pith. sign in

arxiv: 2606.27411 · v1 · pith:YWEZHQVYnew · submitted 2026-06-25 · 🪐 quant-ph · cs.AI· eess.IV

Compression-Driven Anomaly Detection in Brain MRI Using an Interpretable Quantum Autoencoder

Santanu Ganguly , Xing Liang , Dimitrios Makris This is my paper

Pith reviewed 2026-06-29 02:08 UTC · model grok-4.3

classification 🪐 quant-ph cs.AIeess.IV
keywords quantum autoencoderanomaly detectionbrain MRIquantum machine learningcompressionmedical imaginginterpretable model
0
0 comments X

The pith

A quantum autoencoder detects brain MRI anomalies by scoring resistance to compression into a learned normal representation.

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

The paper trains a variational quantum autoencoder on normal brain MRI patches using angle encoding and trash qubits to discard information. Anomaly scores come from the residual information in the trash qubits, which rises when inputs deviate from the normal manifold. On public DICOM datasets this yields slice-level ROC-AUC near 0.95 and patch-level ROC-AUC near 0.813, exceeding classical autoencoder and PCA baselines. Parameter analysis reveals that detection arises from structured encoder compression rather than decoder size, producing a controllable trade-off and localized heatmaps that align with tumors. The work frames quantum autoencoders as an interpretable tool for anomaly detection grounded in incompressibility.

Core claim

A quantum autoencoder trained to compress normal brain MRI patches via angle encoding and auxiliary trash qubits produces anomaly scores from the uncompressible residual; on public datasets this reaches slice-level ROC-AUC of approximately 0.95 and patch-level ROC-AUC of approximately 0.813, outperforming classical baselines, while learned encoder-decoder asymmetry shows that effective detection stems from controlled compression in the encoder and yields spatially localized heatmaps aligned with tumor regions.

What carries the argument

The variational quantum autoencoder with trash qubits whose measured information quantifies incompressibility of an input state relative to the learned normal latent representation.

If this is right

  • Slice-level detection reaches ROC-AUC of approximately 0.95 on public brain MRI data.
  • Patch-level detection reaches ROC-AUC of approximately 0.813 and exceeds classical autoencoder and PCA performance.
  • Encoder-decoder asymmetry shows anomaly detection arises from structured encoder compression rather than decoder expressivity.
  • The resulting compression-reconstruction trade-off supplies a clear regime for threshold selection.
  • Qualitative heatmaps localize high scores to tumorous regions.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same trash-qubit measure could be tested on other imaging modalities to check whether incompressibility remains a reliable anomaly signal.
  • Focusing quantum resources primarily on the encoder stage may improve similar detection tasks.
  • The interpretable operating regime could be combined with classical post-processing to produce confidence intervals on anomaly scores.
  • If the asymmetry holds across datasets, it would suggest redesigning variational circuits to allocate fewer parameters to the decoder.

Load-bearing premise

Resistance to compression measured by trash-qubit information corresponds to clinically meaningful deviations rather than dataset artifacts or training instabilities.

What would settle it

On a held-out set of artifact-containing but tumor-free MRI slices, the quantum autoencoder would assign systematically high anomaly scores.

read the original abstract

We study a quantum autoencoder (QAE) for compression-driven anomaly detection in brain MRI data. The approach leverages angle encoding to map image patches into quantum states, followed by a variational encoder-decoder architecture trained to discard information via auxiliary trash qubits. Anomaly scores reflect the degree to which inputs resist compression relative to normal data, with higher scores corresponding to deviations from the learned normal manifold. Evaluated on publicly available brain MRI DICOM datasets, the method achieves a slice-level ROC-AUC of approximately 0.95 and a patch-level ROC-AUC of approximately 0.813, outperforming classical autoencoder and PCA baselines. Analysis of the learned parameters reveals a pronounced encoder-decoder asymmetry, where effective anomaly detection arises from structured information compression within the encoder rather than increased parameter magnitude or decoder expressivity. This results in a controlled compression-reconstruction trade-off with a clear operating regime that supports principled threshold selection. Qualitative evaluation further shows that the QAE produces spatially localized anomaly heatmaps aligned with tumorous regions. The results, supported by promising baseline performances, demonstrate that quantum autoencoders provide an interpretable and controllable mechanism for anomaly detection based on incompressibility with respect to a learned latent representation. This work highlights the potential of quantum autoencoders as a principled tool for studying compression dynamics in quantum machine learning, with promising implications for decision support in medical imaging workflows.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 2 minor

Summary. The paper proposes a quantum autoencoder (QAE) for anomaly detection in brain MRI slices and patches. It employs angle encoding of image patches into quantum states, a variational encoder-decoder circuit with auxiliary trash qubits to enforce compression, and anomaly scores derived from reconstruction error or trash-qubit information content. On public DICOM brain MRI datasets the method reports slice-level ROC-AUC ≈ 0.95 and patch-level ROC-AUC ≈ 0.813, outperforming classical autoencoders and PCA; it further claims an interpretable encoder-decoder asymmetry that enables controlled compression and spatially localized heatmaps aligned with tumors.

Significance. If the reported performance and interpretability claims are substantiated with reproducible experimental protocols, the work would constitute a concrete demonstration that quantum compression can serve as a controllable anomaly detector in medical imaging. The emphasis on encoder-decoder asymmetry and the explicit link between incompressibility and anomaly scoring would be a useful addition to the quantum machine-learning literature on variational autoencoders.

major comments (3)
  1. [Methods] Methods (training protocol): the abstract and any referenced experimental section supply no circuit depth, number of qubits, variational ansatz, optimizer, learning-rate schedule, batch size, or number of training epochs. These parameters are load-bearing for the central claim that the QAE outperforms classical baselines at the stated ROC-AUC values; without them the numerical results cannot be verified or reproduced.
  2. [Results] Results (anomaly-score validation): the mapping from trash-qubit information (or reconstruction error) to clinically meaningful tumor regions is asserted but not accompanied by explicit threshold-selection procedure, precision-recall curves, or statistical tests against ground-truth segmentations. This directly affects the weakest assumption that resistance to compression reliably indicates pathology rather than dataset artifacts.
  3. [Results] Results (baseline comparison): the classical autoencoder and PCA baselines are stated to be outperformed, yet no description of their architectures, hyperparameter tuning, or input preprocessing is provided. Without these details the outperformance claim cannot be assessed as fair or robust.
minor comments (2)
  1. [Abstract] Abstract: report exact rather than approximate ROC-AUC values together with standard deviations or confidence intervals obtained from multiple runs or cross-validation folds.
  2. [Methods] Notation: the precise mathematical definition of the anomaly score (e.g., whether it is the expectation value on trash qubits, the reconstruction fidelity, or a combination) should be stated explicitly with an equation number.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for their thorough review and valuable suggestions. We address each of the major comments below and have prepared revisions to the manuscript to improve the reproducibility and validation aspects of our work.

read point-by-point responses

  1. Referee: [Methods] Methods (training protocol): the abstract and any referenced experimental section supply no circuit depth, number of qubits, variational ansatz, optimizer, learning-rate schedule, batch size, or number of training epochs. These parameters are load-bearing for the central claim that the QAE outperforms classical baselines at the stated ROC-AUC values; without them the numerical results cannot be verified or reproduced.

    Authors: We agree that detailed training protocol information is necessary to allow verification and reproduction of the reported results. In the revised manuscript, we will include a new subsection in the Methods section that specifies all relevant parameters: the number of qubits, circuit depth, the variational ansatz structure, the optimizer and its learning-rate schedule, batch size, and the number of training epochs. revision: yes

  2. Referee: [Results] Results (anomaly-score validation): the mapping from trash-qubit information (or reconstruction error) to clinically meaningful tumor regions is asserted but not accompanied by explicit threshold-selection procedure, precision-recall curves, or statistical tests against ground-truth segmentations. This directly affects the weakest assumption that resistance to compression reliably indicates pathology rather than dataset artifacts.

    Authors: We acknowledge the importance of rigorous validation for the anomaly scoring. In the revised manuscript, we will add precision-recall curves, an explicit description of the threshold selection procedure (e.g., based on a validation set), and statistical tests comparing the anomaly scores to ground-truth segmentations to better substantiate that the scores align with pathological regions. revision: yes

  3. Referee: [Results] Results (baseline comparison): the classical autoencoder and PCA baselines are stated to be outperformed, yet no description of their architectures, hyperparameter tuning, or input preprocessing is provided. Without these details the outperformance claim cannot be assessed as fair or robust.

    Authors: We agree that complete details on the baseline methods are required for a fair assessment. In the revised manuscript, we will expand the Results section to describe the architectures of the classical autoencoder and PCA, including their hyperparameters, tuning procedures, and the input preprocessing steps applied to ensure comparable conditions. revision: yes

Circularity Check

0 steps flagged

No significant circularity detected

full rationale

The abstract and provided text present empirical ROC-AUC results on public brain MRI datasets, with anomaly scores defined via compression resistance in a QAE architecture. No equations, derivation steps, or self-citations appear that reduce any claimed prediction or result to fitted inputs by construction. Performance claims are framed as comparisons against external baselines (classical autoencoder, PCA), and interpretability follows from observed parameter asymmetry rather than any self-referential loop. The derivation chain is therefore self-contained against external benchmarks, with no load-bearing steps matching the enumerated circularity patterns.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The central claim rests on standard variational quantum circuit assumptions and the premise that compression resistance maps to anomaly status. Free parameters are the trained variational angles; no new physical entities are introduced.

free parameters (1)
  • variational parameters of encoder-decoder circuit
    Angles and other trainable parameters optimized on normal MRI patches to achieve compression.
axioms (1)
  • standard math Standard quantum mechanics and the variational principle for circuit training
    Angle encoding and trash-qubit discarding presuppose the usual rules of quantum state evolution and measurement.

pith-pipeline@v0.9.1-grok · 5785 in / 1394 out tokens · 34462 ms · 2026-06-29T02:08:22.369200+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

141 extracted references · 107 canonical work pages · 9 internal anchors

  1. [1]

    A., & Chuang, I

    Nielsen, M. A., & Chuang, I. L. (2010). Quantum computation and quantum information . Cambridge Cambridge University Press

    2010

  2. [2]

    Kulshrestha, A., Liu, X., Ushijima -Mwesigwa, H., & Safro, I. (2025). Neural Architecture Search Algorithms for Quantum Autoencoders. IEEE Transactions on Quantum Engineer- ing, 1–18. https://doi.org/10.1109/tqe.2025.3615886

    work page doi:10.1109/tqe.2025.3615886 2025

  3. [3]

    Shor, P. W. (1997). Polynomial-Time Algorithms for Prime Factorization and Discrete Log- arithms on a Quantum Computer. SIAM Journal on Computing , 26(5), 1484 –1509. https://doi.org/10.1137/s0097539795293172

    work page doi:10.1137/s0097539795293172 1997

  4. [4]

    G., Laflamme, R., Knill, E., Viola, L., Havel, T

    Cory, D. G., Laflamme, R., Knill, E., Viola, L., Havel, T. F., Boulant, N., Boutis, G., Fortu- nato, E., Lloyd, S., Martinez, R., Negrevergne, C., Pravia, M., Sharf, Y., Teklemariam, G., Weinstein, Y. S., & Zurek, W. H. (2000). NMR Based Quantum Information Processing: Achievements and Prospects. Fortschritte Der Physik , 48(9-11), 875 –907. https://doi.o...

    work page doi:10.1002/1521-3978(200009)48:9/11 2000

  5. [5]

    Yan, F., Iliyasu, A.M., Liu, Z., Salama, A.S., Dong, F., & Hirota, K. (2015). Bloch Sphere- Based Representation for Quantum Emotion Space. J. Adv. Comput. Intell. Intell. Informat- ics, 19, 134-142. https://doi.org/10.20965/jaciii.2015.p0134

    work page doi:10.20965/jaciii.2015.p0134 2015

  6. [6]

    Yang, Y.-G., Xia, J., Jia, X., & Zhang, H. (2013). Novel image encryption/decryption based on quantum Fourier transform and double phase encoding. Quantum Information Pro- cessing, 12(11), 3477–3493. https://doi.org/10.1007/s11128-013-0612-y

    work page doi:10.1007/s11128-013-0612-y 2013

  7. [7]

    Zhou, R.-G., Wu, Q., Zhang, M.-Q., & Shen, C.-Y. (2012). Quantum Image Encryption and Decryption Algorithms Based on Quantum Image Geometric Transformations. Internatio- nal Journal of Theoretical Physics, 52(6), 1802–1817. https://doi.org/10.1007/s10773-012- 1274-8

    work page doi:10.1007/s10773-012- 2012

  8. [8]

    Leymann, F., & Barzen, J. (2020). The bitter truth about gate -based quantum algorithms in the NISQ era. Quantum Science and Technology , 5(4), 044007. https://doi.org/10.1088/2058-9565/abae7d

    work page doi:10.1088/2058-9565/abae7d 2020

  9. [9]

    V., & Markov, I

    Shende, V. V., & Markov, I. L. (2005). Quantum circuits for incompletely specified two - qubit operators. Quantum Information and Computation , 5(1), 48 –56. https://doi.org/10.26421/qic5.1-5

    work page doi:10.26421/qic5.1-5 2005

  10. [10]

    Benedetti, M., Lloyd, E., Sack, S., & Fiorentini, M. (2019). Parameterized quantum circuits as machine learning models. Quantum Science and Technology , 4(4), 043001. https://doi.org/10.1088/2058-9565/ab4eb5

    work page doi:10.1088/2058-9565/ab4eb5 2019

  11. [11]

    Farhi, E., & Neven, H. (2018). Classification with Quantum Neural Networks on Near Term Processors. arXiv: Quantum Physics . https://doi.org/10.48550/arXiv.1802.06002, last ac- cessed 2025/11/10

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.1802.06002 2018

  12. [12]

    Younes *, A., & Miller, J. F. (2004). Representation of Boolean quantum circuits as reed – Muller expansions. International Journal of Electronics , 91(7), 431 –444. https://doi.org/10.1080/00207210412331272643

    work page doi:10.1080/00207210412331272643 2004

  13. [13]

    Huang, H., Broughton, M., Mohseni, M., Babbush, R., Boixo, S., Neven, H., & McClean, J.R. (2020). Power of data in quantum machine learning. Nature Communications, 12 . https://doi.org/10.1038/s41467-021-22539-9

    work page doi:10.1038/s41467-021-22539-9 2020

  14. [14]

    Fashion MINIST original dataset: https://github.com/zalandoresearch/fashion-mnist, last accessed 2025/11/10 29

    2025

  15. [15]

    Tensorflow quantum: A software framework for quantum machine learning.arXiv preprint arXiv:2003.02989, 2020

    Broughton, M., Verdon, G., McCourt, T., Martinez, A. J., Yoo, J. H., Isakov, S. V., Massey, P., Niu, M. Y., Ramin Halavati, Peters, E., Leib, M., Skolik, A., Streif, M., Dollen, D. V., McClean, J. R., Boixo, S., Bacon, D., Ho, A. K., Neven, H., & Mohseni, M. (2020). Tensor- Flow Quantum: A Software Framework for Quantum Machine Learning. ArXiv (Cornell Un...

    work page doi:10.48550/arxiv.2003.02989 2020

  16. [16]

    Schuld, M., & Killoran, N. (2019). Quantum Machine Learning in Feature Hilbert Spaces. Physical Review Letters, 122(4). https://doi.org/10.1103/physrevlett.122.040504

    work page doi:10.1103/physrevlett.122.040504 2019

  17. [17]

    Ganguly, S. (2021). Quantum Machine Learning: An Applied Approach. Apress

    2021

  18. [18]

    Schuld, M., & Petruccione, F. (2018). Supervised Learning with Quantum Computers. In Quantum Science and Technology . Springer International Publishing. https://doi.org/10.1007/978-3-319-96424-9

    work page doi:10.1007/978-3-319-96424-9 2018

  19. [19]

    D., Temme, K., Harrow, A

    Havlícek, V., Córcoles, A. D., Temme, K., Harrow, A. W., Kandala, A., Chow, J. M., & Gambetta, J. M. (2019). Supervised learning with quantum -enhanced feature spaces. Na- ture, 567, 209–212. https://doi.org/10.1038/s41586-019-0980-2

    work page doi:10.1038/s41586-019-0980-2 2019

  20. [20]

    Lloyd, S., Mohseni, M., & Rebentrost, P. (2014). Quantum principal component analy- sis. Nature Physics, 10(9), 631–633. https://doi.org/10.1038/nphys3029

    work page doi:10.1038/nphys3029 2014

  21. [21]

    Williams, C. K. I. (2003). Learning With Kernels: Support Vector Machines, Regularization, Optimization, and Beyond. Journal of the American Statistical Association , 98(462), 489–

    2003

  22. [22]

    https://doi.org/10.1198/jasa.2003.s269

    work page doi:10.1198/jasa.2003.s269 2003

  23. [23]

    Bernhard Schölkopf, Ralf Herbrich, & Smola, A. J. (2001). A Generalized Representer The- orem. 416–426. https://doi.org/10.1007/3-540-44581-1_27

    work page doi:10.1007/3-540-44581-1_27 2001

  24. [24]

    Schuld, M., Fingerhuth, M., & Petruccione, F. (2017). Quantum machine learning with small-scale devices: Implementing a distance -based classifier with a quantum interference circuit. ArXiv (Cornell University), , last accessed 2025/12/17

    2017

  25. [25]

    Arjovsky, M., Shah, A., & Yoshua Bengio. (2016). Unitary evolution recurrent neural net- works. 1120–1128

    2016

  26. [26]

    McWeeny, R. (1960). Some Recent Advances in Density Matrix Theory. Reviews of Mod- ern Physics, 32(2), 335–369. https://doi.org/10.1103/revmodphys.32.335

    work page doi:10.1103/revmodphys.32.335 1960

  27. [27]

    Zhang, Y., Lu, K., Gao, Y., & Wang, M. (2013). NEQR: a novel enhanced quantum repre- sentation of digital images. Quantum Information Processing , 12(8), 2833 –2860. https://doi.org/10.1007/s11128-013-0567-z

    work page doi:10.1007/s11128-013-0567-z 2013

  28. [28]

    Le Quang Phuc, Fangyang, D., Arai Yoshinori, & Hirota Kaoru. (2009). Flexible Represen- tation of Quantum Images and Its Computational Complexity Analysis . 25, 185 –185. https://doi.org/10.14864/fss.25.0.185.0

    work page doi:10.14864/fss.25.0.185.0 2009

  29. [29]

    M., Yan, F., Dong, F., & Hirota, K

    Sun, B., Iliyasu, A. M., Yan, F., Dong, F., & Hirota, K. (2013). An RGB Multi -Channel Representation for Images on Quantum Computers. Journal of Advanced Computational In- telligence and Intelligent Informatics , 17(3), 404 –417. https://doi.org/10.20965/jaciii.2013.p0404

    work page doi:10.20965/jaciii.2013.p0404 2013

  30. [30]

    E., & Ball, J

    Venegas-Andraca, S. E., & Ball, J. L. (2009). Processing images in entangled quantum sys- tems. Quantum Information Processing , 9(1), 1 –11. https://doi.org/10.1007/s11128 -009- 0123-z

    work page doi:10.1007/s11128 2009

  31. [31]

    Latorre, J. (2005). Image compression and entanglement . https://doi.org/10.48550/ar- Xiv.quant-ph/0510031, last accessed 2025/12/17

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/ar- 2005

  32. [32]

    Grover, L. K. (1997). Quantum Mechanics Helps in Searching for a Needle in a Hay- stack. Physical Review Letters, 79(2), 325–328. https://doi.org/10.1103/physrevlett.79.325. 30

    work page doi:10.1103/physrevlett.79.325 1997

  33. [33]

    Qiskit API documentation: https://qiskit.org/documentation/, last accessed 2025/12/17

    2025

  34. [34]

    Quantum computing with Qiskit

    Javadi-Abhari, A., Treinish, M., Krsulich, K., Wood, C. J., Lishman, J., Gacon, J., Martiel, S., Nation, P., Bishop, L. S., Cross, A. W., Johnson, B. R., & Gambetta, J. M. (2024, May 15). Quantum computing with Qiskit. https://doi.org/10.48550/arXiv.2405.08810, last ac- cessed 2025/12/17

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.2405.08810 2024

  35. [36]

    N., Duvenaud, D., Hernández -Lobato, J

    Gómez-Bombarelli, R., Wei, J. N., Duvenaud, D., Hernández -Lobato, J. M., Sánchez - Lengeling, B., Sheberla, D., Aguilera -Iparraguirre, J., Hirzel, T. D., Adams, R. P., & As- puru-Guzik, A. (2018). Automatic Chemical Design Using a Data-Driven Continuous Rep- resentation of Molecules. ACS Central Science , 4(2), 268 –276. https://doi.org/10.1021/acscents...

    work page doi:10.1021/acscentsci.7b00572 2018

  36. [37]

    R., Venkatesh, S., & Anton

    Gong, D., Liu, L., Le, V., Saha, B., Mansour, M. R., Venkatesh, S., & Anton. (2019). Mem- orizing Normality to Detect Anomaly: Memory-augmented Deep Autoencoder for Unsuper- vised Anomaly Detection. ArXiv (Cornell University) . https://doi.org/10.48550/arxiv.1904.02639

    work page doi:10.48550/arxiv.1904.02639 2019

  37. [38]

    Chen, Z., Yeo, C., Lee, B., & Lau, C.T. (2018). Autoencoder -based network anomaly de- tection. 2018 Wireless Telecommunications Symposium (WTS) , 1 -5. https://doi.org/10.1109/WTS.2018.8363930

    work page doi:10.1109/wts.2018.8363930 2018

  38. [39]

    Yu, J., Li, S., Liu, X., Li, H., Ma, M., Liu, P., & You, L. (2024). Residual squeeze-and- excitation convolutional auto-encoder for fault detection and diagnosis in complex industrial processes. Engineering Applications of Artificial Intelligence , 136, 108872 –108872. https://doi.org/10.1016/j.engappai.2024.108872

    work page doi:10.1016/j.engappai.2024.108872 2024

  39. [40]

    Harrou, F., Dairi, A., Taghezouit, B., Khaldi, B., & Sun, Y. (2024). Automatic fault detec- tion in grid-connected photovoltaic systems via variational autoencoder -based monitoring. Energy Conversion and Management , 314, 118665. https://doi.org/10.1016/j.encon- man.2024.118665

    work page doi:10.1016/j.encon- 2024

  40. [41]

    (2017, November 1)

    Wan, Z., Zhang, Y., & He, H. (2017, November 1). Variational autoencoder based synthetic data generation for imbalanced learning . IEEE Xplore. https://doi.org/10.1109/SSCI.2017.8285168

    work page doi:10.1109/ssci.2017.8285168 2017

  41. [42]

    Majumdar, A. (2019). Blind Denoising Autoencoder. IEEE Transactions on Neural Net- works and Learning Systems, 30(1), 312–317. https://doi.org/10.1109/tnnls.2018.2838679

    work page doi:10.1109/tnnls.2018.2838679 2019

  42. [43]

    Jiang, J., Ren, H., & Zhang, M. (2021). A Convolutional Autoencoder Method for Simulta- neous Seismic Data Reconstruction and Denoising. IEEE Geoscience and Remote Sensing Letters, 19, 1–5. https://doi.org/10.1109/lgrs.2021.3073560

    work page doi:10.1109/lgrs.2021.3073560 2021

  43. [44]

    L., & Greco, S

    Torabi, H., Mirtaheri, S. L., & Greco, S. (2023). Practical autoencoder based anomaly de- tection by using vector reconstruction error. Cybersecurity, 6(1). https://doi.org/10.1186/s42400-022-00134-9

    work page doi:10.1186/s42400-022-00134-9 2023

  44. [45]

    Gogoi, M., & Begum, S.A. (2017). Image Classification Using Deep Autoencoders. 2017 IEEE International Conference on Computational Intelligence and Computing Research (ICCIC), 1-5. https://doi.org/10.1109/iccic.2017.8524276

    work page doi:10.1109/iccic.2017.8524276 2017

  45. [46]

    Luo, W., Li, J., Yang, J., Xu, W., & Zhang, J. (2017). Convolutional Sparse Autoencoders for Image Classification. IEEE Transactions on Neural Networks and Learning Systems, 1–

    2017

  46. [47]

    https://doi.org/10.1109/tnnls.2017.2712793 31

    work page doi:10.1109/tnnls.2017.2712793 2017

  47. [48]

    Zhou, P., Han, J., Cheng, G., & Zhang, B. (2019). Learning Compact and Discriminative Stacked Autoencoder for Hyperspectral Image Classification . 57(7), 4823 –4833. https://doi.org/10.1109/tgrs.2019.2893180

    work page doi:10.1109/tgrs.2019.2893180 2019

  48. [49]

    Sun, Y., Xue, B., Zhang, M., & Yen, G. G. (2019). A Particle Swarm Optimization -Based Flexible Convolutional Autoencoder for Image Classification. IEEE Transactions on Neural Networks and Learning Systems , 30(8), 2295 –2309. https://doi.org/10.1109/tnnls.2018.2881143

    work page doi:10.1109/tnnls.2018.2881143 2019

  49. [50]

    P., & Aspuru -Guzik, A

    Romero, J., Olson, J. P., & Aspuru -Guzik, A. (2017). Quantum autoencoders for efficient compression of quantum data. Quantum Science and Technology , 2(4), 045001. https://doi.org/10.1088/2058-9565/aa8072

    work page doi:10.1088/2058-9565/aa8072 2017

  50. [51]

    Liu, H., Gao, Y., Shi, L., Wei, L., Shan, Z., & Zhao, B. (2023). HM-QCNN: Hybrid Multi- branches Quantum-Classical Neural Network for Image Classification. International Con- ference on Advanced Data Mining and Applications

    2023

  51. [52]

    Cong, I., Choi, S., & Lukin, M. D. (2019). Quantum convolutional neural networks. Nature Physics, 15(12), 1273–1278. https://doi.org/10.1038/s41567-019-0648-8

    work page doi:10.1038/s41567-019-0648-8 2019

  52. [53]

    Wijanarko, H., Calista, E., Chen, L., & Chen, Y. (2024). Tri-VAE: Triplet Variational Au- toencoder for Unsupervised Anomaly Detection in Brain Tumor MRI. 2024 IEEE/CVF Con- ference on Computer Vision and Pattern Recognition Workshops (CVPRW), 3930-3939

    2024

  53. [54]

    Bayraktar, H.H., Charara, A., Clark, D., Cohen, S., Costa, T.B., Fang, Y.L., Gao, Y., Guan, J., Gunnels, J.A., Haidar, A., Hehn, A., Hohnerbach, M., Jones, M.T., Lubowe, T., Lyakh, D.I., Morino, S., Springer, P.L., Stanwyck, S.W., Terentyev, I.S., Varadhan , S., Wong, J., & Yamaguchi, T. (2023). cuQuantum SDK: A High -Performance Library for Accelerating ...

    2023

  54. [55]

    Vidal, G. (2008). Class of Quantum Many-Body States That Can Be Efficiently Simulated. Physical Review Letters, 101(11). https://doi.org/10.1103/physrevlett.101.110501

    work page doi:10.1103/physrevlett.101.110501 2008

  55. [56]

    Q., Dong, F., & Hirota, K

    Le, P. Q., Dong, F., & Hirota, K. (2010). A flexible representation of quantum images for polynomial preparation, image compression, and processing operations. Quantum Infor- mation Processing, 10(1), 63–84. https://doi.org/10.1007/s11128-010-0177-y

    work page doi:10.1007/s11128-010-0177-y 2010

  56. [57]

    Yang, J., Shi, R., & Ni, B. (2020). MedMNIST Classification Decathlon: A Lightweight AutoML Benchmark for Medical Image Analysis. 2021 IEEE 18th International Symposium on Biomedical Imaging (ISBI), 191-195

    2020

  57. [58]

    Yang, J., Shi, R., Wei, D., Liu, Z., Zhao, L., Ke, B., Pfister, H., & Ni, B. (2023). MedMNIST v2 - A large-scale lightweight benchmark for 2D and 3D biomedical image classification. Scientific Data, 10(1), 41. https://doi.org/10.1038/s41597-022-01721-8

    work page doi:10.1038/s41597-022-01721-8 2023

  58. [59]

    Bergholm, V., Izaac, J., Schuld, M., Gogolin, C., Ahmed, S., Pankajam, A. V., M. Sohaib Alam, Alonso-Linaje, G., AkashNarayanan, B., Asadi, A., Juan Miguel Arrazola, Azad, U., Banning, S., Blank, C., Bromley, T. R., Cordier, B. A., Ceroni, J., Delgado, A., Olivia Di Matteo, & Amintor Dusko. (2018). PennyLane: Automatic differentiation of hybrid quan- tum-...

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.1811.04968 2018

  59. [60]

    (2018, March 14)

    File:X-ray of the hand of a 4 year old male - lateral.jpg - Wikimedia Commons . (2018, March 14). Wikimedia.org. https://commons.wikimedia.org/wiki/File:X- ray_of_the_hand_of_a_4_year_old_male_-_lateral.jpg 32

    2018

  60. [61]

    Azevedo, V., Silva, C., & Dutra, I. (2022). Quantum transfer learning for breast cancer de- tection. Quantum Machine Intelligence, 4(1). https://doi.org/10.1007/s42484-022-00062-4

    work page doi:10.1007/s42484-022-00062-4 2022

  61. [62]

    Y., Strahm, M., Kazdaghli, S., Prakash, A., & Kerenidis, I

    Landman, J., Mathur, N., Li, Y. Y., Strahm, M., Kazdaghli, S., Prakash, A., & Kerenidis, I. (2022). Quantum Methods for Neural Networks and Application to Medical Image Classi- fication. Quantum, 6, 881. https://doi.org/10.22331/q-2022-12-22-881

    work page doi:10.22331/q-2022-12-22-881 2022

  62. [63]

    Matondo-Mvula, N., & Khaled Elleithy. (2024). Breast Cancer Detection with Quanvolu- tional Neural Networks. Entropy, 26(8), 630–630. https://doi.org/10.3390/e26080630

    work page doi:10.3390/e26080630 2024

  63. [64]

    Chalumuri, A., Kune, R., & Manoj, B. S. (2021). A hybrid classical-quantum approach for multi-class classification. Quantum Information Processing, 20(3). https://doi.org/10.1007/s11128-021-03029-9

    work page doi:10.1007/s11128-021-03029-9 2021

  64. [65]

    M., Ortega, G., Orts, F., Garzón, E

    Donaire, L. M., Ortega, G., Orts, F., Garzón, E. M., & Filatovas, E. (2026). A hybrid quan- tum-classical approach for liver disease detection using quantum machine learning. Engi- neering Applications of Artificial Intelligence, 164, 113240. https://doi.org/10.1016/j.en- gappai.2025.113240

    work page doi:10.1016/j.en- 2026

  65. [66]

    Henderson, M., Shakya, S., Pradhan, S., & Cook, T. (2020). Quanvolutional neural net- works: powering image recognition with quantum circuits. Quantum Machine Intelligence, 2(1), 1–9. https://doi.org/10.1007/s42484-020-00012-y

    work page doi:10.1007/s42484-020-00012-y 2020

  66. [67]

    Senokosov, A., Sedykh, A., Sagingalieva, A., Kyriacou, B., & Melnikov, A. (2024). Quan- tum machine learning for image classification. Machine Learning: Science and Technology, 5(1), 015040. https://doi.org/10.1088/2632-2153/ad2aef

    work page doi:10.1088/2632-2153/ad2aef 2024

  67. [68]

    Alvarez-Estevez, D. (2025). Benchmarking quantum machine learning kernel training for classification tasks. IEEE Transactions on Quantum Engineering, 1 –15. https://doi.org/10.1109/tqe.2025.3541882

    work page doi:10.1109/tqe.2025.3541882 2025

  68. [69]

    & Schuch, N

    Mari, A., Bromley, T. R., Izaac, J., Schuld, M., & Killoran, N. (2020). Transfer learning in hybrid classical -quantum neural networks. Quantum, 4, 340. https://doi.org/10.22331/q- 2020-10-09-340

    work page doi:10.22331/q- 2020

  69. [70]

    Li, Y., Hao, X., Liu, G., Shang, R., & Jiao, L. (2024). QEA-QCNN: optimization of quantum convolutional neural network architecture based on quantum evolution. Memetic Compu- ting, 16(3), 233–254. https://doi.org/10.1007/s12293-024-00417-3

    work page doi:10.1007/s12293-024-00417-3 2024

  70. [71]

    Shi, S., Wang, Z., Li, J., Li, Y., Shang, R., Zhong, G., & Gu, Y. (2024). Quantum convolu- tional neural networks for multiclass image classification. Quantum Information Processing, 23(5). https://doi.org/10.1007/s11128-024-04360-7

    work page doi:10.1007/s11128-024-04360-7 2024

  71. [72]

    Slabbert, D., & Petruccione, F. (2025). Classical-quantum approach to image classification: Autoencoders and quantum SVMs. AVS Quantum Science, 7(2). https://doi.org/10.1116/5.0261885

    work page doi:10.1116/5.0261885 2025

  72. [73]

    Singh, G., Jin, H., & Merz, K. (2025). Benchmarking MedMNIST dataset on real quantum hardware. ArXiv, abs/2502.13056. Last accessed 2025/12/18

    arXiv 2025

  73. [74]

    Ganguly, S. (2022). Classification of NEQR Processed Classical Images using Quantum Neural Networks (QNN). ArXiv, abs/2204.02797. Last accessed 2025/12/15

    arXiv 2022

  74. [75]

    Mathur, N., Landman, J., Li, Y., Strahm, M., Kazdaghli, S., Prakash, A., & Kerenidis, I. (2021). Medical image classification via quantum neural networks. Last accessed 2025/12/15

    2021

  75. [76]

    Chen, G., Chen, Q., Long, S., Zhu, W., Yuan, Z., & Wu, Y. (2022). Quantum convolutional neural network for image classification. Pattern Analysis and Applications. https://doi.org/10.1007/s10044-022-01113-z 33

    work page doi:10.1007/s10044-022-01113-z 2022

  76. [77]

    Zhuang, S., Wu, Y., Cadet, X.F., Huynh, D., Liu, W., Charton, P., Damour, C., Cadet, F., & Wang, J. (2025). Quantum Autoencoder: An efficient approach to quantum feature map gen- eration. Last accessed 2025/12/18

    2025

  77. [78]

    Basit, J., Hanif, D., & Arshad, M. (2024). Quantum Variational Autoencoders for Predictive Analytics in High Frequency Trading Enhancing Market Anomaly Detection. International Journal of Emerging Multidisciplinaries: Computer Science & Artificial Intellig ence, 3(1),

    2024

  78. [79]

    https://doi.org/10.54938/ijemdcsai.2024.03.1.319

    work page doi:10.54938/ijemdcsai.2024.03.1.319 2024

  79. [80]

    Khoshaman, A., Vinci, W., Denis, B., Andriyash , E., Sadeghi, H., & Amin, M. H. (2018). Quantum variational autoencoder. Quantum Science and Technology, 4(1), 014001. https://doi.org/10.1088/2058-9565/aada1f

    work page doi:10.1088/2058-9565/aada1f 2018

  80. [81]

    Bravo-Prieto, C. (2021). Quantum autoencoders with enhanced data encoding. Machine Learning: Science and Technology, 2(3), 035028. https://doi.org/10.1088/2632- 2153/ac0616

    work page doi:10.1088/2632- 2021

Showing first 80 references.