OSCR

Human-in-the-Loop Enhances Machine Learning Inference in Intraoperative Optical Coherence Tomography Glioma Imaging.

Code ↔ Paper

2 matches between paragraphs of the paper and lines of its authors' code, computed by the harvester (lexical-v1). Click a colored paragraph or line to see its counterpart.

The 2 matches
  1. [1] § 2. Materials and Methods › 2.3. OCT Scans Preprocessing and Parametric Maps ↔ Part3_Processor.py, lines 65 to 99 · score 0.59 · generate OAC, speckle contrast, intensity, window, RSC, OCT · Report a mismatch
  2. [2] § 2. Materials and Methods › 2.1. OCT Setup ↔ Part1_Generator.py, lines 5 to 45 · score 0.55 · central wavelength, laterally, beam, pixel, depth, scans · Report a mismatch

Paper

Loaded from Europe PMC by your browser; OSCR does not store it: doi.org · Europe PMC

The paper is loaded when this pane is shown.

The authors' code

Python · 106 lines · 3.6 KB · no license · 1 match

  1. import numpy as np
  2. import matplotlib.pyplot as plt
  3. from scipy.ndimage import uniform_filter
  4. import os
  6. # ==========================================
  7. # CONFIGURATION (Dataset Specific)
  8. # ==========================================
  9. HPX = 6.0 # Pixel size (microns) - Updated to Zenodo spec
  10. WINDOW_SIZE = 20 # Window for Speckle Contrast
  12. def load_and_linearize_image(filename):
  13. """
  14. Loads PNG and restores Linear Intensity.
  15. Assumes PNG is 0..1 mapping to 0..40dB.
  16. """
  17. if not os.path.exists(filename):
  18. raise FileNotFoundError(f"File {filename} not found.")
  20. img = plt.imread(filename)
  21. if img.ndim == 3:
  22. if img.shape[2] == 4: img = img[:, :, :3]
  23. img = np.mean(img, axis=2)
  25. if img.max() > 1.0:
  26. img = img / 255.0
  28. # Linearization: I = 10^(4 * P) -> 4 comes from 40dB / 10
  29. intensity = 10.0 ** (img * 4.0)
  30. return intensity
  32. def calculate_oac(intensity, h_px):
  33. epsilon = 1e-10
  34. I_rev = intensity[::-1, :]
  35. cumsum_rev = np.cumsum(I_rev, axis=0)
  36. cumsum_from_bottom = cumsum_rev[::-1, :]
  37. denom_term = cumsum_from_bottom - 0.5 * intensity
  38. mu = intensity / (2 * h_px * (denom_term + epsilon))
  39. return mu
  41. def calculate_speckle_contrast_map(data, window_size=20):
  42. mean_val = uniform_filter(data, size=window_size, mode='reflect')
  43. mean_sq_val = uniform_filter(data**2, size=window_size, mode='reflect')
  44. var_val = mean_sq_val - mean_val**2
  45. var_val[var_val < 0] = 0
  46. std_val = np.sqrt(var_val)
  47. sc_map = std_val / (mean_val + 1e-10)
  49. # Crop borders to avoid artifacts
  50. border = window_size // 2
  51. sc_map_cropped = sc_map[border:-border, border:-border]
  53. # Pad back to original size to keep image dimensions consistent for SSIM
  54. sc_map_padded = np.pad(sc_map_cropped, border, mode='edge')
  55. return sc_map_padded
  57. def save_map(data, filename, vmin=None, vmax=None):
  58. """Saves a normalized grayscale map for metric comparison."""
  59. if vmin is None: vmin = np.min(data)
  60. if vmax is None: vmax = np.max(data)
  62. norm_data = np.clip((data - vmin) / (vmax - vmin + 1e-10), 0, 1)
  63. plt.imsave(filename, norm_data, cmap='gray')
  65. def generate_maps(input_image_path):
  66. """
  67. Generates OAC, SC, and RSC maps from a structural OCT image.
  68. Saves them with suffixes _OAC.png, _SC.png, _RSC.png.
  69. Returns a dictionary of generated file paths.
  70. """
  71. base_name = os.path.splitext(input_image_path)[0]
  73. # 1. Load
  74. intensity = load_and_linearize_image(input_image_path)
  76. # 2. OAC
  77. mu_map = calculate_oac(intensity, HPX)
  78. oac_path = base_name + "_OAC.png"
  79. # OAC limits: usually 0 to 99th percentile
  80. save_map(mu_map, oac_path, vmin=np.min(mu_map), vmax=np.percentile(mu_map, 99))
  82. # 3. SC (on Intensity)
  83. sc_map = calculate_speckle_contrast_map(intensity, WINDOW_SIZE)
  84. sc_path = base_name + "_SC.png"
  85. save_map(sc_map, sc_path, vmin=0.5, vmax=5.0) # Standard limits
  87. # 4. RSC (on OAC)
  88. rsc_map = calculate_speckle_contrast_map(mu_map, WINDOW_SIZE)
  89. rsc_path = base_name + "_RSC.png"
  90. save_map(rsc_map, rsc_path, vmin=0.5, vmax=5.0)
  92. print(f"[Processor] Generated maps for {input_image_path}")
  94. return {
  95. 'Struct': input_image_path,
  96. 'OAC': oac_path,
  97. 'SC': sc_path,
  98. 'RSC': rsc_path
  99. }
  101. if __name__ == "__main__":
  102. # Test run (requires Test_Scan.png from Part 2)
  103. if os.path.exists("Test_Scan.png"):
  104. generate_maps("Test_Scan.png")
  105. else:
  106. print("Run Part 2 first to generate Test_Scan.png")

Part3_Processor.py at commit 9508e27, no license · at the source

Overview

Authors: Radik Zinatullin1,2, Alexander Sovetsky1, Artem Grishin2, Elena Kiseleva2, Liudmila Kukhnina2, Svetlana Korikova2, Alexander Matveyev1, Vladimir Zaitsev1, Konstantin Yashin1,2, Lev Matveev1
ORCID iDs: Radik Zinatullin, Alexander Sovetsky, Artem Grishin, Elena Kiseleva, Liudmila Kukhnina, Svetlana Korikova, Alexander Matveyev, Vladimir Zaitsev, Konstantin Yashin, Lev Matveev
  1. Russian Academy of Sciences, Institute of Applied Physics, 46 Ulyanov Str., 603951 Nizhny Novgorod, Russia; (R.Z.); (A.S.); (A.M.); (V.Z.); (K.Y.)
  2. Department of Neurosurgery, Privolzhsky Research Medical University, 10/1, Minin and Pozharsky Sq., 603950 Nizhny Novgorod, Russia; (A.G.); (E.K.); (L.K.); (S.K.)
Institutions: Privolzhsky Research Medical University (Russia); Russian Academy of Sciences (Russia); Institute of Applied Physics (Russia)
Journal: Medical sciences (Basel, Switzerland), volume 14, issue 2, article 263
Dates: received 16 March 2026; accepted 15 May 2026; published online 20 May 2026
Type: Research article · Language: English
License: CC BY
Identifiers: DOI 10.3390/medsci14020263 · PMID 42201055 · PMCID PMC13214701 · OpenAlex W7161819289
Open access: gold, a free copy (OpenAlex)
Status: code verified (how the code is found and verified)
Categories: other (modality), human (organism), other condition (population)
Methods: Connectivity, Machine learning, Statistics, fMRI & imaging
Keywords: human-in-the-loop, optical coherence tomography, glioma, machine learning, intraoperative imaging, optical biopsy
MeSH: Brain Neoplasms*, Glioma*, Machine Learning*, Tomography, Optical Coherence*, Female, Humans, Male, Reproducibility of Results, Retrospective Studies (* major topic)
Topic: Retinal Imaging and Analysis (Radiology, Nuclear Medicine and Imaging, Medicine), according to OpenAlex
Funding: Government of the Nizhny Novgorod Region and the Russian Science Foundation (RSF) (25-12-20032)
Citations: not cited yet (Europe PMC); 20 references in the paper

Abstract

Background/Objectives: The integration of Artificial Intelligence (AI) into clinical workflows raises critical questions regarding decision-making responsibility, as fully autonomous systems inevitably carry a margin of error that can be fatal in high-stakes fields like surgery. This study addresses this challenge by evaluating a “Human-in-the-Loop” (HITL) workflow, using intraoperative Optical Coherence Tomography (OCT) for glioma detection. We aimed to determine if integrating Machine Learning (ML)-generated segmentation maps with human contextual analysis resolves the tension between automation and clinical responsibility, yielding superior diagnostic reliability compared to structural or quantitative imaging alone. Methods: We retrospectively analyzed 86 intraoperative OCT scans from 27 patients. Five neurosurgeons blindly assessed the data across three progressive levels of processing: (1) structural scans, (2) physics-based parametric maps, and (3) SVM-based generated segmentation maps. Crucially, the HITL inference performance on segmentation maps was benchmarked against “models-only” inference pipeline: a SVM and a state-of-the-art multimodal reasoning model, Gemini 3.1 Pro. To evaluate interpretability and the operator’s ability to confidently exercise their authority, we measured inter-rater consistency alongside diagnostic performance. Results: The results demonstrate that, while quantitative parametric maps improved Global Accuracy (87% [95% CI: 82–92%]) compared to structural scans (80% [95% CI: 73–86%]), they suffered from an “interpretability gap,” resulting in a moderate inter-rater consistency of 0.68 [95% CI: 0.59–0.78]. In contrast, the HITL approach using segmentation maps maximized consensus to 0.98 [95% CI: 0.95–1.00] and achieved the highest performance (Accuracy 94% [95% CI: 88–98%] and Sensitivity 98% [95% CI: 92–100%]). Compared to the standalone models, the HITL approach significantly outperformed the SVM baseline (Accuracy 84% [95% CI: 81–87%]; Sensitivity 83% [95% CI: 78–88%]). Furthermore, it surpassed the SOTA Gemini 3.1 Pro model (Accuracy 90% [95% CI: 83–95%]; Sensitivity 86% [95% CI: 74–95%]). While the HITL sensitivity demonstrated a definitive and statistically significant edge over the Gemini model, the accuracy improvement fell just slightly short of undisputed statistical significance due to overlapping confidence intervals. Conclusions: By utilizing their clinical domain knowledge of tumor invasion patterns and topological priors, surgeons effectively filtered algorithmic noise—overriding ML errors in 69% (9 out of 13) false positive cases that models alone could not resolve. This demonstrates exactly how and where HITL optimally utilizes human contextual intelligence to outperform autonomous “models-only” pipelines, confirming a human-ML synergy that augments the objectivity of machine learning with human domain knowledge. This paradigm ensures that the ultimate responsibility for diagnostic inference remains safely and practically in human hands. Open Data Initiative: To ensure essential reproducibility, enable independent multi-center validation and support open science, all examples of intraoperative in vivo OCT brain scans used in this study are made publicly available. To the best of our knowledge, this represents the first open-access data of its kind globally.

Reproduced under the paper's license (CC BY), from the paper cited above.

Repositories

Its files are read in the Code ↔ Paper reader above, with 2 matches between paragraphs and lines of code.

SynthOCTChallenge/SynthOCT_Baseline

License: none: the authors keep all their rights
State: the link answers, verified on 28 September 2026
Evidence: files inventoried
Commit: 9508e27c4b0bd54e0d76172a1c85af329472f2fb, 15 June 2026
Languages: Python (6)
Size: 38 files, 6 scripts
Software Heritage: not archived
Found in: “Data Availability Statement”
Holds: README
Not found: license file, CITATION.cff, environment file, tests, continuous integration, documentation
Tools: NumPy (6 files), Matplotlib (4 files), pandas (3 files), PyTorch (2 files), scikit-image (2 files), SciPy (2 files), seaborn (1 file)
Availability: 1 check, the latest on 28 September 2026: the link answers
  • 28 September 2026: the link answers
7 files
At the source: github.com/SynthOCTChallenge/SynthOCT_Baseline

accounts.opticelastograph.com

License: none: the authors keep all their rights
State: the link answers, verified on 28 September 2026
Evidence: the link answers
Software Heritage: not checked
Found in: “Data Availability Statement”
Not found: README, license file, CITATION.cff, environment file, tests, continuous integration, documentation
Availability: 1 check, the latest on 28 September 2026: the link answers (HTTP 200)
  • 28 September 2026: the link answers (HTTP 200)
At the source: accounts.opticelastograph.com

The paper's code and data availability statement is in the Data section.

Tracing map

Proposed by the machine: these links were found in the paper and verified at the source, without human review. The map will receive a Zenodo DOI once one of the paper's authors has validated it with their ORCID.

What the map holds:

  • 2 repositories of the authors' code, each at its verified commit, with its license and how the link was found in the paper;
  • 6 scripts, each with its path and the digest of its content;
  • 2 matches between paragraphs of the paper and lines of the code (method lexical-v1);
  • neither the text of the paper nor the code itself.

Its JSON (tracing-map.json) is deposited on Zenodo with its DOI once the map is validated.

Data

No dataset and no data link were found in the paper.

Data Availability Statement

The data presented in this study are available in this article or Supplementary Material. All scan examples used in this study are available in the Supplementary Material. To support reviewer-requested reproducibility and open science, and to facilitate independent multi-center validation, the intraoperative in vivo OCT scans used in this study are attached to this paper under a Conditional Non-Exclusive License. To the best of our knowledge, this dataset represents the first publicly available intraoperative in vivo OCT scans of human brain gliomas globally. Under this Conditional Non-Exclusive License, the use of these materials is permitted subject to prior notification of the authors and on the condition that this article is appropriately cited. The data used in this retrospective study were originally acquired during RSF project No. 23-75-10068; however, that project did not provide financial support for the specific research presented in this paper. The code that converts OCT scans into physics-based maps (OAC and RSC) can be found at https://github.com/SynthOCTChallenge/SynthOCT_Baseline (accessed on 28 April 2026) (Part3_Processor.py) and on the OpticElastograph LLC cloud-based platform: https://accounts.opticelastograph.com (accessed on 28 April 2026) (requires registration). These solutions were developed under RSF project No. 25-12-20032, which solely funded this research.

Reproduced under the paper's license (CC BY), from the paper cited above.

Versions

The history of this record: each version stored by the harvester or made by a correction of its authors or of the maintainers of its code, and what changed in its facts. The texts of the paper (its abstract, its availability statements) are not part of it; versions that changed only those are not listed.

Version 1, 28 September 2026: the first record

Recorded: type, language, journal, volume, issue, pages, dates, 10 authors, 6 keywords, 9 MeSH terms, 1 funder, 18 references.

Cite

This paper

Zinatullin, R., Sovetsky, A., Grishin, A., Kiseleva, E., Kukhnina, L., Korikova, S., Matveyev, A., Zaitsev, V., Yashin, K., & Matveev, L. (2026). Human-in-the-Loop Enhances Machine Learning Inference in Intraoperative Optical Coherence Tomography Glioma Imaging. Medical sciences (Basel, Switzerland), 14(2), 263. https://doi.org/10.3390/medsci14020263

BibTeX

@article{zinatullin2026human,
author = {Zinatullin, Radik and Sovetsky, Alexander and Grishin, Artem and Kiseleva, Elena and Kukhnina, Liudmila and Korikova, Svetlana and Matveyev, Alexander and Zaitsev, Vladimir and Yashin, Konstantin and Matveev, Lev},
title = {{Human-in-the-Loop Enhances Machine Learning Inference in Intraoperative Optical Coherence Tomography Glioma Imaging}},
journal = {Medical sciences (Basel, Switzerland)},
year = {2026},
month = may,
volume = {14},
number = {2},
pages = {263},
publisher = {Multidisciplinary Digital Publishing Institute (MDPI)},
issn = {2076-3271},
doi = {10.3390/medsci14020263},
url = {https://doi.org/10.3390/medsci14020263},
pmid = {42201055},
pmcid = {PMC13214701}
}

RIS

TY - JOUR
AU - Zinatullin, Radik
AU - Sovetsky, Alexander
AU - Grishin, Artem
AU - Kiseleva, Elena
AU - Kukhnina, Liudmila
AU - Korikova, Svetlana
AU - Matveyev, Alexander
AU - Zaitsev, Vladimir
AU - Yashin, Konstantin
AU - Matveev, Lev
TI - Human-in-the-Loop Enhances Machine Learning Inference in Intraoperative Optical Coherence Tomography Glioma Imaging
T2 - Medical sciences (Basel, Switzerland)
J2 - Med Sci (Basel)
PY - 2026
DA - 2026/05/20
VL - 14
IS - 2
SP - 263
SN - 2076-3271
PB - Multidisciplinary Digital Publishing Institute (MDPI)
DO - 10.3390/medsci14020263
UR - https://doi.org/10.3390/medsci14020263
LA - en
ER -

CSL-JSON

{
"id": "10.3390/medsci14020263",
"type": "article-journal",
"title": "Human-in-the-Loop Enhances Machine Learning Inference in Intraoperative Optical Coherence Tomography Glioma Imaging",
"container-title": "Medical sciences (Basel, Switzerland)",
"author": [
{
"family": "Zinatullin",
"given": "Radik"
},
{
"family": "Sovetsky",
"given": "Alexander"
},
{
"family": "Grishin",
"given": "Artem"
},
{
"family": "Kiseleva",
"given": "Elena"
},
{
"family": "Kukhnina",
"given": "Liudmila"
},
{
"family": "Korikova",
"given": "Svetlana"
},
{
"family": "Matveyev",
"given": "Alexander"
},
{
"family": "Zaitsev",
"given": "Vladimir"
},
{
"family": "Yashin",
"given": "Konstantin"
},
{
"family": "Matveev",
"given": "Lev"
}
],
"container-title-short": "Med Sci (Basel)",
"volume": "14",
"issue": "2",
"page": "263",
"DOI": "10.3390/medsci14020263",
"PMID": "42201055",
"PMCID": "PMC13214701",
"ISSN": "2076-3271",
"publisher": "Multidisciplinary Digital Publishing Institute (MDPI)",
"URL": "https://doi.org/10.3390/medsci14020263",
"language": "en",
"issued": {
"date-parts": [
[
2026,
5,
20
]
]
}
}

The tracing map gets a citation of its own once an author has validated it and it has a DOI.

Similar papers

The papers with a page that share the most with this one: the tools found in their code, their categories, datasets, cited references and authors, the rarest counting most.

[1] doi:10.1109/jtehm.2024.3430035 [code]
XAI-Based Assessment of the AMURA Model for Detecting Amyloid-β and Tau Microstructural Signatures in Alzheimer's Disease.
Journal: IEEE journal of translational engineering in health and medicine
In common: scikit-image, PyTorch, pandas, 3 other tools, 1 reference
[2] doi:10.3389/fnins.2026.1803154 [code]
Multimodal machine learning reveals neurobiological signatures of binge-type eating disorders.
Journal: Frontiers in neuroscience
In common: seaborn, pandas, SciPy, 2 other tools, other condition, 1 reference
[3] doi:10.1038/s41586-023-06615-2 [code]
Ultra-fast deep-learned CNS tumour classification during surgery.
Journal: Nature
In common: PyTorch, pandas, Matplotlib, 1 other tool, other condition, 1 reference
[4] doi:10.1186/s12967-023-04222-3 [code]
Assessment of brain cancer atlas maps with multimodal imaging features.
Journal: Journal of translational medicine
In common: pandas, SciPy, Matplotlib, 1 other tool, other condition, 1 reference
[5] doi:10.1093/braincomms/fcaf346 [code]
Presurgical structural connectivity predicts postsurgical cognitive impairment in glioma patients.
Journal: Brain communications
In common: pandas, Matplotlib, NumPy, other condition, 1 reference
[6] doi:10.1016/j.isci.2026.117281 [code]
Spatial pattern-driven interpretable model and biological correlates in brain glioblastoma-lymphoma differentiation.
Journal: iScience
In common: pandas, NumPy, other condition, 1 reference
[7] doi:10.1016/j.dib.2026.112478
BraTioUS: A multicenter dataset of baseline intraoperative brain tumor ultrasound images.
Journal: Data in brief
In common: other, other condition, 1 reference
[8] doi:10.3171/case24139 [code]
Advanced tractography-guided laser ablation of a perirolandic long-term epilepsy-associated tumor: illustrative case.
Journal: Journal of neurosurgery. Case lessons
In common: NumPy, other condition, 1 reference
[9] doi:10.1364/boe.513877
Brain tumor grading diagnosis using transfer learning based on optical coherence tomography.
Journal: Biomedical optics express
In common: other, other condition, 1 reference
[10] doi:10.1117/1.jbo.30.1.010501
Quantitative assessment of the generalizability of a brain tumor Raman spectroscopy machine learning model to various tumor types including astrocytoma and oligodendroglioma.
Journal: Journal of biomedical optics
In common: other condition, 1 reference

Contribute

The authors of this paper can claim it, correct its record and validate its tracing map, and the maintainers of its code (its owner, or a public member of its organization) correct what it says of their repository; anyone signed in can ask for its removal. Every request goes to OSCR's own machine, which answers it; your account page follows them.

Sign in with ORCID to claim this paper as one of its authors, correct its record or validate its tracing map: when the paper's metadata lists your ORCID iD, you are recognized at once. Maintainers of its code: sign in with GitHub, then claim the repository on your account page.

Add the badge to the code's README

OSCR: the badge of this paper

If you are an author or a maintainer of this paper's code, you can add this badge. It says what OSCR knows of the paper (its code, its validated tracing map) and links the code to this page. Paste one of these snippets under the README's title. Keep the two marks around it, so that the badge can be replaced in place and is never added twice. OSCR changes nothing in your repository.

Markdown
<!-- oscr-badge:start -->
[![OSCR](https://openscicode.org/badge/doi/10.3390/medsci14020263.svg)](https://openscicode.org/badge/doi/10.3390/medsci14020263)
<!-- oscr-badge:end -->
reStructuredText
.. oscr-badge:start

.. image:: https://openscicode.org/badge/doi/10.3390/medsci14020263.svg
   :target: https://openscicode.org/badge/doi/10.3390/medsci14020263
   :alt: OSCR

.. oscr-badge:end
HTML
<!-- oscr-badge:start -->
<a href="https://openscicode.org/badge/doi/10.3390/medsci14020263"><img src="https://openscicode.org/badge/doi/10.3390/medsci14020263.svg" alt="OSCR"></a>
<!-- oscr-badge:end -->

On GitHub, open the README, choose Edit (the pencil), paste the snippet under the title, then "Commit changes".

If you choose to, you can have the badge kept up to date for you, with oscr badge --write from the command line or with the GitHub Action in your repository's workflows. Every other form and option (a card, AsciiDoc, LaTeX, shields.io) is on the badge page.

Request its removal

To ask OSCR to remove this record, the copies of its authors' scripts or its tracing map, use the removal request page: signed in, you say who you are, what to remove and why, then review and confirm the request. Published rules decide every request (how).

Discussion, reproductions, activity

Discussion: code errors and mismatches between the code and the paper about this paper and its code, from readers and its authors, beside the paper and its tracing map.

Report a mismatch between the code and the paper · Report a code error (signed in)

Reproductions: reports from readers who ran the authors' code and did not get the paper's result, what they ran, with which environment, commit and data.

Report a reproduction failure (signed in)

Versions of its code: the releases of its code tied to a version of this paper (the preprint, the accepted manuscript, the version of record…), each with the tracing map versioned with it.

Activity: what happens around this paper: new versions of its record, its map's validation, discussions and reproductions. It opens with sign-in.