https://doi.org/10.1140/epjp/s13360-025-07244-6
Regular Article
Optimization with R2U-Net nuclei segmentation and NASLe forward fractional network cell classification in pan-cancer histology images
1
Department of Computer Science and Engineering, Saveetha School of Engineering, Saveetha Institute of Medical and Technical Sciences (SIMATS), 602105, Chennai, Tamil Nadu, India
2
Department of Computer Science and Engineering, Lendi Institute of Engineering and Technology, 535005, Vizianagaram, India
3
Department of Information Technology, School of Computer Science Engineering and Information Systems, Vellore Institute of Technology, 632014, Vellore, India
4
Department of Computer Science and Engineering, SRM Institute of Science and Technology, Ramapuram Campus, Chennai, India
5
Department of Information Technology, Saveetha Engineering College, Chennai, India
6
Department of Electronics and Communication Engineering, Mahendra Engineering College, Namakkal, India
a
This email address is being protected from spambots. You need JavaScript enabled to view it.
Received:
25
September
2025
Accepted:
19
December
2025
Published online:
14
January
2026
Abstract
Nuclei segmentation and cell classification in histopathologic images (HI) are considered a vital and significant task for cancer analysis and prediction. However, it is a demanding task owing to the nuclei clustering, as well as dimensions with overlapping borders. In this research, the Taylor elk herd optimization with recurrent residual U-Net is developed for nuclei segmentation, and the neuron attention stage-by-stage LeNet forward fractional network (NASLe FF-Net) is developed for cell classification in pan-cancer histology images. At first, the collected HIs are pre-preprocessed with the adaptive Gaussian filtering, and then nuclei segmentation is done through the recurrent residual U-Net with the developed Taylor elk herd optimization, which is the combination of both Taylor series and elk herd optimizer. Subsequently, data augmentation and feature extraction are performed. Furthermore, the extracted features are classified based on the NASLe FF-Net, which combined the neuron attention stage-by-stage net as well as LeNet with fractional calculus. Thus, the NASLe FF-Net achieved an accuracy of 0.944, a true positive rate of 0.968, and a true negative rate of 0.928 with an image size of 40.
Copyright comment Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
© The Author(s), under exclusive licence to Società Italiana di Fisica and Springer-Verlag GmbH Germany, part of Springer Nature 2026
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
