MultiMagNet: Multi-Magnification Fusion with Availability Masking, Magnitude Dropout, and Feature Alignment Regularisation for Robust Breast Cancer Histopathology Classification

Automated breast cancer diagnosis from haematoxylin and eosin (H&E) histopathology slides is limited by a magnification-scale gap: almost all deep learning systems train at a single, arbitrarily chosen zoom level, discarding the multi-scale information that pathologists routinely combine across $40\times$, $100\times$, $200\times$, and $400\times$ magnification. We introduce MultiMagNet, a four-stream late-fusion framework that jointly encodes all four magnifications through frozen ImageNet-pretrained ResNet-50 backbones augmented with three components: (i) availability masking, an explicit binary signal distinguishing a genuinely absent stream from an informative zero-valued embedding; (ii) magnitude dropout, which stochastically erases entire streams during training ($p=0.10$), implicitly ensembling over all $2^4-1=15$ missing-view patterns at essentially zero extra cost; and (iii) feature alignment regularisation (FAR), an ${\ell }_2$-alignment auxiliary loss enforcing coherence between co-available stream embeddings. On BreakHis (7,909 images, 82 patients), under strict patient-level partitioning, naive four-stream fusion already exceeds the strongest single-stream baseline ($200\times$, AUC $=0.890$) by 4.0 AUC points. Adding our proposed components yields a set of trade-offs rather than a single optimal model: magnitude dropout provides the best robustness to missing views, while a gated variant (E10) reaches the highest full-stream AUC-ROC ($0.935\pm 0.003$) but is fragile under partial availability. Transfer to the BACH dataset yields a macro AUC of $0.882\pm 0.016$; because BACH is single-magnification and the task is four-class, this is not directly comparable to the binary BreakHis result, but demonstrates retained discriminative capacity out-of-distribution. We evaluate the framework's clinical reliability using Monte Carlo Dropout uncertainty decomposition, Mahalanobis out-of-distribution detection (AUROC $=1.000$), temperature calibration, and stream attribution methods, resulting in a robust suite of engineering choices deployable at an inference cost of $\approx 30$ ms per sample on a single GPU.