aiqbench analysis
This study introduces AMN, a hybrid segmentation framework that effectively combines Swin Transformer and ResNet-50 encoders to simultaneously capture local texture and global spatial context in histopathology images. By employing a dynamic gating mechanism and a multi-objective loss function that includes uncertainty modulation, the model achieves superior generalization across datasets without retraining, addressing limitations of isolated encoder architectures. The results demonstrate robust performance on challenging nuclei classes, offering a practical advancement for medical imaging applications requiring high-fidelity boundary detection and classification.
AMN: An Adaptive Multi-Scale Fusion Network with Boundary and Uncertainty Modeling for Nuclei Segmentation
arXiv:2606.07633v1 Announce Type: cross Abstract: Accurate classification of nuclei subtypes in histopathology images is critical for downstream tasks including tumor grading, immune infiltrate quantification, and prognosis prediction. Existing approaches rely on either convolutional or transformer-based encoders in isolation, limiting their ability to simultaneously capture fine-grained local texture and long-range spatial context. We present AMN (Adaptive Multi-Scale Nuclei Network), a dual-encoder segmentation framework that jointly leverages a Swin Transformer and a ResNet-50 feature pyramid, fused via a learned per-channel gating mechanism that dynamically weighs each encoder's contribution at every scale. AMN is trained with a multi-objective loss combining class-weighted focal loss, boundary-aware loss with positive-pixel emphasis, and a novel uncertainty-modulated classification term that suppresses overconfident erroneous predictions. Evaluated on the CoNIC benchmark across seven nuclei classes, AMN achieves a mean Dice of 0.82 and mean F1 of 0.68, with an F1 of 0.67 on the diagnostically challenging lymphocyte class. AMN outperforms eight baseline models spanning pure-CNN, pure-transformer, and recent hybrid architectures: U-Net, ResU-Net, DeepLabV3+, SegNet, ViT-Small, HmsU-Net, ConvFormer-UNet, and BEFUnet. Cross-dataset evaluation on MoNuSeg demonstrates strong generalization without retraining and validating the domain robustness of the learned representations.
Abstract:Accurate classification of nuclei subtypes in histopathology images is critical for downstream tasks including tumor grading, immune infiltrate quantification, and prognosis prediction. Existing approaches rely on either convolutional or transformer-based encoders in isolation, limiting their ability to simultaneously capture fine-grained local texture and long-range spatial context. We present AMN (Adaptive Multi-Scale Nuclei Network), a dual-encoder segmentation framework that jointly leverages a Swin Transformer and a ResNet-50 feature pyramid, fused via a learned per-channel gating mechanism that dynamically weighs each encoder's contribution at every scale. AMN is trained with a multi-objective loss combining class-weighted focal loss, boundary-aware loss with positive-pixel emphasis, and a novel uncertainty-modulated classification term that suppresses overconfident erroneous predictions. Evaluated on the CoNIC benchmark across seven nuclei classes, AMN achieves a mean Dice of 0.82 and mean F1 of 0.68, with an F1 of 0.67 on the diagnostically challenging lymphocyte class. AMN outperforms eight baseline models spanning pure-CNN, pure-transformer, and recent hybrid architectures: U-Net, ResU-Net, DeepLabV3+, SegNet, ViT-Small, HmsU-Net, ConvFormer-UNet, and BEFUnet. Cross-dataset evaluation on MoNuSeg demonstrates strong generalization without retraining and validating the domain robustness of the learned representations.
Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
I.4.6; I.4.8; I.2.10
arXiv:2606.07633 [cs.CV]
(or arXiv:2606.07633v1 [cs.CV] for this version)
https://doi.org/10.48550/arXiv.2606.07633 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 285 words
Related coverage
Amazon Research Awards recipients announced
Amazon Science · May 27, 2026
AI and Quantum Computing Briefing
livetradingnews.com · Aug 4, 2026
State of the Field: AI for Epistemics and Coordination
LessWrong · Oct 2, 2026
Edge-Aware and Content-Adaptive Infrared Gas Leak Detection for Industrial Safety Monitoring
arXiv · Oct 1, 2026