HistoEffiCrossFormer: Cross-Attention CNN–Transformer Fusion with Multi-Scale Tokens for Ovarian Cancer Histopathology Classification

Roseline Oluwaseun Ogundokun, Rotimi-Williams Bello, Pius Adewale Owolawi, Chunling Tu, Sunday Agbolade, Tosho Abdulahi AbdulRahman
Proceedings of IndabaX Nigeria 2026: Building Scalable AI That Works: From Research to Deployment in Resource-Constrained Environments, PMLR 319:320-330, 2026.

Abstract

We propose HistoEffiCrossFormer, a lightweight CNN–Transformer hybrid for five-class classification of ovarian cancer histopathology images. The pipeline uses Macenko-inspired stain normalisation, EfficientNet-B0 feature extraction, squeeze-and-excitation channel attention, multi-scale tokenisation, a 2-layer transformer encoder, and cross-attention fusion. Benchmarked against SqueezeNet, ShuffleNetV2, AlexNet, and Xception, HistoEffiCrossFormer achieves 0.8267 test accuracy with AUC 0.974, outperforming lightweight CNN baselines (0.7067–0.7867 accuracy) and closely matching Xception’s AUC (0.975), motivating further validation in external cohorts and whole-slide pipelines.

Cite this Paper

BibTeX
@InProceedings{pmlr-v319-ogundokun26a, title = {{HistoEffiCrossFormer}: Cross-Attention {CNN}–{Transformer} Fusion with Multi-Scale Tokens for Ovarian Cancer Histopathology Classification}, author = {Ogundokun, Roseline Oluwaseun and Bello, Rotimi-Williams and Owolawi, Pius Adewale and Tu, Chunling and Agbolade, Sunday and AbdulRahman, Tosho Abdulahi}, booktitle = {Proceedings of IndabaX Nigeria 2026: Building Scalable AI That Works: From Research to Deployment in Resource-Constrained Environments}, pages = {320--330}, year = {2026}, editor = {Folorunso, Sakinat and Ogundokun, Roseline and Oladipo, Francisca}, volume = {319}, series = {Proceedings of Machine Learning Research}, month = {11--14 May}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v319/main/assets/ogundokun26a/ogundokun26a.pdf}, url = {https://proceedings.mlr.press/v319/ogundokun26a.html}, abstract = {We propose HistoEffiCrossFormer, a lightweight CNN–Transformer hybrid for five-class classification of ovarian cancer histopathology images. The pipeline uses Macenko-inspired stain normalisation, EfficientNet-B0 feature extraction, squeeze-and-excitation channel attention, multi-scale tokenisation, a 2-layer transformer encoder, and cross-attention fusion. Benchmarked against SqueezeNet, ShuffleNetV2, AlexNet, and Xception, HistoEffiCrossFormer achieves 0.8267 test accuracy with AUC 0.974, outperforming lightweight CNN baselines (0.7067–0.7867 accuracy) and closely matching Xception’s AUC (0.975), motivating further validation in external cohorts and whole-slide pipelines.} }
Endnote
%0 Conference Paper %T HistoEffiCrossFormer: Cross-Attention CNN–Transformer Fusion with Multi-Scale Tokens for Ovarian Cancer Histopathology Classification %A Roseline Oluwaseun Ogundokun %A Rotimi-Williams Bello %A Pius Adewale Owolawi %A Chunling Tu %A Sunday Agbolade %A Tosho Abdulahi AbdulRahman %B Proceedings of IndabaX Nigeria 2026: Building Scalable AI That Works: From Research to Deployment in Resource-Constrained Environments %C Proceedings of Machine Learning Research %D 2026 %E Sakinat Folorunso %E Roseline Ogundokun %E Francisca Oladipo %F pmlr-v319-ogundokun26a %I PMLR %P 320--330 %U https://proceedings.mlr.press/v319/ogundokun26a.html %V 319 %X We propose HistoEffiCrossFormer, a lightweight CNN–Transformer hybrid for five-class classification of ovarian cancer histopathology images. The pipeline uses Macenko-inspired stain normalisation, EfficientNet-B0 feature extraction, squeeze-and-excitation channel attention, multi-scale tokenisation, a 2-layer transformer encoder, and cross-attention fusion. Benchmarked against SqueezeNet, ShuffleNetV2, AlexNet, and Xception, HistoEffiCrossFormer achieves 0.8267 test accuracy with AUC 0.974, outperforming lightweight CNN baselines (0.7067–0.7867 accuracy) and closely matching Xception’s AUC (0.975), motivating further validation in external cohorts and whole-slide pipelines.
APA
Ogundokun, R.O., Bello, R., Owolawi, P.A., Tu, C., Agbolade, S. & AbdulRahman, T.A.. (2026). HistoEffiCrossFormer: Cross-Attention CNN–Transformer Fusion with Multi-Scale Tokens for Ovarian Cancer Histopathology Classification. Proceedings of IndabaX Nigeria 2026: Building Scalable AI That Works: From Research to Deployment in Resource-Constrained Environments, in Proceedings of Machine Learning Research 319:320-330 Available from https://proceedings.mlr.press/v319/ogundokun26a.html.

Related Material