, S. Paper sensor modiﬁed with mos2 for detection of dopamine using a machine-intelligent web app interface. ACS Applied Materials & Interfaces 15, 43060–43074 (2023)

2023
[39]

& Hugonin, J.-P

Lalanne, P ., Y an, W., Vynck, K., Sauvan, C. & Hugonin, J.-P . Light interaction with photonic and plasmonic resonances. Laser & Photonics Reviews 12, 1700113 (2018)

2018
[40]

& Miroshnichenko, A

Kivshar, Y . & Miroshnichenko, A. Meta-optics with mie resonances. Optics and Photonics News 28, 24–31 (2017)

2017
[41]

& Zhao, S

Wang, Y ., Y ao, H. & Zhao, S. Auto-encoder based dimensionality reduction. Neurocomputing 184, 232–242 (2016). 18

2016
[42]

Theory of the backpropagation neural network

Hecht-Nielsen, R. Theory of the backpropagation neural network. In Neural networks for perception, 65–93 (Elsevier, 1992)

1992
[43]

Lundberg, S. M. et al. From local explanations to global understanding with explainable ai for trees. Nature machine intelligence 2, 56–67 (2020)

2020
[44]

J., Frankel, M

Liu, P .-L., Williams, K. J., Frankel, M. Y . & Esman, R. D. Saturation characteristics of fast photodetectors. IEEE Transactions on microwave theory and techniques 47, 1297–1303 (1999)

1999
[45]

& Fratalocchi, A

Makarenko, M., Wang, Q., Burguete-Lopez, A., Getman, F. & Fratalocchi, A. Ro- bust and scalable ﬂat-optics on ﬂexible substrates via evolutionary neural networks. Advanced Intelligent Systems 3, 2100105 (2021)

2021
[46]

We arrange the set of S1, ...S N projecting surfaces into a matrix of sub-pixel arrays projecting on the camera pixels (Fig. 1g). The image readout from the camera acquires the latent vector z and ﬂattens it into an electric signal, which the regression network processes to pr...
[47]

& Geladi, P

Wold, S., Esbensen, K. & Geladi, P . Principal component analysis. Chemometrics and intelligent laboratory systems 2, 37–52 (1987)

1987
[48]

Makarenko, M. et al. Real-time Hyperspectral Imag- ing in Hardware via Trained Metasurface Encoders. In 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , 12682– 12692 (IEEE, New Orleans, LA, USA, 2022)

2022
[49]

& Baltrusaitis, J

Fairley, N., Bargiela, P ., Huang, W.-M. & Baltrusaitis, J. Principal component analysis (pca) unravels spectral components present in xps spectra of complex oxide ﬁlms on iron foil. Applied Surface Science Advances 17, 100447 (2023). URL https://urldefense.com/v3/__https://ww...

2023
[50]

Spectrax: A straightforward tool for principal component analysis-based spectral analysis

Y u, J.et al. Spectrax: A straightforward tool for principal component analysis-based spectral analysis. Journal of the American Society for Mass Spectrometry 35, 397–400 (2024). PMID: 38217511. 19

2024
[51]

Hamoudi, H. et al. Self-assembly of 1, 4-benzenedimethanethiol self-assembled monolayers on gold. Langmuir 26, 7242–7247 (2010)

2010
[52]

& Y an, Q

Sundararajan, M., Taly, A. & Y an, Q. Axiomatic attribution for deep networks. In International conference on machine learning, 3319–3328 (PMLR, 2017)

2017
[53]

Wu, L.-N. et al. Dopamine sensor based on a hybrid material composed of cuprous oxide hollow microspheres and carbon black. Microchimica Acta 182, 1361–1369 (2015)

2015
[54]

Dass, A. et al. Gold nanoparticles with perﬂuorothiolate ligands. Langmuir 24, 310–315 (2008)

2008
[55]

Y uan, Q. et al. Highly stable and regenerative graphene–diamond hybrid electrochemical biosensor for fouling target dopamine detection. Biosensors and Bioelectronics 111, 117–123 (2018). 20

2018
[56]

& Jha, S

Roychoudhury, A., Basu, S. & Jha, S. K. Dopamine biosensor based on surface functionalized nanostructured nickel oxide platform. Biosensors and Bioelectronics 84, 72–81 (2016)

2016
[57]

Sun, H. et al. Gold nanoparticle-decorated mos 2 nanosheets for simultaneous detection of ascorbic acid, dopamine and uric acid. Rsc Advances 4, 27625–27629 (2014)

2014
[58]

A., Desprez, V ., Kilmartin, P

Xu, G., Jarjes, Z. A., Desprez, V ., Kilmartin, P . A. & Travas-Sejdic, J. Sensitive, selective, disposable electrochemical dopamine sensor based on pedot-modiﬁed laser scribed graphene. Biosensors and Bioelectronics 107, 184–191 (2018)

2018
[59]

In-situ growth of gold nanoparticles on a 3d-network consisting of a mos2/rgo nanocomposite for simultaneous voltammetric determination of ascorbic acid, dopamine and uric acid

Zhao, Y .et al. In-situ growth of gold nanoparticles on a 3d-network consisting of a mos2/rgo nanocomposite for simultaneous voltammetric determination of ascorbic acid, dopamine and uric acid. Microchimica Acta 186, 1–10 (2019)

2019
[60]

Y ang, T. et al. Using poly (m-aminobenzenesulfonic acid)-reduced mos2 nanocomposite synergistic electrocatalysis for determination of dopamine. Sensors and Actuators B: Chemical 249, 451–457 (2017)

2017
[61]

Demirkan, B. et al. Palladium supported on polypyrrole/reduced graphene oxide nanopar- ticles for simultaneous biosensing application of ascorbic acid, dopamine, and uric acid. Scientiﬁc reports 10, 2946 (2020)

2020
[62]

& Ren, H

Cui, X., Fang, X., Zhao, H., Li, Z. & Ren, H. An electrochemical sensor for dopamine based on polydopamine modiﬁed reduced graphene oxide anchored with tin dioxide and gold nanoparticles. Analytical methods 9, 5322–5332 (2017)

2017
[63]

& Chuanuwatanakul, S

Kunpatee, K., Traipop, S., Chailapakul, O. & Chuanuwatanakul, S. Simultaneous determination of ascorbic acid, dopamine, and uric acid using graphene quantum dots/ionic liquid modiﬁed screen-printed carbon electrode. Sensors and Actuators B: Chemical 314, 128059 (2020)

2020
[64]

& Liu, S

Zhang, H. & Liu, S. Electrochemical sensors based on nitrogen-doped reduced graphene oxide for the simultaneous detection of ascorbic acid, dopamine and uric acid. Journal of Alloys and Compounds 842, 155873 (2020). 21

2020
[65]

Xie, W. et al. Label-free and highly selective mofs-based dopamine detection in urine of parkinson’s patients. Chemical Engineering Journal 443, 136371 (2022)

2022
[66]

Ji, W. et al. Carbonized silk fabric-based ﬂexible organic electrochemical transistors for highly sensitive and selective dopamine detection. Sensors and Actuators B: Chemical 304, 127414 (2020)

2020
[67]

Schuster, B. et al. Dopaminergic modulation of dynamic emotion perception. Journal of Neuroscience 42, 4394–4400 (2022)

2022
[68]

E., Okazaki, H., Y agishita, S

Kasai, H., Ziv, N. E., Okazaki, H., Y agishita, S. & Toyoizumi, T. Spine dynamics in the brain, mental disorders and artiﬁcial neural networks. Nature Reviews Neuroscience 22, 407–422 (2021)

2021
[69]

Bido, S. et al. Microglia-speciﬁc overexpression of α -synuclein leads to severe dopaminergic neurodegeneration by phagocytic exhaustion and oxidative toxicity. Nature communications 12, 6237 (2021)

2021
[70]

Mining imaging and clinical data with machine learning approaches for the diagnosis and early detection of parkinson’s disease

Zhang, J. Mining imaging and clinical data with machine learning approaches for the diagnosis and early detection of parkinson’s disease. npj Parkinson’s Disease 8, 13 (2022)

2022
[71]

Cheung, G. et al. Physiological synaptic activity and recognition memory require astroglial glutamine. Nature communications 13, 753 (2022)

2022
[72]

Bangasser, D. A. & Cuarenta, A. Sex differences in anxiety and depression: circuits and mechanisms. Nature Reviews Neuroscience 22, 674–684 (2021). 22

2021
[73]

Propper, D. J. & Balkwill, F. R. Harnessing cytokines and chemokines for cancer therapy. Nature reviews Clinical oncology 19, 237–253 (2022). 23 Acknowledgements. We conducted the hardware encoder’s design using computing resources provided by KAUST Supercomputing Core Lab. We...

2022
[74]

Barbour, A. J. et al. Seizures exacerbate excitatory: inhibitory imbalance in alzheimer’s disease and 5xfad mice. Brain 147, 2169–2184 (2024)

2024
[152]

6◦ with the relative standard deviation (RSD) of 0.36% computed from 10 different measurements, showing super-hydrophobic behavior (Suppl

3 ± 0. 6◦ with the relative standard deviation (RSD) of 0.36% computed from 10 different measurements, showing super-hydrophobic behavior (Suppl. Fig. S3). The CA observed for the FDTS-treated cover glass is 102. 2 ± 0. 5◦ (Suppl. Fig. S4), slightly lower than PFDT but still w...

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.

[1] [1]

Tümay, S. O. & Y e¸ silot, S. Small molecule based water-soluble ﬂuorescence material for highly selective and ultra-sensitive detection of tnt: Design and spectroﬂuorimetric determination in real samples. Sensors and Actuators B: Chemical 343, 130088 (2021)

work page 2021

[2] [2]

C., Ben-Jaber, S

To, K. C., Ben-Jaber, S. & Parkin, I. P . Recent Developments in the Field of Explosive Trace Detection. ACS Nano 14, 10804–10833 (2020)

work page 2020

[3] [3]

Dai, L. et al. Ultrasensitive Physical, Bio, and Chemical Sensors Derived from 1-, 2-, and 3-D Nanocellulosic Materials. Small 16, 1906567 (2020)

work page 2020

[4] [4]

& Bischof, J

Liu, Y ., Zhan, L., Qin, Z., Sackrison, J. & Bischof, J. C. Ultrasensitive and Highly Speciﬁc Lateral Flow Assays for Point-of-Care Diagnosis. ACS Nano 15, 3593–3611 (2021)

work page 2021

[5] [5]

Kang, H. et al. Ultrasensitive Detection of SARS-CoV -2 Antibody by Graphene Field-Effect Transistors. Nano Letters 21, 7897–7904 (2021)

work page 2021

[6] [6]

Mejía-Salazar, J. R. & Oliveira, O. N. J. Plasmonic Biosensing. Chemical Reviews 118, 10617–10625 (2018)

work page 2018

[7] [7]

Sun, Z. et al. Skin-like ultrasensitive strain sensor for full-range detection of human health monitoring. ACS applied materials & interfaces 12, 13287–13295 (2020)

work page 2020

[8] [8]

Feng, L. et al. Ultrasensitive and highly selective detection of strontium ions. Nature Sustainability 1–8 (2023). 14

work page 2023