Search papers, labs, and topics across Lattice.
This paper introduces a hybrid deep learning framework for white blood cell classification that addresses extreme class imbalance and long-tail distributions. The framework combines Pix2Pix restoration, a Swin Transformer ensemble with MedSigLIP contrastive embeddings, and a biologically-inspired refinement step using geometric spikiness and Mahalanobis distance. Results on the WBCBench 2026 challenge demonstrate a Macro-F1 score of 0.77139, showcasing the effectiveness of integrating biological priors into deep learning for improved rare-class generalization in hematological image analysis.
By injecting biological heuristics into a deep learning pipeline, this method achieves state-of-the-art performance in classifying rare white blood cell subtypes, a task where standard deep learning models often fail.
Automated white blood cell (WBC) classification is essential for leukemia screening but remains challenged by extreme class imbalance, long-tail distributions, and domain shift, leading deep models to overfit dominant classes and fail on rare subtypes. We propose a hybrid framework for rare-class generalization that integrates a generative Pix2Pix-based restoration module for artifact removal, a Swin Transformer ensemble with MedSigLIP contrastive embeddings for robust representation learning, and a biologically-inspired refinement step using geometric spikiness and Mahalanobis-based morphological constraints to recover out-of-distribution predictions. Evaluated on the WBCBench 2026 challenge, our method achieves a Macro-F1 of 0.77139 on the private leaderboard, demonstrating strong performance under severe imbalance and highlighting the value of incorporating biological priors into deep learning for hematological image analysis.
