EUVIMEDEUVIMED entdecken
Crossref

MOSAIC: Multi-scale orientation-aware segmentation and instance classification network for histopathological image analysis

Arbab Sufyan Wadood, Mohammad Faizal Ahmad Fauzi, Lai Kuan Wong, Jenny Tung Hiong Lee, See Yee Khor, Lai-Meng Looi

Computer Methods and Programs in Biomedicine · 2026-06 · 280 · 109334

DOI: 10.1016/j.cmpb.2026.109334 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Computer Methods and Programs in Biomedicine, 2026-06. Verfasst von Arbab Sufyan Wadood und 5 weiteren Autor:innen. Beteiligte Einrichtung: Faculty of Artificial Intelligence and Engineering, Multimedia University, 63100, Cyberjaya, Selangor, Malaysia; Department of Computer Science, Balo…. Hauptthemen laut MEDLINE: Image Processing, Computer-Assisted. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 61 Literaturangaben. Gefördert von: Research and Development.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

BACKGROUND AND OBJECTIVE Accurate nuclei segmentation and instance classification are fundamental tasks in biomedical image analysis; however, many existing computational models exhibit limited robustness when confronted with scale variability, morphological heterogeneity, and arbitrary rotational orientations commonly observed in histopathological images. The objective of this work is to develop a unified computational framework that is robust to effective magnification variability, arbitrary orientations, and long-range contextual dependencies, without relying on multi-magnification supervision or magnification-specific retraining. METHODS We propose a multi-scale orientation-aware segmentation and instance classification (MOSAIC) framework, which integrates hierarchical context extraction, rotation-aware feature fusion, and transformer-based long-range contextual modeling within a single encoder-decoder architecture. The proposed model combines large-, medium-, and small-scale contextual cues derived from a single native training magnification to enable robust learning across effective magnifications. The proposed method is evaluated on an institutional estrogen receptor immunohistochemistry cohort, the multi-organ nuclei segmentation and classification dataset, and the colorectal nuclei segmentation and phenotypes dataset. RESULTS The proposed model outperforms baseline methods, achieving a mean Dice coefficient of 0.862, an Aggregated Jaccard Index of 0.721, and a Panoptic Quality score of 0.647, with consistent improvements of 3%-7% across datasets. The model also demonstrates favorable computational cost relative to representative baselines, with an inference time of 0.175 s per 512 × 512 image patch and a peak memory footprint of 3.7 GB. CONCLUSIONS The results demonstrate that orientation-aware multi-scale fusion and long-range contextual modeling improve boundary precision, instance separation, and classification consistency across heterogeneous nuclear morphologies. These improvements indicate that the proposed design generalizes reliably across challenging tissue appearances.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Image Processing, Computer-AssistedHumansCell NucleusAlgorithms

Stichwörter der Autor:innen

Breast cancerComputational pathologyDigital pathologyInstance classificationNuclei segmentation

Autor:innen und Institutionen

  • Faculty of Artificial Intelligence and Engineering, Multimedia University, 63100, Cyberjaya, Selangor, Malaysia; Department of Computer Science, Balochistan University of Information Technology, Engineering and Manageme…
  • Faculty of Artificial Intelligence and Engineering, Multimedia University, 63100, Cyberjaya, Selangor, Malaysia; School of Digital Health, KPJU Healthcare University, 71800 Nilai, Negeri Sembilan, Malaysia. Electronic a…
  • Centre for Image and Vision Computing, CoE for Artificial Intelligence, Multimedia University, 63100, Cyberjaya, Selangor, Malaysia; Faculty of Computing & Informatics, Multimedia University, 63100, Cyberjaya, Malaysia.…
  • Department of Pathology, Sarawak General Hospital, Kuching, 93586, Sarawak, Malaysia. Electronic address: hionghiong@moh.gov.my.
  • Department of Pathology, Hospital Seberang Jaya, Jalan Tun Hussein Onn, Seberang Jaya, 13700, Penang, Malaysia. Electronic address: sykhor25@gmail.com.
  • Department of Pathology, University Malaya Medical Centre, Kuala Lumpur, 59100, Malaysia. Electronic address: looilm@ummc.edu.my.

Förderung

  • Research and Development

Zitieren

APA

Wadood, A. S., Fauzi, M. F. A., Wong, L. K., Lee, J. T. H., Khor, S. Y., & Looi, L. M. (2026). MOSAIC: Multi-scale orientation-aware segmentation and instance classification network for histopathological image analysis. Computer Methods and Programs in Biomedicine, 280, 109334. https://doi.org/10.1016/j.cmpb.2026.109334

Vancouver

Wadood AS, Fauzi MFA, Wong LK, Lee JTH, Khor SY, Looi LM. MOSAIC: Multi-scale orientation-aware segmentation and instance classification network for histopathological image analysis. Computer Methods and Programs in Biomedicine. 2026;280:109334. doi:10.1016/j.cmpb.2026.109334

BibTeX
@article{Wadood2026,
  author = {Wadood, Arbab Sufyan and Fauzi, Mohammad Faizal Ahmad and Wong, Lai Kuan and Lee, Jenny Tung Hiong and Khor, See Yee and Looi, Lai-Meng},
  title = {MOSAIC: Multi-scale orientation-aware segmentation and instance classification network for histopathological image analysis},
  journal = {Computer Methods and Programs in Biomedicine},
  year = {2026},
  volume = {280},
  pages = {109334},
  doi = {10.1016/j.cmpb.2026.109334}
}
Kostenlos starten →