OSCR

A multi-domain graph-integrated neural framework for robust acoustic anomaly detection under adverse environmental conditions.

Code ↔ Paper

30 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 30 matches · 8 of them tie a paragraph to a whole file, not to given lines: weak matches, whose lines are not tinted
  1. [1] § Materials and methods › Model architecture › Input representations ↔ utils/preprocessing.py, the whole file · a weak match · score 0.91 · sinusoidal reparameterization, Gammatone Filterbank Processing, gammatone filters, gammatone spectrogram, Phase Encoding, STFT
  2. [2] § Materials and methods › Theoretical foundations and novelty analysis › Graph-Theoretic Multi-Aspect Feature Integration (GTMAFI) ↔ models/loss_functions.py, the whole file · a weak match · score 0.91 · inverse square law, learned adjacency matrix, attenuation coefficient, physics informed regularization, governing, enforces
  3. [3] § Materials and methods › Model architecture ↔ models/model_architecture.py, the whole file · a weak match · score 0.90 · Hierarchical Cross Modal, Graph Theoretic Multi, graph integrated, AnomalyAudioNet, FiLM, acoustic event
  4. [4] § Materials and methods › Theoretical foundations and novelty analysis › Triple-stream encoders ↔ models/backbone.py, lines 40–77 · score 0.90 · Auditory Perceptual Representation, ResNet, filterbank response, log energy, auditory spectrogram, backbone
  5. [5] § Materials and methods › Theoretical foundations and novelty analysis › Hierarchical Cross-Domain Transformer (HCMT) ↔ models/fusion.py, lines 58–115 · score 0.88 · Gated Modality Refinement, hierarchical fused embedding, triadic fusion, sigmoid, aggregate, pairwise
  6. [6] § Materials and methods › Theoretical foundations and novelty analysis › Memory-Augmented Contrastive Learning Module (MACLM) ↔ models/memory.py, lines 38–167 · score 0.86 · momentum updated, prototype contrast, class prototype, event classes, contrastive loss, memory
  7. [7] § Materials and methods › Model architecture ↔ models/model_architecture.py, the whole file · a weak match · score 0.85 · Hierarchical Cross Modal, Graph Theoretic Multi, calibrated uncertainty, AnomalyAudioNet, FiLM, complex spectrogram
  8. [8] § Materials and methods › Model architecture ↔ models/fusion.py, lines 58–115 · score 0.85 · gated refinement, pairwise cross attention, Hierarchical Cross Modal, triadic fusion, aggregates, HCMT
  9. [9] § Materials and methods › Theoretical foundations and novelty analysis › Memory-Augmented Contrastive Learning Module (MACLM) ↔ models/memory.py, lines 38–167 · score 0.84 · rare class memory, memory augmented contrastive, contrastive objective, rare event, momentum, boundary
  10. [10] § Materials and methods › Training configuration and protocol › Multi-task joint optimization (Stage 2) ↔ training/trainer.py, lines 51–164 · score 0.84 · cosine annealing, AdamW, weight decay, scheduled, epochs, validation
  11. [11] § Materials and methods › Theoretical foundations and novelty analysis › Triple-stream encoders ↔ utils/preprocessing.py, the whole file · a weak match · score 0.78 · concatenated coefficient maps, discrete wavelet transform, Daubechies, stack, signal, temporal
  12. [12] § Materials and methods › Theoretical foundations and novelty analysis › Environmental Conditioning (E-FiLM) ↔ models/memory.py, lines 5–35 · score 0.78 · dynamically adapts, Linear Modulation, wind direction, FiLM, ambient, speed
  13. [13] § Materials and methods › Experimental setup › Hyperparameter configuration ↔ training/trainer.py, lines 51–164 · score 0.77 · cosine annealing learning, AdamW, weight decay, scheduler, optimization
  14. [14] § Materials and methods › Model architecture ↔ models/fusion.py, lines 118–184 · score 0.76 · physics informed graph, Graph Theoretic Multi, fused representations, refined, edges, nodes
  15. [15] § Materials and methods › Theoretical foundations and novelty analysis › Triple-stream encoders ↔ models/backbone.py, lines 80–96 · score 0.74 · phase amplitude coupling, complex convolutional layers, imaginary, GeLU, magnitude, channels
  16. [16] § Materials and methods › Theoretical foundations and novelty analysis › Graph-Theoretic Multi-Aspect Feature Integration (GTMAFI) ↔ models/loss_functions.py, the whole file · a weak match · score 0.73 · energy decay, spatial distance, physics informed, proxy, attenuation, adjacency
  17. [17] § Materials and methods › Training configuration and protocol › Fine-tuning with environmental adaptation (stage 3) ↔ training/trainer.py, lines 167–208 · score 0.72 · adaptation loss, Fine tuning, FiLM, distillation, model
  18. [18] § Materials and methods › Theoretical foundations and novelty analysis › Graph-Theoretic Multi-Aspect Feature Integration (GTMAFI) ↔ models/fusion.py, lines 118–184 · score 0.72 · Global Graph Pooling, GCN, learnable, semantically, node, weighted
  19. [19] § Materials and methods › Evaluation metrics ↔ evaluation/metrics.py, lines 5–31 · score 0.72 · AUC ROC, F1 score, macro, recall, precision, metrics
  20. [20] § Materials and methods › Datasets and preprocessing › Data splitting protocol ↔ scripts/download_dataset.sh, the whole file · a weak match · score 0.70 · FSD50K, VGGSound, UrbanSound8K, MIMII, DCASE, ESC
  21. [21] § Materials and methods › Theoretical foundations and novelty analysis › Triple-stream encoders ↔ models/backbone.py, lines 5–37 · score 0.68 · coefficient maps, Multi Resolution, dilated, dilations, stack, convolutional
  22. [22] § Materials and methods › Distance label acquisition and annotation ↔ scripts/download_dataset.sh, the whole file · a weak match · score 0.68 · FSD50K, VGGSound, UrbanSound8K, ESC, audio
  23. [23] § Materials and methods › Theoretical foundations and novelty analysis › Triple-stream encoders ↔ models/backbone.py, lines 127–162 · score 0.63 · Cross Domain Harmonization, layer normalization, linear, embeddings, stream, encoders
  24. [24] § Materials and methods › Theoretical foundations and novelty analysis › Triple-stream encoders ↔ models/backbone.py, lines 127–162 · score 0.63 · embedding manifold, triple stream, geometry, complex spectrogram, encoder, phase
  25. [25] § Experimental results › Limitation ↔ inference/predict.py, lines 8–62 · score 0.62 · wind speed, Environmental metadata, wind direction, inference, humidity, temperature
  26. [26] § Materials and methods › Theoretical foundations and novelty analysis › Environmental Conditioning (E-FiLM) ↔ models/memory.py, lines 5–35 · score 0.62 · acoustic feature, FiLM, rescaling, translation, dynamically, latent
  27. [27] § Materials and methods › Training configuration and protocol ↔ training/trainer.py, lines 167–208 · score 0.59 · environmental adaptation, fine tuning, optimize, HCMT, training, GTMAFI
  28. [28] § Experimental results › Ablation study ↔ evaluation/metrics.py, lines 5–31 · score 0.54 · AUC ROC, F1 score, metrics, accuracy
  29. [29] § Materials and methods › Training configuration and protocol ↔ evaluation/evaluate.py, lines 9–64 · score 0.53 · Expected Calibration Error, AnomalyAudioNet, ECE, model
  30. [30] § Materials and methods › Model architecture › Input representations ↔ models/backbone.py, lines 40–77 · score 0.53 · auditory perception, Gammatone Filterbank, spectrogram

Paper

Loaded from Europe PMC by your browser, not stored by OSCR: doi.org · Europe PMC

The paper is loaded when this pane is shown.

The authors' code

Python · 162 lines · 6.1 KB · MIT · 6 matches

  1. import torch
  2. import torch.nn as nn
  3. import torch.nn.functional as F
  5. class WaveletEncoder(nn.Module):
  6. """
  7. Multi-Resolution Time-Frequency Decomposition.
  8. Dilated 1D convolutions over wavelet coefficient maps.
  9. """
  10. def __init__(self, in_channels, hidden_dim=512):
  11. super().__init__()
  12. # Dilations: d = {1, 2, 4, 8, 16}
  13. dilations = [1, 2, 4, 8, 16]
  14. layers = []
  16. # We start with 5 in_channels (4-level DB8 wavedec gives 5 coefficient arrays: cA4, cD4, cD3, cD2, cD1)
  17. # after resampling they are stacked.
  18. c_in = in_channels
  19. c_out = 64
  21. for d in dilations:
  22. layers.append(nn.Conv1d(c_in, c_out, kernel_size=3, padding=d, dilation=d))
  23. layers.append(nn.BatchNorm1d(c_out))
  24. layers.append(nn.GELU())
  25. c_in = c_out
  26. c_out = min(hidden_dim, c_out * 2)
  28. self.dilated_stack = nn.Sequential(*layers)
  29. self.proj = nn.Linear(c_in, hidden_dim)
  31. def forward(self, x):
  32. # x: [B, in_channels (e.g. 5), T]
  33. out = self.dilated_stack(x) # [B, C, T']
  34. # Global average pooling over time for a compact representation
  35. out = out.mean(dim=-1) # [B, C]
  36. out = self.proj(out) # [B, hidden_dim]
  37. return out
  40. class GammatoneEncoder(nn.Module):
  41. """
  42. Auditory-Perceptual Representation (Modified ResNet-18 Backbone).
  43. Processes log-energy gammatone filterbank responses.
  44. """
  45. def __init__(self, in_channels=1, hidden_dim=512):
  46. super().__init__()
  47. # Simplified ResNet-style block for auditory spectrograms
  48. self.conv1 = nn.Conv2d(in_channels, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3))
  49. self.bn1 = nn.BatchNorm2d(64)
  50. self.relu = nn.GELU()
  51. self.pool = nn.MaxPool2d(kernel_size=3, stride=2, padding=1)
  53. self.layer1 = self._make_layer(64, 128, stride=2)
  54. self.layer2 = self._make_layer(128, 256, stride=2)
  55. self.layer3 = self._make_layer(256, hidden_dim, stride=2)
  57. self.global_pool = nn.AdaptiveAvgPool2d((1, 1))
  59. def _make_layer(self, in_c, out_c, stride=1):
  60. return nn.Sequential(
  61. nn.Conv2d(in_c, out_c, kernel_size=3, stride=stride, padding=1),
  62. nn.BatchNorm2d(out_c),
  63. nn.GELU(),
  64. nn.Conv2d(out_c, out_c, kernel_size=3, padding=1),
  65. nn.BatchNorm2d(out_c),
  66. nn.GELU()
  67. )
  69. def forward(self, x):
  70. # x: [B, 1, M_bands, T]
  71. x = self.pool(self.relu(self.bn1(self.conv1(x))))
  72. x = self.layer1(x)
  73. x = self.layer2(x)
  74. x = self.layer3(x)
  76. x = self.global_pool(x).flatten(1) # [B, hidden_dim]
  77. return x
  80. class ComplexConv2d(nn.Module):
  81. """
  82. Complex convolutional layer mimicking operation on real and imaginary parts.
  83. Since phase is reparameterized as (Magnitude, Cos, Sin), this layer works on 3 channel inputs
  84. and learns grouped mappings equivalent to complex correlation.
  85. """
  86. def __init__(self, in_channels, out_channels, kernel_size=3, stride=1, padding=1):
  87. super().__init__()
  88. # In a strict algebraic sense, true complex convolutions operate on Re/Im.
  89. # Here we map [Mag, Cos, Sin] (3 channels) using a standard 2D conv with 3 input channels
  90. # which effectively learns phase-amplitude couplings.
  91. self.conv = nn.Conv2d(in_channels, out_channels, kernel_size, stride, padding)
  92. self.cbn = nn.BatchNorm2d(out_channels) # Complex Batch Norm simplified to 2D standard
  93. self.act = nn.GELU()
  95. def forward(self, x):
  96. return self.act(self.cbn(self.conv(x)))
  99. class ComplexSpectrogramEncoder(nn.Module):
  100. """
  101. Phase-Preserving Representation.
  102. Operates on [Magnitude, Cos(Phase), Sin(Phase)] using 'Complex' Convolutions.
  103. """
  104. def __init__(self, channels=3, hidden_dim=512):
  105. super().__init__()
  106. # Input channels = 3
  107. self.conv1 = ComplexConv2d(channels, 64, kernel_size=(5, 5), stride=(2, 2), padding=2)
  108. self.pool1 = nn.MaxPool2d(kernel_size=2, stride=2)
  110. self.conv2 = ComplexConv2d(64, 128, kernel_size=3, stride=2)
  111. self.conv3 = ComplexConv2d(128, 256, kernel_size=3, stride=2)
  112. self.conv4 = ComplexConv2d(256, hidden_dim, kernel_size=3, stride=2)
  114. self.global_pool = nn.AdaptiveAvgPool2d((1, 1))
  116. def forward(self, x):
  117. # x: [B, 3, F, T]
  118. x = self.pool1(self.conv1(x))
  119. x = self.conv2(x)
  120. x = self.conv3(x)
  121. x = self.conv4(x)
  123. x = self.global_pool(x).flatten(1) # [B, hidden_dim]
  124. return x
  127. class TripleStreamEncoder(nn.Module):
  128. """
  129. Unifies Wavelet, Gammatone, and Complex Spectrogram streams into a
  130. common embedding manifold geometry via learned cross-domain projections.
  131. """
  132. def __init__(self, hidden_dim=512):
  133. super().__init__()
  134. # Using 5 input channels for Daubechies 4-level decomposition (cA4, cD4, cD3, cD2, cD1)
  135. self.wavelet_enc = WaveletEncoder(in_channels=5, hidden_dim=hidden_dim)
  137. # Using 1 channel for the approximated 2D Gammatone-like spectrogram
  138. self.gammatone_enc = GammatoneEncoder(in_channels=1, hidden_dim=hidden_dim)
  140. # Using 3 channels for Magnitude, Cos(Phase), Sin(Phase)
  141. self.complex_enc = ComplexSpectrogramEncoder(channels=3, hidden_dim=hidden_dim)
  143. # Cross-Domain Harmonization and Projection
  144. self.ln_w = nn.LayerNorm(hidden_dim)
  145. self.ln_g = nn.LayerNorm(hidden_dim)
  146. self.ln_c = nn.LayerNorm(hidden_dim)
  148. self.proj_w = nn.Linear(hidden_dim, hidden_dim)
  149. self.proj_g = nn.Linear(hidden_dim, hidden_dim)
  150. self.proj_c = nn.Linear(hidden_dim, hidden_dim)
  152. def forward(self, x_wav, x_gam, x_spec):
  153. z_w = self.wavelet_enc(x_wav)
  154. z_g = self.gammatone_enc(x_gam)
  155. z_c = self.complex_enc(x_spec)
  157. # Harmonize and Project (Eq 24)
  158. z_w = self.ln_w(self.proj_w(z_w))
  159. z_g = self.ln_g(self.proj_g(z_g))
  160. z_c = self.ln_c(self.proj_c(z_c))
  162. return z_w, z_g, z_c

backbone.py at commit 39deafc, under MIT · at the source

Overview

Authors: Ayan Sar1, Sumit Aich1, Pranav Singh Puri1, Sampurna Roy1, Tanupriya Choudhury1, Lubna Abdelkareim Gabralla2, Minakshi3, Bhupesh Kumar Dewangan4
  1. School of Computer Science, University of Petroleum and Energy Studies,Dehradun, 248007 Uttarakhand India
  2. Department of Computer Science, Applied College, Princess Nourah bint Abdulrahman University,P.O. Box 84428, Riyadh, 11671 Saudi Arabia
  3. College of Computer Science, King Khalid University,Abha, Saudi Arabia
  4. Symbiosis Institute of Technology, Nagpur Campus, Symbiosis International (Deemed University),Pune, India
Institutions: University of Petroleum and Energy Studies (India); Princess Nourah bint Abdulrahman University (Saudi Arabia); King Khalid University (Saudi Arabia); Symbiosis International University (India)
Journal: Scientific reports, volume 16, issue 1, article 23837
Dates: received 23 November 2025; accepted 14 May 2026; published online 25 May 2026
Type: Research article · Language: English
License: CC BY
Identifiers: DOI 10.1038/s41598-026-53876-8 · PMID 42185565 · PMCID PMC13433785 · OpenAlex W7162335557
Open access: gold, a free copy (OpenAlex)
Status: code verified
Categories: Parkinson's (population), methods / tools (subfield)
Methods: Connectivity, Preprocessing, Spectral & time-frequency, Statistics, Machine learning, Smoothing, state filtering, decompositions
Keywords: Deep learning, Early diagnosis, Multimodal fusion, Neurodegenerative disease, Parkinson’s detection, Engineering, Mathematics and computing
Topic: Music and Audio Processing (Signal Processing, Computer Science), according to OpenAlex
Funding: Princess Nourah bint Abdulrahman University Researchers Supporting Project (PNURSP2025R178)
Citations: not cited yet (Europe PMC); 30 references in the paper

Abstract

Acoustic event understanding and anomaly detection play a critical role in security monitoring, defence applications, and urban safety, yet current approaches struggle to generalise across noisy environments, rare event categories, and varying distances. To address these limitations, we propose a multi-domain, graph-integrated neural framework that unifies spectral, temporal, and phase representations for robust acoustic modeling. Our architecture combines a triple-stream decomposition - wavelet, gammatone, and complex spectrogram encoders - with a hierarchical cross-modal transformer for multi-scale fusion. Graph-theoretic feature integration, informed by physical propagation constraints, enables robust representation learning, while a memory-augmented contrastive module enhances recognition of rare events. The framework is trained with multi-task objectives encompassing classification, uncertainty-aware distance estimation, and environment-conditioned adaptation. Evaluations across seven benchmark datasets, including UrbanSound8K, ESC-50, FSD50K, DCASE, and MAD, demonstrate strong multi-task performance across classification, distance estimation, and uncertainty quantification. The framework achieves robust generalization under 0-10 dB noise degradation with relative performance degradation below 12%, mean absolute error of 0.73-1.12m for controlled-condition distance estimation on datasets with ground-truth spatial annotatins (MAD, DCASE, MIMII), and 1.24-1.68m on ground-truth annotations from extended range intervals with aggregate MAE of 6.39m across all datasets inclusive of those with physics-simulation-derived labels and superior calibration (ECE < 0.04 on primary benchmarks). Furthermore, the model achieves superior calibration and rare-event detection compared to leading transformer-based baselines. These results demonstrate that multi-domain, physics-aware acoustic modeling yields substantial robustness and multi-task advantages over single-task classification baselines, particularly under adverse environmental conditions and for rare event categories, with implications for real-time deployment in defense monitoring, disaster response, and smart city surveillance.

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

Repository

Its files are read in the Code ↔ Paper reader above, with 30 matches between paragraphs and lines of code.

sumit945/Anomaly-Detection

License: MIT
State: the link answers, verified on 28 September 2026
Evidence: files inventoried
Commit: 39deafc204114c444a169eac91f6da022beb9d93, 13 March 2026
Languages: Python (19), Shell (2), Jupyter (1)
Size: 31 files, 22 scripts
Software Heritage: not archived
Found in: “Code availability”
Holds: README, CITATION.cff, environment (environment.yml, requirements.txt, setup.py), tests, documentation, 1 notebook
Not found: license file, continuous integration
Tools: PyTorch (14 files), NumPy (3 files), PyTorch Geometric (2 files), PyWavelets (2 files), SciPy (2 files), scikit-learn (1 file)
Availability: 1 check, the latest on 28 September 2026: the link answers
  • 28 September 2026: the link answers
23 files
At the source: github.com/sumit945/Anomaly-Detection

Code availability

The complete implementation of the AnomalyAudioNet framework, including all model architectures, training scripts, preprocessing pipelines, evaluation protocols, and configuration files for reproducing all reported experiments, is publicly available at Github Link (https://github.com/sumit945/Anomaly-Detection).

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

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:

  • 1 repository of the authors' code, each at its verified commit, with its license and how the link was found in the paper;
  • 22 scripts, each with its path and the digest of its content;
  • 30 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

Datasets cited

Data availability

The data used in this research are compiled and available at AudioSet-Military Dataset (https://www.kaggle.com/datasets/junewookim/mad-dataset-military-audio-dataset), UrbanSound8K Dataset (https://urbansounddataset.weebly.com/urbansound8k.html), ESC-50 Dataset (https://github.com/karolpiczak/ESC-50), FSD50K Dataset (https://zenodo.org/records/4060432), DCASE 2020 Task 1 A Dataset (https://dcase.community/challenge2020/index), VGGSound Dataset (https://www.kaggle.com/datasets/codebreaker619/vggsound), MIMII Dataset (https://zenodo.org/records/3384388).

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, 8 authors, 7 keywords, 1 funder, 21 references.

Cite

This paper

Sar, A., Aich, S., Puri, P. S., Roy, S., Choudhury, T., Gabralla, L. A., Minakshi, & Dewangan, B. K. (2026). A multi-domain graph-integrated neural framework for robust acoustic anomaly detection under adverse environmental conditions. Scientific reports, 16(1), 23837. https://doi.org/10.1038/s41598-026-53876-8

BibTeX

@article{sar2026multi,
author = {Sar, Ayan and Aich, Sumit and Puri, Pranav Singh and Roy, Sampurna and Choudhury, Tanupriya and Gabralla, Lubna Abdelkareim and Minakshi and Dewangan, Bhupesh Kumar},
title = {{A multi-domain graph-integrated neural framework for robust acoustic anomaly detection under adverse environmental conditions}},
journal = {Scientific reports},
year = {2026},
month = may,
volume = {16},
number = {1},
pages = {23837},
publisher = {Nature Publishing Group},
issn = {2045-2322},
doi = {10.1038/s41598-026-53876-8},
url = {https://doi.org/10.1038/s41598-026-53876-8},
pmid = {42185565},
pmcid = {PMC13433785}
}

RIS

TY - JOUR
AU - Sar, Ayan
AU - Aich, Sumit
AU - Puri, Pranav Singh
AU - Roy, Sampurna
AU - Choudhury, Tanupriya
AU - Gabralla, Lubna Abdelkareim
AU - Minakshi
AU - Dewangan, Bhupesh Kumar
TI - A multi-domain graph-integrated neural framework for robust acoustic anomaly detection under adverse environmental conditions
T2 - Scientific reports
J2 - Sci Rep
PY - 2026
DA - 2026/05/25
VL - 16
IS - 1
SP - 23837
SN - 2045-2322
PB - Nature Publishing Group
DO - 10.1038/s41598-026-53876-8
UR - https://doi.org/10.1038/s41598-026-53876-8
LA - en
ER -

CSL-JSON

{
"id": "10.1038/s41598-026-53876-8",
"type": "article-journal",
"title": "A multi-domain graph-integrated neural framework for robust acoustic anomaly detection under adverse environmental conditions",
"container-title": "Scientific reports",
"author": [
{
"family": "Sar",
"given": "Ayan"
},
{
"family": "Aich",
"given": "Sumit"
},
{
"family": "Puri",
"given": "Pranav Singh"
},
{
"family": "Roy",
"given": "Sampurna"
},
{
"family": "Choudhury",
"given": "Tanupriya"
},
{
"family": "Gabralla",
"given": "Lubna Abdelkareim"
},
{
"family": "Minakshi"
},
{
"family": "Dewangan",
"given": "Bhupesh Kumar"
}
],
"container-title-short": "Sci Rep",
"volume": "16",
"issue": "1",
"page": "23837",
"DOI": "10.1038/s41598-026-53876-8",
"PMID": "42185565",
"PMCID": "PMC13433785",
"ISSN": "2045-2322",
"publisher": "Nature Publishing Group",
"URL": "https://doi.org/10.1038/s41598-026-53876-8",
"language": "en",
"issued": {
"date-parts": [
[
2026,
5,
25
]
]
}
}

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:10.1038/s41598-026-68186-2 [code]
NeuroStream: spectral-spatio-temporal deep learning for visual stimulus classification from EEG.
Journal: Scientific reports
In common: PyWavelets, PyTorch, scikit-learn, 2 other tools, methods / tools
[2] doi:10.1038/s41598-026-55584-9 [code]
Reproducible benchmark of wavelet-enhanced intrabody communication biometric identification.
Journal: Scientific reports
In common: PyWavelets, PyTorch, scikit-learn, 2 other tools, methods / tools
[3] doi:10.1038/s41592-026-03057-2 [code]
CREsted: modeling genomic and synthetic cell-type-specific enhancers across tissues and species.
Journal: Nature methods
In common: PyWavelets, PyTorch, scikit-learn, 2 other tools, methods / tools
[4] doi:10.1093/bioinformatics/btag553 [code]
Spatial-spectral fusion enables drug repositioning by capturing indirect and long-range associations in biological networks.
Journal: Bioinformatics (Oxford, England)
In common: PyTorch Geometric, PyTorch, scikit-learn, 2 other tools, Parkinson's
[5] doi:10.1093/bib/bbag118 [code]
Drug screening for α-synuclein aggregation inhibitors via multimodal graph neural network.
Journal: Briefings in bioinformatics
In common: PyTorch Geometric, PyTorch, scikit-learn, 2 other tools, Parkinson's
[6] doi:10.3389/fsysb.2026.1873899 [code]
A systems microbiology framework for reproducible multi-dataset omics integration with application to long COVID.
Journal: Frontiers in systems biology
In common: PyTorch Geometric, PyTorch, scikit-learn, 2 other tools, methods / tools
[7] doi:10.1002/advs.77003 [code]
SemanticST: A Scalable Multi-Contextual Graph Learning Framework for Uncovering Spatial Niches and Robust Multi-Sample Integration in Spatial Transcriptomics.
Journal: Advanced science (Weinheim, Baden-Wurttemberg, Germany)
In common: PyTorch Geometric, PyTorch, scikit-learn, 2 other tools, methods / tools
[8] doi:10.1038/s41592-026-03194-8 [code]
Beyond benchmarking: an expert-guided consensus approach to spatially aware clustering.
Journal: Nature methods
In common: PyTorch Geometric, PyTorch, scikit-learn, 2 other tools, methods / tools
[9] doi:10.1093/bioinformatics/btag540 [code]
Deciphering spatial heterogeneity by multimodal spatial transcriptomics modelling with SpatialModal.
Journal: Bioinformatics (Oxford, England)
In common: PyTorch Geometric, PyTorch, scikit-learn, 2 other tools, methods / tools
[10] doi:10.21203/rs.3.rs-9676637/v1 [code]
A Comprehensive Benchmarking of Spatial Deconvolution and Domain Detection Methods across Diverse Tissues and Spatial Transcriptomic Technologies
Journal: Research Square (preprint)
In common: PyTorch Geometric, PyTorch, scikit-learn, 2 other tools, methods / tools

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.

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: questions and error reports about this paper and its code, from signed-in readers and its authors. It opens with sign-in.

Reproductions: reports from readers who ran the authors' code: what they reproduced, with which environment, commit and data. It opens with sign-in.

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