Characterizing Multi-Domain Representations for Gastric Histopathological Image Classification
- 1Centro de Investigación en Computación, Instituto Politécnico Nacional, Ciudad de México 07738, Mexico
- 2Escuela Superior de Ingeniería Mecánica y Eléctrica (ESIME), Unidad Zacatenco, Instituto Politécnico Nacional, Ciudad de México 07738, Mexico
Abstract
This study characterizes spatial, spectral, relational, and contextual representations for binary gastric histopathological image classification using FLIGENT-Net. The framework combines an ImageNet-pretrained convolutional backbone, patch-level fast Fourier transform (FFT) descriptors, graph convolution over a cosine-similarity graph, and a Transformer Encoder. Experiments use the 160-by-160 GasHisSDB subset, a persisted stratified 60%/20%/20% partition, and one training run per configuration with seed 42. In the nine completed configurations of the revised evaluation, FLIGENT-Net achieved 97.31% accuracy, 96.57% precision, 96.61% recall, 96.59% F1-score for the Abnormal class, and 0.9964 AUC-ROC. An additional native-grid FFT control yielded 97.43% accuracy and 96.74% F1 under an alternative spectral protocol that preserved the 160-by-160 sampling grid while changing the local FFT window sizes and spectral discretization. The conventional CNN baseline (ResNet18) obtained the highest accuracy and F1-score among these completed runs, at 98.51% and 98.13%, respectively. These results do not establish a predictive advantage for the complete multi-domain architecture; rather, they characterize how the added representation domains affect predictive behavior, graph topology, computational cost, and attribution response under a common protocol. The study contributes a reproducible integration of the four representation domains and a controlled empirical characterization of their combinations. Transformer attention visualizations describe information allocation among contextualized nodes, while a common input-gradient deletion protocol examines attribution behavior. Together, these analyses identify the behavior and limitations of the implemented fusion strategy and provide a basis for its further development. Performance estimates are conditional on the available patch-level split and should not be interpreted as patient-, slide-, or source-image-independent generalization; run-to-run variability is also not estimated because one fixed seed is used per configuration.
Related
- The classification of taut surface singularities via representations from logarithmic vector fields
- Evaluation of TF-IDF and Word2Vec Representations Using Support Vector Machine for Multiclass Product Category Classification of Tokopedia Reviews
- Domain-Adaptive Deep Joint Source-Channel Coding for Image Classification
- An Uncertainty-Aware Hybrid Symbolic-Lexicon Framework for Multi-Domain Sentiment Classification
- Evaluating the effectiveness of open-weight LLMs for multi-domain Bengali text classification with zero-shot and few-shot prompting
BibTeX
@article{reyesvera2026characterizing,
title = {{Characterizing Multi-Domain Representations for Gastric Histopathological Image Classification}},
author = {Zayra Reyes-Vera and Jesús Yaljá Montiel-Pérez and José Luis López-Bonilla and Elsa Rubio-Espino and Juan Humberto Sossa-Azuela and Víctor Hugo Ponce-Ponce and Osvaldo Espinosa-Sosa and Sergio Vidal-Beltrán},
journal = {Mathematics},
year = {2026},
month = oct,
doi = {10.3390/math14213867},
url = {https://www.mdpi.com/2227-7390/14/21/3867},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Characterizing Multi-Domain Representations for Gastric Histopathological Image Classification
AU - Zayra Reyes-Vera
AU - Jesús Yaljá Montiel-Pérez
AU - José Luis López-Bonilla
AU - Elsa Rubio-Espino
AU - Juan Humberto Sossa-Azuela
AU - Víctor Hugo Ponce-Ponce
AU - Osvaldo Espinosa-Sosa
AU - Sergio Vidal-Beltrán
PY - 2026
DA - 2026/10/23
JO - Mathematics
DO - 10.3390/math14213867
UR - https://www.mdpi.com/2227-7390/14/21/3867
AB - This study characterizes spatial, spectral, relational, and contextual representations for binary gastric histopathological image classification using FLIGENT-Net. The framework combines an ImageNet-pretrained convolutional backbone, patch-level fast Fourier transform (FFT) descriptors, graph convolution over a cosine-similarity graph, and a Transformer Encoder. Experiments use the 160-by-160 GasHisSDB subset, a persisted stratified 60%/20%/20% partition, and one training run per configuration with seed 42. In the nine completed configurations of the revised evaluation, FLIGENT-Net achieved 97.31% accuracy, 96.57% precision, 96.61% recall, 96.59% F1-score for the Abnormal class, and 0.9964 AUC-ROC. An additional native-grid FFT control yielded 97.43% accuracy and 96.74% F1 under an alternative spectral protocol that preserved the 160-by-160 sampling grid while changing the local FFT window sizes and spectral discretization. The conventional CNN baseline (ResNet18) obtained the highest accuracy and F1-score among these completed runs, at 98.51% and 98.13%, respectively. These results do not establish a predictive advantage for the complete multi-domain architecture; rather, they characterize how the added representation domains affect predictive behavior, graph topology, computational cost, and attribution response under a common protocol. The study contributes a reproducible integration of the four representation domains and a controlled empirical characterization of their combinations. Transformer attention visualizations describe information allocation among contextualized nodes, while a common input-gradient deletion protocol examines attribution behavior. Together, these analyses identify the behavior and limitations of the implemented fusion strategy and provide a basis for its further development. Performance estimates are conditional on the available patch-level split and should not be interpreted as patient-, slide-, or source-image-independent generalization; run-to-run variability is also not estimated because one fixed seed is used per configuration.
SN - 2227-7390
ER -