Pith. sign in

REVIEW 4 major objections 6 minor 38 references

Automated Phytosensing: Ozone Exposure Classification Based on Plant Electrical Signals

T0 review · 4 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 generic automated toolchain classifies plant ozone exposure from electrical signals with up to 94.6% accuracy.

desk verdict The 94.6% 'unseen data' accuracy is a validation-set number; the paper's own held-out tests land at 57-77%, so the abstract overstates the result, but the toolchain and honest discussion of meta-overfitting make it worth refereeing. read the letter →

arxiv 2412.13312 v1 pith:IS63VJVL submitted 2024-12-17 cs.LG eess.SP

Till Aust , Eduard Buss , Felix Mohr , Heiko Hamann This is my paper · ORCID
classification cs.LGeess.SP
keywords phytosensingozonedetectionplantelectrophysiologyautomatedmachinelearningtsfreshfeatureextractionairqualitymonitoringHederahelixelectricaldifferentialpotential
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 sets out to prove that a living plant can be turned into an automated ozone detector without hand-crafted features or manual model selection. The authors record electric differential potentials from ivy plants inside a Faraday cage, expose the plants to ten-minute ozone pulses, and feed the signals into a toolchain that extracts 787 generic time-series features per slice, searches for a classification pipeline with automated machine learning, and prunes the features with forward selection. They report separating ozone-exposed from non-exposed ten-minute windows with accuracies up to 94.6% and ROC AUC values up to 0.9985 on the analysis split, and lower held-out test accuracies (76.96% leaf, 77.30% combined, 57.21% stem) that they attribute to meta-overfitting. If these claims hold, pollutant monitoring in cities could be built from dense networks of low-cost plant-based sensors rather than sparse, expensive stations.

What carries the argument

The load-bearing mechanism is a three-stage toolchain applied to 10-minute slices of electrical differential potential. First, the tsfresh library computes 787 generic time-series features per slice, with background subtraction that removes the features of a slice taken 10 to 20 minutes before the stimulus to reduce inter-plant variance. Second, Naive AutoML greedily searches over preprocessing steps and scikit-learn classifiers, optimizing either ROC AUC or accuracy with five stratified 80/20 validation splits. Third, a semi-greedy forward feature selection iteratively adds the feature that most improves ROC AUC while keeping the $n$ best candidate sets per feature-count to allow temporary suboptimality; the selected 62 to 94 features are what the final models use.

What would settle it

Run a sham-exposure control in which the ozone generator and air circulation are activated for ten minutes without producing ozone, on the same schedule. If a classifier trained on real-ozone versus pre-exposure baselines also separates sham-exposure from pre-exposure windows with comparable accuracy, the method is detecting the procedure rather than ozone.

Watch

Extended reading notes

Core claim

The central claim, in the paper's own terms, is that a generic automatic toolchain can classify whether a plant is being exposed to ozone from the electric differential potential recorded by two needle electrodes. The pipeline computes 787 tsfresh features per 10-minute signal slice, uses Naive AutoML to select and tune a classifier, and then applies a semi-greedy forward feature selection that keeps several candidate feature sets per size. On the analysis (validation) split, feature selection lifts ROC AUC to 0.9901 for the leaf, 0.9063 for the stem, and 0.9985 for the combined dataset, with accuracies of 94.60%, 82.64%, and 89.34% respectively. On the held-out 20% test split the selected-feature models reach ROC AUC 0.8790 (leaf), 0.6665 (stem), and 0.9111 (combined), with accuracies of 76.96%, 57.21%, and 77.30%; the authors explain the drop as meta-overfitting of the pipeline. They also report 95.39% accuracy for a wind-versus-no-wind classification on ZZ plants measured with a different phytosensor, which they present as evidence of generalizability.

Load-bearing premise

The experiment assumes that the ten minutes immediately before each scheduled ozone pulse are a clean, ozone-free resting baseline, meaning two hours of recovery fully clears the previous dose, the plant is not anticipating the procedure, and the ozone generator and airflow themselves do not alter the electrical signal.

Editorial extensions

If this is right

  • A new plant species or stimulus can be turned into a classifier by rerunning the same three-stage pipeline, without choosing features or models by hand.
  • Leaf measurements carry more ozone information than stem measurements, so sensor placement near leaves should be prioritized in future phytosensing devices.
  • Combining leaf and stem signals is at least as accurate as leaf alone, so multi-point measurement is a safe design choice.
  • The low-cost PhytoNode hardware records enough signal detail for classification, supporting the feasibility of dense urban monitoring networks.
  • Performance on held-out data is substantially below validation performance, so pipelines trained on small datasets should expect a meta-overfitting penalty when deployed.

Reading between the lines

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

  • The reported 94.6% figure is an analysis-split number; a deployed monitor should be benchmarked against the held-out numbers, roughly 77% accuracy, until the pipeline is re-validated in situ.
  • Because the pre-exposure 10-minute window is used as the non-ozone class, a sham-exposure control is needed to rule out the alternative that the classifier learns the generator start or the airflow pattern rather than ozone physiology.
  • The generic tsfresh features are not physiologically interpretable, so the toolchain cannot by itself say which plant electrical response carries the ozone information; targeted experiments would be needed to close that gap.
  • The peak exposure of about 1,447 ppb in these experiments is far above typical urban ozone levels near health thresholds, so real-world accuracy at 60 to 200 ppb remains an open empirical question that the paper's outlook already identifies.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 6 minor

Summary. The paper proposes an automated toolchain for classifying ozone exposure in ivy (Hedera helix) from plant electrical differential potentials. The pipeline uses tsfresh to extract 787 features per 10-minute signal slice, Naive AutoML to select and optimize classifiers, and a forward feature selection procedure to reduce the feature set. Experiments cover leaf, stem, and combined electrode placements, with an additional claim that the same toolchain transfers to wind detection in a different plant species. The headline result is a classification accuracy of up to 94.6% on unseen data, but the paper's own held-out test results are substantially lower (76.96% Leaf, 77.30% Combined, and 57.21% Stem with selected features).

Significance. If the reported performance were supported, the paper would make a useful practical contribution: an automated, species- and stimulus-generic pipeline for phytosensing that avoids hand-crafted features and manual model selection. Strengths include public release of the code, use of generic time-series features, a reasonable choice of AutoML framework, and an unusually candid discussion of meta-overfitting and learning-curve variance in Sections III-B and III-C. However, two load-bearing problems currently block the stated significance: the '94.6% on unseen data' figure is a validation result, not a held-out test result, and the no-ozone baseline is temporally confounded with the exposure procedure because no sham-exposure control is described. The generalizability claim in Section IV is also reported without sufficient detail to be evaluated. The toolchain may still be valuable at the honestly reported lower accuracy or as a methodological template, but the manuscript as written overstates what is established.

major comments (4)
  1. [Abstract; Section III-B; Section III-C] The abstract's central claim of 'accuracies of up to 94.6% on unseen data' is not supported by the paper's own evaluation. The 94.6% value appears in Section III-B as the average validation accuracy of the best forward-selected feature subset (62 leaf features) over 100 random 80/20 splits of the analysis dataset, evaluated during model and feature selection. Section III-C reports the held-out test accuracies with the selected features as 76.96% for Leaf, 77.30% for Combined, and 57.21% for Stem, with 57.21% near chance for a balanced two-class problem. The abstract and the concluding paragraph in Section III-C must be rewritten to report the test-set performance, or the evaluation must be redone with a nested/selection-aware protocol so that the reported number is a true unbiased estimate.
  2. [Section II-A] The no-ozone class is defined as the 10 minutes immediately before each scheduled ozone exposure, and no sham-exposure control is described. Because the positive and negative slices are adjacent in a repeating 2-hour cycle, the classifier can learn temporal position within the cycle, recovery from the previous ozone dose, or procedural artifacts from the ozone generator and airflow rather than a physiological response to ozone. This is not a minor detail: it directly undermines the attribution of the classification to ozone exposure. A sham-exposure condition in which the generator/fan is operated without ozone would be the standard control. At minimum, the manuscript must explicitly state this as a limiting condition and soften the causal claim that the model detects 'ozone exposure'.
  3. [Section II-A; Section III-C] The random 80/20 split of samples across plants places the ozone and pre-ozone slices of the same exposure episode into both training and test partitions. Since these slices are 10 minutes apart and come from the same plant, the classifier can exploit strong temporal correlation and plant-specific signal baselines, inflating the reported test accuracy. To support the claim that the method generalizes to unseen data, the authors should report an evaluation using leave-one-exposition-out or leave-one-plant-out splits, or at least analyze how performance changes when adjacent slices are kept together. This is a load-bearing issue for the 'unseen data' wording.
  4. [Section IV] The claimed transfer result of 95.39% accuracy for wind detection in ZZ plants is presented without the experimental details needed for assessment: no dataset size, no number of exposures, no description of preprocessing or feature extraction, no split protocol, and no statement of whether this is a test-set or validation number. Since 'our approach can be used for other plant species and stimuli' is one of the three bullet contributions in the introduction, this result cannot be left as an unreviewable aside; either full details must be provided or the claim should be removed or explicitly labeled preliminary.
minor comments (6)
  1. [Section II-B] The ±200 mV cutoff is described as removing 'physically illogical values,' but no justification or count of removed samples is given; please report the fraction of removed measurements and the sensitivity of the results to this threshold.
  2. [Table I; Figure 3; Section III-C] The word 'accuracy' is used for both validation and test results without consistent labeling; for example, Table I and Figure 3 report analysis-data validation scores while Section III-C reports test scores. Add explicit labels such as 'validation' and 'test' in all captions and in the text.
  3. [Section III-A; Figure 4] The threshold model's maximum accuracy of 70.83% should be clearly labeled as a validation result on the analysis data, since the paragraph currently reads as if it were a standalone result without indicating the split.
  4. [Section II-C] The sentence 'ROC AUC captures this uncertainty' is imprecise; ROC AUC measures ranking quality across classification thresholds, not predictive uncertainty. Please rephrase.
  5. [Figure 2] The y-axis is labeled 'EDP [normalized]' while the feature extraction is performed in mV; please clarify whether normalization is applied to the data used for feature extraction and, if so, where it enters the pipeline.
  6. [Section III-B] The phrase 'This maybe counter-intuitive finding' contains a typo and should read 'This may be counter-intuitive'.

Circularity Check

1 steps flagged · score 6.0 of 10

The 94.6% 'unseen data' accuracy is the best validation score from the forward feature-selection loop, not the held-out test result; the paper's own test accuracies are 76.96% (Leaf), 77.30% (Combined), and 57.21% (Stem), so the headline number reduces to the fitting process.

  1. fitted input called prediction [Abstract; Sec. III-B; Sec. III-C]
    "We show that our approach successfully classifies plant ozone exposure with accuracies of up to 94.6% on unseen data. ... The best score for the Leaf (ROC AUC = 0.9901, ACC = 94.60%), Stem (ROC AUC = 0.9063, ACC = 82.64%), and Combined analysis dataset (ROC AUC = 0.9985, ACC = 89.34%) is reached using 62, 69, and 94 features, respectively. All results are averaged over 100 independent repetitions (analysis data were randomly split into 80%/20% training and validation data)."

    The 94.6% is the maximum accuracy among feature subsets evaluated on 100 random 80/20 validation splits of the analysis dataset, and the same validation procedure drives forward feature selection (Sec. II-D). Hence the number is a selected optimum of the fitting/selection loop, not a measure on data untouched by that loop. The paper's held-out evaluation (Sec. III-C) gives Leaf 76.96%, Combined 77.30%, and Stem 57.21% with the selected features, and attributes the gap to 'meta overfitting'. Presenting the validation-selected 94.6% as 'on unseen data' is therefore a fitted input called prediction: by construction it is the best of the selection runs, so it does not independently support the generalization claim.

full rationale

The paper contains no equation-level derivation, so no self-definitional or uniqueness-imported circularity is present. The main circularity-adjacent issue is a single, load-bearing reporting choice: the abstract's headline 'accuracies of up to 94.6% on unseen data' is taken from the analysis-data validation phase (Sec. III-B), where the 94.6% is the best accuracy over feature subsets and random splits that were themselves used to select those features and pipelines. This makes the headline an in-sample selected optimum, not an unbiased prediction; the paper's own held-out test results are considerably lower. The self-citation to Buss et al. [22] for background subtraction and for the ZZ-plant generalization experiment is not itself circular: the ZZ data are separate measurements, and the generalization result (95.39% for wind/no-wind) is reported as a new application of the toolchain rather than imported as a theorem. The 'no ozone' class being defined as the 10 minutes before exposure is a possible confound (procedural/temporal artifacts) but is not a definitional equivalence and is better treated as a validity concern than as circularity. Overall, the central quantitative claim as stated reduces to the validation/fitting process, while the paper does provide an honest, lower test-set estimate, yielding partial circularity with independent content.

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

The central claim rests on experimental design choices: the no-ozone baseline is the pre-exposure window, background subtraction assumes stable non-stimulus periods, and the ozone dose is fixed at a high concentration. Feature subset sizes are selected from the analysis data. No new entities are introduced. The main burden is confounding control, not mathematical circularity.

free parameters (5)
  • Best feature subset size = 62 (leaf), 69 (stem), 94 (combined)
    Selected by maximizing ROC AUC on the analysis dataset (Sec III-B); the value is a validation-time choice, not a fixed scientific constant.
  • Ozone exposure concentration = 1,447 ppb ± 376 ppb
    Chosen exposure level (Sec II-A); roughly 24x the EU target value, so the classifier's operating point is far from ambient urban ozone.
  • Recovery time between exposures = 2 hours
    Set in Sec II-A; if recovery is incomplete, the no-ozone baseline may include residual ozone response.
  • Background subtraction window = 10-20 minutes before stimulus
    Chosen in Sec II-B to compute feature_sub; assumes this window is an ozone-free stable reference.
  • Preprocessing constants = rolling median window 10, downsampling 2 Hz, ±200 mV cutoff
    Hand-chosen in Sec II-B; no sensitivity analysis is provided.
assumptions (5)
  • domain assumption The pre-exposure window is a clean no-ozone resting state.
    Sec II-A defines the no-ozone class; plants may still be recovering from previous exposure and the exposure apparatus may change state before stimulation.
  • domain assumption Background subtraction removes inter-plant variance without removing stimulus-relevant signal.
    Sec II-B uses feature_sub = feature_target - feature_background; if slow circadian drift or recovery trends differ between the target and background windows, this subtraction biases the features.
  • ad hoc to paper The ±200 mV cutoff only removes physically illogical values.
    Sec II-B; no physiological calibration is given for this threshold.
  • domain assumption tsfresh features are plant- and stimulus-generic enough to transfer across species.
    Sec II-B justifies tsfresh as generic; the only cross-species evidence is a summarized ZZ-plant wind result in Sec IV.
  • domain assumption The classifier learns plant physiology rather than experimental artifacts.
    No sham-exposure control is described (Sec II-A); generator operation, airflow, or electrode settling could be confounded with ozone.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Automated Phytosensing: Ozone Exposure Classification Based on Plant Electrical Signals." pith.science (2026). https://pith.science/paper/IS63VJVL

@misc{pith2026241213312,
  author       = {Pith},
  title        = {Pith review of: Automated Phytosensing: Ozone Exposure Classification Based on Plant Electrical Signals},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/IS63VJVL}},
  note         = {Machine review of arXiv:2412.13312}
}
read the original abstract

In our project WatchPlant, we propose to use a decentralized network of living plants as air-quality sensors by measuring their electrophysiology to infer the environmental state, also called phytosensing. We conducted in-lab experiments exposing ivy (Hedera helix) plants to ozone, an important pollutant to monitor, and measured their electrophysiological response. However, there is no well established automated way of detecting ozone exposure in plants. We propose a generic automatic toolchain to select a high-performance subset of features and highly accurate models for plant electrophysiology. Our approach derives plant- and stimulus-generic features from the electrophysiological signal using the tsfresh library. Based on these features, we automatically select and optimize machine learning models using AutoML. We use forward feature selection to increase model performance. We show that our approach successfully classifies plant ozone exposure with accuracies of up to 94.6% on unseen data. We also show that our approach can be used for other plant species and stimuli. Our toolchain automates the development of monitoring algorithms for plants as pollutant monitors. Our results help implement significant advancements for phytosensing devices contributing to the development of cost-effective, high-density urban air monitoring systems in the future.

Figures

Figures reproduced from arXiv: 2412.13312 by the authors.

Figure 1
Figure 1. Experimental setup to expose an ivy plant to ozone. The ivy plant [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Example data of the first experiment with 17 expositions. The [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Each ROC curve is averaged (solid line) over 500 splits using [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: Accuracy results for a simple threshold model (blue: leaf, orange: [PITH_FULL_IMAGE:figures/full_fig_p004_4.png]
Figure 6
Figure 6. Figure 6: Learning curves for classifiers trained with (blue) 62 features of the [PITH_FULL_IMAGE:figures/full_fig_p005_6.png]
Figure 5
Figure 5. Figure 5: Mean ROC AUC score (solid line) for the all feature feature subsets of [PITH_FULL_IMAGE:figures/full_fig_p005_5.png]
Figure 7
Figure 7. Figure 7: ROC curves for the test data using (a) all features (same setting as [PITH_FULL_IMAGE:figures/full_fig_p006_7.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

38 extracted references · 34 canonical work pages

  1. [1]

    Advances in smart environment monitoring systems using iot and sensors,

    S. L. Ullo and G. R. Sinha, “Advances in smart environment monitoring systems using iot and sensors,” Sensors, vol. 20, no. 11, p. 3113, 2020

    work page 2020

  2. [2]

    A review of urban air pollution monitoring and exposure assessment methods,

    X. Xie, I. Semanjski, S. Gautama, E. Tsiligianni, N. Deligiannis, R. Rajan, F. Pasveer, and W. Philips, “A review of urban air pollution monitoring and exposure assessment methods,” ISPRS International Journal of Geo-Information , vol. 6, no. 12, p. 389, 2017

    work page 2017

  3. [3]

    The changing paradigm of air pollution monitoring,

    E. G. Snyder, T. H. Watkins, P. A. Solomon, E. D. Thoma, R. W. Williams, G. S. Hagler, D. Shelow, D. A. Hindin, V . J. Kilaru, and P. W. Preuss, “The changing paradigm of air pollution monitoring,” Environmental Science & Technology, vol. 47, no. 20, pp. 11 369–11 377, 2013

    work page 2013

  4. [4]

    Biohybrid systems for environmental intelligence on living plants: WatchPlant project,

    L. Garc ´ıa-Carmona, S. Bogdan, A. Diaz-Espejo, M. Dobielewski, H. Hamann, V . Hernandez-Santana, A. Kernbach, S. Kernbach, A. Quijano-L ´opez, N. Roxhed, B. Salamat, and M. Wahby, “Biohybrid systems for environmental intelligence on living plants: WatchPlant project,” in Proceedings of the Conference on Information Technology for Social Good , ser. GoodI...

    work page 2021

  5. [5]

    WatchPlant: Networked bio-hybrid systems for pollution monitoring of urban areas,

    H. Hamann, S. Bogdan, A. Diaz-Espejo, L. Garc ´ıa-Carmona, V . Hernandez-Santana, S. Kernbach, A. Kernbach, A. Quijano-L ´opez, B. Salamat, and M. Wahby, “WatchPlant: Networked bio-hybrid systems for pollution monitoring of urban areas,” in ALIFE 2022: The 2022 Conference on Artificial Life , vol. ALIFE 2021: The 2021 Conference on Artificial Life, 2021, ...

    work page 2022

  6. [6]

    Phytonodes for environmental monitoring: Stimulus classification based on natural plant signals in an interactive energy- efficient bio-hybrid system,

    E. Buss, T.-L. Rabbel, V . Horvat, M. Kriˇzmanˇci´c, S. Bogdan, M. Wahby, and H. Hamann, “Phytonodes for environmental monitoring: Stimulus classification based on natural plant signals in an interactive energy- efficient bio-hybrid system,” in Proceedings of the 2022 ACM Confer- ence on Information Technology for Social Good . ACM, 2022, pp. 258–264

    work page 2022

  7. [7]

    Phyton- ode upgraded: Energy-efficient long-term environmental monitoring using phytosensing,

    E. Buss, T. Aust, O. Hamburger, C. K. Heck, and H. Hamann, “Phyton- ode upgraded: Energy-efficient long-term environmental monitoring using phytosensing,” in Advances in Information and Communication. FICC 2025. Lecture Notes in Networks and Systems . Springer Nature Switzerland, 2025, (in press)

    work page 2025

  8. [8]

    Phytosensors: Harnessing plants to understand the world around us,

    A. C. Pfotenhauer and S. C. Lenaghan, “Phytosensors: Harnessing plants to understand the world around us,” Current Opinion in Biotechnology , vol. 87, p. 103134, 2024

    work page 2024

Show all 38 references
  1. [9]

    Mathematical models of electrical activity in plants,

    E. Sukhova, E. Akinchits, and V . Sukhov, “Mathematical models of electrical activity in plants,” The Journal of Membrane Biology , vol. 250, no. 5, pp. 407–423, 2017

    2017

  2. [10]

    Machine learning: Trends, perspec- tives, and prospects,

    M. I. Jordan and T. M. Mitchell, “Machine learning: Trends, perspec- tives, and prospects,” Science, vol. 349, no. 6245, pp. 255–260, 2015

    2015

  3. [11]

    Deep learning,

    Y . LeCun, Y . Bengio, and G. Hinton, “Deep learning,” nature, vol. 521, no. 7553, pp. 436–444, 2015

    2015

  4. [12]

    Research on classification of water stress state of plant electrical signals based on pso-svm,

    L. Tian, C. Shang, M. Li, and Y . Wang, “Research on classification of water stress state of plant electrical signals based on pso-svm,” IEEE Access, vol. 11, pp. 125 021–125 032, 2023

    2023

  5. [13]

    Time series data modelling for classifi- cation of drought in tomato plants,

    K. Sai, N. Sood, and I. Saini, “Time series data modelling for classifi- cation of drought in tomato plants,” Theoretical and Experimental Plant Physiology, vol. 35, pp. 379–394, 2023

    2023

  6. [14]

    Classification of plant electrophysiology signals for detection of spider mites infestation in tomatoes,

    E. Najdenovska, F. Dutoit, D. Tran, C. Plummer, N. Wallbridge, C. Camps, and L. E. Raileanu, “Classification of plant electrophysiology signals for detection of spider mites infestation in tomatoes,” Applied Sciences, vol. 11, no. 4, 2021

    2021

  7. [15]

    Electrophysiological assess- ment of plant status outside a Faraday cage using supervised machine learning,

    D. Tran, F. Dutoit, E. Najdenovska, N. Wallbridge, C. Plummer, M. Mazza, L. E. Raileanu, and C. Camps, “Electrophysiological assess- ment of plant status outside a Faraday cage using supervised machine learning,” Scientific Reports, vol. 9, no. 1, p. 17073, 2019

    2019

  8. [16]

    Using a one-dimensional convolutional neural network with a conditional generative adversarial network to classify plant electrical signals,

    X.-H. Qin, Z.-Y . Wang, J.-P. Yao, Q. Zhou, P.-F. Zhao, Z.-Y . Wang, and L. Huang, “Using a one-dimensional convolutional neural network with a conditional generative adversarial network to classify plant electrical signals,” Computers and Electronics in Agriculture, vol. 174,...

    2020

  9. [17]

    A deep learning method for the long-term prediction of plant electrical signals under salt stress to identify salt tolerance,

    J.-P. Yao, Z.-Y . Wang, R. F. de Oliveira, Z.-Y . Wang, and L. Huang, “A deep learning method for the long-term prediction of plant electrical signals under salt stress to identify salt tolerance,” Computers and Electronics in Agriculture, vol. 190, p. 106435, 2021

    2021

  10. [18]

    De- tecting stress caused by nitrogen deficit using deep learning techniques applied on plant electrophysiological data,

    D. Gonz ´alez I Jucl`a, E. Najdenovska, F. Dutoit, and L. E. Raileanu, “De- tecting stress caused by nitrogen deficit using deep learning techniques applied on plant electrophysiological data,” Scientific Reports , vol. 13, p. 9633, 2023

    2023

  11. [19]

    Clivia biosensor: Soil moisture identification based on electrophysiol- ogy signals with deep learning,

    J. Qi, C. Liu, Q. Wang, Y . Shi, X. Xia, H. Wang, L. Sun, and H. Men, “Clivia biosensor: Soil moisture identification based on electrophysiol- ogy signals with deep learning,” Biosensors and Bioelectronics, vol. 262, p. 116525, 2024

    2024

  12. [20]

    Automatic classification of plant electrophysiological responses to environmental stimuli using machine learning and interval arithmetic,

    D. R. Pereira, J. P. Papa, G. F. R. Saraiva, and G. M. Souza, “Automatic classification of plant electrophysiological responses to environmental stimuli using machine learning and interval arithmetic,” Computers and Electronics in Agriculture, vol. 145, pp. 35–42, 2018

    2018

  13. [21]

    Abiotic stress classification through spectral analysis of enhanced electrophysiological signals of plants,

    K. Sai, N. Sood, and I. Saini, “Abiotic stress classification through spectral analysis of enhanced electrophysiological signals of plants,” Biosystems Engineering, vol. 219, pp. 189–204, 2022

    2022

  14. [22]

    Stimulus classification with electrical potential and impedance of living plants: Comparing discriminant analysis and deep-learning methods,

    E. Buss, T. Aust, M. Wahby, T.-L. Rabbel, S. Kernbach, and H. Hamann, “Stimulus classification with electrical potential and impedance of living plants: Comparing discriminant analysis and deep-learning methods,” Bioinspiration & Biomimetics , vol. 18, no. 2, p. 025003, 2023

    2023

  15. [23]

    Plant electrophysiol- ogy: Bibliometric analysis, methods and applications in the monitoring of plant-environment interactions,

    S. L. Chaparro-C ´ardenas, J. A. Ramirez-Bautista, W. Gamboa-Contreras, A. L. Moreno-Chac ´on, and F. C. Vargas-Tangua, “Plant electrophysiol- ogy: Bibliometric analysis, methods and applications in the monitoring of plant-environment interactions,” Dyna, vol. 88, no. 218, pp....

    2021

  16. [24]

    Exploring strategies for classification of external stimuli using statistical features of the plant electrical response,

    S. K. Chatterjee, S. Das, K. Maharatna, E. Masi, L. Santopolo, S. Man- cuso, and A. Vitaletti, “Exploring strategies for classification of external stimuli using statistical features of the plant electrical response,” Journal of the Royal Society Interface , vol. 12, no. 104, ...

    2015

  17. [25]

    Comparison of decision tree based classification strategies to detect external chemical stimuli from raw and filtered plant electrical response,

    S. K. Chatterjee, S. Das, K. Maharatna, E. Masi, L. Santopolo, I. Colzi, S. Mancuso, and A. Vitaletti, “Comparison of decision tree based classification strategies to detect external chemical stimuli from raw and filtered plant electrical response,” Sensors and Actuators B: Ch...

    2017

  18. [26]

    Chemical sensing employing plant electrical signal response-classification of stimuli using curve fitting coefficients as features,

    S. K. Chatterjee, O. Malik, and S. Gupta, “Chemical sensing employing plant electrical signal response-classification of stimuli using curve fitting coefficients as features,” Biosensors, vol. 8, no. 3, 2018

    2018

  19. [27]

    Multiclass classification of environmental chemical stimuli from unbalanced plant electrophysiolog- ical data,

    N. Bhadra, S. K. Chatterjee, and S. Das, “Multiclass classification of environmental chemical stimuli from unbalanced plant electrophysiolog- ical data,” PLOS ONE, vol. 18, no. 5, pp. 1–31, 2023

    2023

  20. [28]

    Plant electrical activity analysis for ozone pollution critical level detection,

    M. Dolfi, I. Colzi, S. Morosi, E. Masi, S. Mancuso, E. Del Re, F. Francini, and R. Magliacani, “Plant electrical activity analysis for ozone pollution critical level detection,” in 2015 23rd European Signal Processing Conference (EUSIPCO). IEEE, 2015, pp. 2431–2435

    2015

  21. [29]

    Taking human out of learning applications: A survey on automated machine learning,

    Q. Yao, M. Wang, Y . Chen, W. Dai, Y .-F. Li, W.-W. Tu, Q. Yang, and Y . Yu, “Taking human out of learning applications: A survey on automated machine learning,” arXiv preprint arXiv:1810.13306, vol. 31, 2018

    2018 arXiv

  22. [30]

    Efficient and robust automated machine learning,

    M. Feurer, A. Klein, K. Eggensperger, J. Springenberg, M. Blum, and F. Hutter, “Efficient and robust automated machine learning,” in Advances in Neural Information Processing Systems 28, 2015, pp. 2962– 2970

    2015

  23. [31]

    Auto-sklearn 2.0: Hands-free automl via meta-learning,

    M. Feurer, K. Eggensperger, S. Falkner, M. Lindauer, and F. Hutter, “Auto-sklearn 2.0: Hands-free automl via meta-learning,” Journal of Machine Learning Research , vol. 23, no. 261, pp. 1–61, 2022

    2022

  24. [32]

    GAMA: Genetic automated machine learning assistant,

    P. Gijsbers and J. Vanschoren, “GAMA: Genetic automated machine learning assistant,” Journal of Open Source Software , vol. 4, no. 33, 2019

    2019

  25. [33]

    Naive automated machine learning,

    F. Mohr and M. Wever, “Naive automated machine learning,” Machine Learning, vol. 112, pp. 1131–1170, 2022

    2022

  26. [34]

    Time series feature extraction on basis of scalable hypothesis tests (tsfresh – a python package),

    M. Christ, N. Braun, J. Neuffer, and A. W. Kempa-Liehr, “Time series feature extraction on basis of scalable hypothesis tests (tsfresh – a python package),” Neurocomputing, vol. 307, pp. 72–77, 2018

    2018

  27. [35]

    Scikit-learn: Machine learning in Python,

    F. Pedregosa, G. Varoquaux, A. Gramfort, V . Michel, B. Thirion, O. Grisel, M. Blondel, P. Prettenhofer, R. Weiss, V . Dubourg, J. Vander- plas, A. Passos, D. Cournapeau, M. Brucher, M. Perrot, and E. Duch- esnay, “Scikit-learn: Machine learning in Python,” Journal of Machine ...

    2011

  28. [36]

    Plant electrical signals: A multidisciplinary challenge,

    J.-H. Li, L.-F. Fan, D.-J. Zhao, Q. Zhou, J.-P. Yao, Z.-Y . Wang, and L. Huang, “Plant electrical signals: A multidisciplinary challenge,” Journal of Plant Physiology , vol. 261, p. 153418, 2021

    2021

  29. [37]

    System Potentials, a Novel Electrical Long-Distance Apoplastic Signal in Plants, Induced by Wounding,

    M. R. Zimmermann, H. Maischak, A. Mith ¨ofer, W. Boland, and H. H. Felle, “System Potentials, a Novel Electrical Long-Distance Apoplastic Signal in Plants, Induced by Wounding,” Plant Physiology , vol. 149, no. 3, pp. 1593–1600, 2009

    2009

  30. [38]

    Kernbach, Differential Impedance Spectrometer for electrochemical and electrophysiological analysis of fluids and organic tissues

    S. Kernbach, Differential Impedance Spectrometer for electrochemical and electrophysiological analysis of fluids and organic tissues. Hand- book and User Manual . CYBRES GmbH, Stuttgart, 2022

    2022

Pith tools

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