Pith. sign in

REVIEW 3 major objections 6 minor 36 references

Label-free SERS Discrimination of Proline from Hydroxylated Proline at Single-molecule Level Assisted by a Deep Learning Model

T0 review · 3 major / 6 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read A plasmonic nanopore and a 1D convolutional network can tell proline from hydroxyproline one molecule at a time, with 96.6% accuracy.

desk verdict A real experimental effort on single-molecule SERS discrimination of proline from hydroxyproline, but the 96.6% accuracy is confounded by device identity and needs a null control before the claim is established. read the letter →

arxiv 2412.18935 v1 pith:E22EZD6H submitted 2024-12-25 physics.chem-ph cs.LGphysics.bio-ph

Yingqi Zhao , Kuo Zhan , Pei-Lin Xin , Zuyan Chen , Shuai Li , Francesco De Angelis , Jianan Huang This is my paper · ORCID
classification physics.chem-phcs.LGphysics.bio-ph
keywords label-freeSERSsingle-moleculedetectionprolinevshydroxyprolineplasmonicnanoporeparticle-in-poresensor1Dconvolutionalneuralnetworkpeak-occurrencefrequencyhistogrampost-translationalmodification
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 aims to establish that proline and hydroxyproline can be told apart one molecule at a time, without labels, by the Raman light they scatter inside a gold nanopore. The authors aggregate thousands of flickering single-molecule spectra into peak-occurrence-frequency histograms and feed them to a one-dimensional convolutional neural network, which separates the two amino acids with 96.6% accuracy. Because proline hydroxylation stabilizes collagen and tags proteins for degradation, a single-molecule readout of this modification would matter for early disease detection and post-translational-modification analysis. The paper also claims this is the first single-molecule discrimination of a hydroxylated versus non-hydroxylated amino acid by surface-enhanced Raman spectroscopy.

What carries the argument

Particle-in-pore plasmonic nanopore sensor: a gold nanoparticle is trapped next to a nanopore side wall, creating a sub-nanometre SERS hot spot in which only a few angstroms of a molecule are excited at a time. Peak-occurrence-frequency histogram: a count, over all effective spectra, of how often a Raman peak appears at each wavenumber, used to suppress intensity noise and recover stable band positions. One-dimensional convolutional neural network: the classifier that learns to separate the two molecules from the 1463-point spectra and tolerates leftover citrate. One-dimensional gradient-weighted feature visualization: the interpretability step that maps which spectral positions drive the classification and connects them to ring and CH2 vibration modes.

What would settle it

Take the same proline solution and measure it on several different particle-in-pore devices, then train the identical 1D CNN to label which device each spectrum came from. If batch-identity classification approaches the 96.6% accuracy achieved for proline versus hydroxyproline, the reported molecular discrimination is not yet separated from device identity. Conversely, measuring both molecules on the same device in random order and seeing the accuracy hold would confirm the molecular origin of the signal.

Watch

Extended reading notes

Core claim

The authors' central claim is that the hydroxyl group on the proline ring produces a learnable, reproducible pattern in otherwise chaotic single-molecule SERS data. In the particle-in-pore sensor, a 50 nm gold nanoparticle trapped beside a 200 nm pore wall creates an electromagnetic hot spot roughly the size of an amino acid, so each recorded spectrum comes from a part of one molecule and changes as the molecule moves. Counting how often each Raman shift appears across tens of thousands of trapping events turns this chaos into stable histograms with sharp bands, such as four resolved CH2 rocking bands around 825-878 cm-1. After 48 hours of incubation with a monolayer of analyte replaces most of the citrate surfactant, the remaining citrate signal becomes a common background and a 1D CNN trained on these spectra classifies proline versus hydroxyproline with 96.6% accuracy. Gradient-weighted feature maps localize the decision to ring-deformation, CH2 twist, and OH-related bands in the 740-1200 cm-1 region, matching the band shifts observed when the OH group is added.

Load-bearing premise

The load-bearing assumption is that the spectral differences the CNN learns come from the hydroxyl group itself and not from stable differences between the separate sensor devices used for the two compounds; the paper does not include a same-molecule, different-device control.

Editorial extensions

If this is right

  • A hydroxylation mark, one of the smallest post-translational modifications, becomes detectable without labels or amplification.
  • The histogram representation lets a classifier use thousands of noisy single-molecule spectra instead of needing one clean average spectrum.
  • Reducing citrate by 48-hour monolayer incubation is enough to keep the classification above 96%, and even incomplete substitution leaves enough molecular information for above-77% accuracy.
  • The feature-weight map points to ring and CH2 vibration bands whose shifts report the added OH group, offering a spectral basis for later site-specific PTM analysis.

Reading between the lines

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

  • Not tested here: because the two molecules were measured in different devices, the 96.6% figure should be re-checked with interleaved devices before it is read as a purely molecular metric.
  • The histogram-plus-CNN recipe could be pointed at harder siblings, such as 3-hydroxy versus 4-hydroxyproline or proline methylation, to test whether the learned bands are truly specific to the added OH group.
  • A practical extension would be counting assigned trapping events to estimate the hydroxylation fraction in a mixed sample, turning a binary classifier into a PTM-ratio assay.
  • The single-molecule assumption inherited from earlier particle-in-pore work could be checked directly here by diluting the analyte until spectra disappear and verifying that the remaining spectra still classify correctly.
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.

Referee Report

3 major / 6 minor

Summary. The paper reports a label-free surface-enhanced Raman spectroscopy (SERS) method for discriminating proline from hydroxyproline at the single-molecule level using a particle-in-pore plasmonic sensor and a one-dimensional convolutional neural network (1D CNN). The authors use peak-occurrence-frequency histograms to characterize spectral fluctuations and to monitor citrate substitution on gold nanoparticles, and they train a CNN on 19,000 effective spectra (with 500 spectra for post-evaluation) to classify the two amino acids. They report 96.6% held-out accuracy for the 48 h monolayer condition, 77.9% and 85.3% accuracy for a 72 h 1/8-monolayer post-evaluation set, and below 70% accuracy for a 24 h 1/8-monolayer set. They also apply 1D gradient-weighted feature visualization to localize Raman bands attributed to the hydroxyl group. The central claim is that this is the first demonstration of single-molecule discrimination of an amino acid with and without hydroxylation.

Significance. If the claims are correct, the combination of the particle-in-pore SERS platform, peak-occurrence-frequency histograms, and 1D CNN would constitute a notable advance: label-free optical discrimination of a challenging post-translational modification (hydroxylation) at the single-molecule level. The manuscript reports quantitative performance with cross-condition evaluation and openly discusses the limitation posed by residual citrate interference, which is commendable. However, the central claim is currently undermined by a device-identity confound: proline and hydroxyproline were measured on separate devices, so the CNN may be learning device-specific baselines rather than molecular structure. The single-molecule premise is also inherited from prior work rather than demonstrated here. The approach is potentially valuable, but the evidence as presented does not yet establish that the discrimination is due to the hydroxyl group.

major comments (3)
  1. [Experimental Section, Raman measurement] The statement "Proline and hydroxyproline were tested in different devices. A new sensor device was used for each AuNP incubation time" perfectly confounds analyte identity with device identity. Because each input spectrum carries the full 1463-feature baseline and pore-response information, the CNN could achieve high accuracy by recognizing device-specific baselines, AuNP batch, pore geometry, or laser alignment rather than the molecular difference from the hydroxyl group. The cross-condition post-evaluation (Figure 5(c,d)) does not remove this confound; it only demonstrates that the device-specific signal is consistent across fabrication/incubation batches. The citrate-substitution control (Figure 4) and the gradient feature visualization (Figure 5(e)) are suggestive, but both could also track device/batch variation. A null experiment is required—for example, classifying spectra of the same molecule measured on two different devices, or training on one device and testing on another—before the 96.6% accuracy can be attributed to hydroxylation.
  2. [Experimental Section, Attachment of amino acid on AuNPs; Raman Data processing] The central claim that the discrimination is at the single-molecule level is not supported by evidence in this manuscript. The text states that the protocol from refs 16 and 17 ensures that "only one molecule occupies the hot spot," but no single-molecule validation (e.g., intensity blinking statistics, concentration-dependent event rate, or comparison with multi-molecule SERS) is provided for proline and hydroxyproline under the exact conditions used here. Additionally, the criteria for selecting "effective spectra" (peak intensity threshold 0.07) and the "peak assignment selected spectra" used for CNN training are not fully defined; if selection depends on the presence of certain peaks, it could bias the data toward particular molecular orientations or events. Please specify the selection criteria and provide evidence that each spectrum corresponds to a single molecule.
  3. [CNN Model] The description of the CNN training and evaluation is ambiguous: "A total of 19000 and 500 peak assignment selected spectra with 1463 features were classified using 5-fold cross-validation for the CNN classification model and post-evaluation model, where 80% of the spectra were used as the training set and 20% of the spectra were used as the test set." It is unclear how many spectra were used for the main model versus the post-evaluation model, how the 5-fold cross-validation relates to the 80/20 split, what "peak assignment selected spectra" means, and how many spectra per class were in each set. Since the central quantitative claim is the 96.6% accuracy, the exact data partitioning and selection steps must be stated unambiguously for the results to be reproducible.
minor comments (6)
  1. [Abstract] The abstract contains a typo: "Discriminating the low-abundance hydroxylated proline from hydroxylated proline" should read "from proline."
  2. [Results and Discussion, Figure 5] The text refers to "Figure 5 (b)" for the feature weight comparison, but the figure caption lists "(e)" for that panel; the panel references should be reconciled.
  3. [Results and Discussion, deep learning analysis] The phrase "We made the temptation to apply it" should be "We made the attempt to apply it."
  4. [Conclusions] In the Conclusions, "small Raman cross-session" should be "small Raman cross-section."
  5. [Results and Discussion, Histogram section] The sentence "The number of spectra events contained in the histograms for proline and hydroxyproline 11002 and 9769 respectively" is missing a verb; it should read "... are 11,002 and 9,769, respectively."
  6. [Figure 5 caption] The terminology "1-CNN" appears in the Figure 5 caption and should be "1D CNN" for consistency with the rest of the text.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the 96.6% accuracy is a held-out empirical measurement with independent post-evaluation, and the self-citations provide independent prior support rather than a derivation collapse.

full rationale

The paper's central quantitative claim is a measured classification accuracy (96.6% on a held-out test split, with 77.9% and 85.3% on post-evaluation data from a different incubation condition), not a derived quantity. The CNN is trained on labelled spectra and evaluated on spectra not used in training; the labels are the molecular identities, and the accuracy is an empirical outcome rather than an identity. No equation in the manuscript defines proline/hydroxyproline discrimination in terms of the CNN output, nor is any fitted parameter renamed as a prediction. The peak-occurrence histograms are descriptive pre-processing, and the gradient-weighted feature weights are post-hoc interpretation; neither enters the accuracy calculation by construction. The main self-citations (refs 16 and 17) establish the single-molecule sensitivity of the particle-in-pore platform and the incubation protocol; they are prior published experiments, independent of the current dataset, and they do not by themselves force the proline/hydroxyproline classification result. The acknowledged limitation that citrate substitution is incomplete is an experimental uncertainty, not a circular step. The device/batch confounding (proline and hydroxyproline measured on different devices) is a real experimental-design risk, but it is not a form of circularity under the definitions used here, because it does not make the prediction equivalent to its input. Verdict: no significant circularity; score 0.

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

The paper introduces no new physical entities. Its claims rest on four domain assumptions imported from prior literature or the authors' own platform work, plus four analysis parameters (peak threshold, monolayer concentration, CNN hyperparameters, effective-spectrum criteria) whose values or audit trail are partly confined to the missing Supporting Information.

free parameters (4)
  • Peak detection intensity threshold = 0.07 (normalized intensity)
    Hand-set threshold to separate real peaks from noise; determines which spectra count as 'effective' and therefore shapes the frequency histograms and the CNN training set.
  • Analyte incubation concentration for monolayer coverage = Not stated numerically; computed from molecular area and AuNP surface area (ref 19)
    Sets the premise of one molecule per hotspot; if the coverage estimate is wrong, the single-molecule interpretation fails.
  • 1D CNN architecture and hyperparameters = Not specified in main text (in Supporting Information)
    Layer count, kernel sizes, learning rate, and the 80/20 split determine the reported accuracy; not auditable from the arXiv version.
  • Effective-spectrum selection criteria = Spectra with at least one detected peak above threshold
    The discarded fraction (raw versus effective counts, SI Table S2) affects both the histograms and the model; the counts are not given in the arXiv version.
assumptions (4)
  • domain assumption Particle-in-pore platform provides single-molecule SERS sensitivity
    The 'single-molecule level' claim is inherited from the authors' own refs 16 and 17 and not re-demonstrated in this paper; invoked in the Introduction and Results (hotspot of 'several angstroms' so that 'usually only part of the molecule is located in the hot spot').
  • domain assumption Monolayer coverage limits the hotspot to one molecule at a time
    The text states that with monolayer capping 'only one molecule occupies the hot spot and generates a single-molecule SERS signal' (Experimental Section); no single-event statistics or dilution series support this in the present data.
  • domain assumption Peak positions fluctuate around a stable central value
    The histogram approach assumes the central Raman position survives statistical averaging: 'the central position can be extracted statistically with sufficient spectra' (Results, Histogram section).
  • domain assumption Residual citrate acts as a common feature the CNN can ignore
    The authors concede citrate cannot be fully removed and treat it as 'a common feature' for the CNN (Results, Citrate substitution); the accuracy drop on citrate-rich data (below 70% for 24h 1/8 monolayer) shows the limits of this assumption.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Label-free SERS Discrimination of Proline from Hydroxylated Proline at Single-molecule Level Assisted by a Deep Learning Model." pith.science (2026). https://pith.science/paper/E22EZD6H

@misc{pith2026241218935,
  author       = {Pith},
  title        = {Pith review of: Label-free SERS Discrimination of Proline from Hydroxylated Proline at Single-molecule Level Assisted by a Deep Learning Model},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/E22EZD6H}},
  note         = {Machine review of arXiv:2412.18935}
}
read the original abstract

Discriminating the low-abundance hydroxylated proline from hydroxylated proline is crucial for monitoring diseases and eval-uating therapeutic outcomes that require single-molecule sensors. While the plasmonic nanopore sensor can detect the hydrox-ylation with single-molecule sensitivity by surface enhanced Raman spectroscopy (SERS), it suffers from intrinsic fluctuations of single-molecule signals as well as strong interference from citrates. Here, we used the occurrence frequency histogram of the single-molecule SERS peaks to extract overall dataset spectral features, overcome the signal fluctuations and investigate the citrate-replaced plasmonic nanopore sensors for clean and distinguishable signals of proline and hydroxylated proline. By ligand exchange of the citrates by analyte molecules, the representative peaks of citrates decreased with incubation time, prov-ing occupation of the plasmonic hot spot by the analytes. As a result, the discrimination of the single-molecule SERS signals of proline and hydroxylated proline was possible with the convolutional neural network model with 96.6% accuracy.

Figures

Figures reproduced from arXiv: 2412.18935 by the authors.

Figure 1
Figure 1. (a) shows the schematic illustration of the particle￾in-pore system. It consists of a gold nanohole array with a 200nm diameter drilled on a 100nm gold film supported on a free-standing SiN membrane with a silicon frame. The SEM image of a typical nanopore is shown in [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Waterfall plot of typical Raman spectra of (a)proline and [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Distribution histograms of Normalized peak oc [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: (a)The normalized peak occurring frequency gen [PITH_FULL_IMAGE:figures/full_fig_p005_4.png]
Figure 5
Figure 5. Figure 5: Confusion matrix of the prediction accuracy for [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

36 extracted references · 31 canonical work pages

  1. [1]

    Frontiers in Oncol- ogy, 2017

    Iommarini, L., et al., Non-Canonical Mechanisms Regulat- ing Hypoxia-Inducible Factor 1 Alpha in Cancer. Frontiers in Oncol- ogy, 2017. 7

    work page 2017

  2. [2]

    Han, and B

    Ma, H., X.X. Han, and B. Zhao, Enhanced Raman spectro- scopic analysis of protein post-translational modifications. TrAC Trends in Analytical Chemistry, 2020. 131: p. 116019

    work page 2020

  3. [3]

    International Journal of Analytical Chemistry,

    Ponomarenko, E., et al., The Size of the Human Proteome: The Width and Depth. International Journal of Analytical Chemistry,

    work page

  4. [4]

    Rodriguez-Larrea, and H

    Rosen, C., D. Rodriguez-Larrea, and H. Bayley, Single-mol- ecule site-specific detection of protein phosphorylation with a na- nopore. Nature Biotechnology, 2014. 32(2): p. 179-181

    work page 2014

  5. [5]

    J Am Chem Soc, 2022

    Ensslen, T., et al., Resolving Isomeric Posttranslational Modifications Using a Biological Nanopore as a Sensor of Molecular Shape. J Am Chem Soc, 2022. 144(35): p. 16060-16068

    2022

  6. [6]

    Biomaterials,

    Zhang, L., et al., Detection of single peptide with only one amino acid modification via electronic fingerprinting using reengi- neered durable channel of Phi29 DNA packaging motor. Biomaterials,

    work page

  7. [7]

    Fahie, M.A. and M. Chen, Electrostatic Interactions be- tween OmpG Nanopore and Analyte Protein Surface Can Distinguish between Glycosylated Isoforms. J Phys Chem B, 2015. 119(32): p. 10198-206

    work page 2015

  8. [8]

    2023: bioRxiv

    Chan Cao , et al., Deep learning-assisted single-molecule detection of protein post-translational modifications with a biological nanopore. 2023: bioRxiv. 8

    work page 2023

Show all 36 references
  1. [9]

    Advanced Functional Materials, 2023

    Iarossi, M., et al., High-Density Plasmonic Nanopores for DNA Sensing at Ultra-Low Concentrations by Plasmon-Enhanced Ra- man Spectroscopy. Advanced Functional Materials, 2023. 33(41)

    2023

  2. [10]

    Crystengcomm, 2018

    Du, J., et al., The role of phosphorylation and dephosphory- lation of shell matrix proteins in shell formation: an in vivo and in vitro study. Crystengcomm, 2018. 20(27): p. 3905-3916

    2018

  3. [11]

    Analytical Chemistry, 2024

    Geddis, A., et al., Label-Free SERS Sensors for Real-Time Monitoring of Tyrosine Phosphorylation. Analytical Chemistry, 2024

    2024

  4. [12]

    Bio- sensors & Bioelectronics, 2015

    Li, Y., et al., Label-free in-situ monitoring of protein tyro- sine nitration in blood by surface-enhanced Raman spectroscopy. Bio- sensors & Bioelectronics, 2015. 69: p. 1-7

    2015

  5. [13]

    Journal of Photochemistry and Photobiology B-Biology, 2018

    Sjöberg, B., et al., Methionine oxidation by hydrogen perox- ide in peptides and proteins: A theoretical and Raman spectroscopy study. Journal of Photochemistry and Photobiology B-Biology, 2018. 188: p. 95-99

    2018

  6. [14]

    Guerrero, A. and R. Aroca, Surface-enhanced Raman scat- tering of hydroxyproline. Journal of Raman Spectroscopy, 2012. 43(4): p. 478-481

    2012

  7. [15]

    A Raman, SERS and theoretical study

    Cárcamo, J., et al., Proline and hydroxyproline deposited on silver nanoparticles. A Raman, SERS and theoretical study. Journal of Raman Spectroscopy, 2012. 43(6): p. 750-755

    2012

  8. [16]

    Nature Communications, 2019

    Huang, J.-A., et al., SERS discrimination of single DNA ba- ses in single oligonucleotides by electro-plasmonic trapping. Nature Communications, 2019. 10(1): p. 5321

    2019

  9. [17]

    An- gewandte Chemie International Edition, 2020

    Huang, J.-A., et al., Multiplexed Discrimination of Single Amino Acid Residues in Polypeptides in a Single SERS Hot Spot. An- gewandte Chemie International Edition, 2020. 59(28): p. 11423-11431

    2020

  10. [18]

    Acs Photonics, 2022

    Zhao, Y., et al., Label-Free Optical Analysis of Biomole- cules in Solid-State Nanopores: Toward Single-Molecule Protein Se- quencing. Acs Photonics, 2022. 9(3): p. 730-742

    2022

  11. [19]

    Plos One, 2013

    Tien, M., et al., Maximum Allowed Solvent Accessibilites of Residues in Proteins. Plos One, 2013. 8(11)

    2013

  12. [20]

    Nature Com- munications, 2020

    Franco-Ulloa, S., et al., Dispersion state phase diagram of citrate-coated metallic nanoparticles in saline solutions. Nature Com- munications, 2020. 11(1)

    2020

  13. [21]

    Buglak, A. and A. Kononov, Comparative study of gold and silver interactions with amino acids and nucleobases. Rsc Advances,

  14. [22]

    Chemical Reviews, 2017

    Zrimsek, A., et al., Single-Molecule Chemistry with Surface- and Tip-Enhanced Raman Spectroscopy. Chemical Reviews, 2017. 117(11): p. 7583-7613

    2017

  15. [23]

    Nano Letters, 2018

    Shin, H., et al., Frequency-Domain Proof of the Existence of Atomic-Scale SERS Hot-Spots. Nano Letters, 2018. 18(1): p. 262-271

    2018

  16. [24]

    Acs Nano, 2024

    Li, W., et al., High-Throughput Single-Molecule Surface- Enhanced Raman Spectroscopic Profiling of Single-Amino Acid Sub- stitutions in Peptides by a Gold Plasmonic Nanopore. Acs Nano, 2024. 18(29): p. 19200-19207

    2024

  17. [25]

    Herlinger, A.W. and T.V. Long, II, Laser-Raman and infra- red spectra of amino acids and their metal complexes. III. Proline and bisprolinato complexes. Journal of the American Chemical Society,

  18. [26]

    Langmuir, 1995

    MUNRO, C., et al., CHARACTERIZATION OF THE SURFACE OF A CITRATE-REDUCED COLLOID OPTIMIZED FOR USE AS A SUBSTRATE FOR SURFACE-ENHANCED RESONANCE RAMAN-SCATTERING. Langmuir, 1995. 11(10): p. 3712-3720

    1995

  19. [27]

    Journal of Physical Chemistry, 1983

    SIIMAN, O., et al., SURFACE-ENHANCED RAMAN- SCATTERING BY CITRATE ON COLLOIDAL SILVER. Journal of Physical Chemistry, 1983. 87(6): p. 1014-1023

    1983

  20. [28]

    Park, J. and J. Shumaker-Parry, Structural Study of Citrate Layers on Gold Nanoparticles: Role of Intermolecular Interactions in Stabilizing Nanoparticles. Journal of the American Chemical Society,

  21. [29]

    Colloids and Sur- faces B-Biointerfaces, 2010

    Majzik, A., et al., Functionalization of gold nanoparticles with amino acid, β-amyloid peptides and fragment. Colloids and Sur- faces B-Biointerfaces, 2010. 81(1): p. 235-241

    2010

  22. [30]

    Jour- nal of Raman Spectroscopy, 2022

    Li, J., et al., Machine learning using convolutional neural networks for SERS analysis of biomarkers in medical diagnostics. Jour- nal of Raman Spectroscopy, 2022. 53(12): p. 2044-2057

    2022

  23. [31]

    Analytical Methods, 2023

    Lu, P., et al., CNN-assisted SERS enables ultra-sensitive and simultaneous detection of Scr and BUN for rapid kidney function as- sessment. Analytical Methods, 2023. 15(3): p. 322-332

    2023

  24. [32]

    Analytical Chemistry, 2023

    Shi, G., et al., 1D Gradient-Weighted Class Activation Map- ping, Visualizing Decision Process of Convolutional Neural Network- Based Models in Spectroscopy Analysis. Analytical Chemistry, 2023. 95(26): p. 9959-9966

    2023

  25. [33]

    Analytical and Bioanalytical Chemistry, 2023

    Xiong, C., et al., Rapid and precise detection of cancers via label-free SERS and deep learning. Analytical and Bioanalytical Chemistry, 2023. 415(17): p. 3449-3462

    2023

  26. [1970]

    6481-6486

    92(22): p. 6481-6486

  27. [2014]

    1907-1921

    136(5): p. 1907-1921

    1907

  28. [2020]

    34149-34160

    10(56): p. 34149-34160

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.