Luping Zhou

2D foundation models can dominate 3D medical imaging tasks without the massive inference overhead of per-slice encoding simply by compressing volume depth via gradient-free multiscale PCA before feature extraction.

Iterative refinement in radiology report generation can significantly enhance accuracy, with DRRG outperforming traditional methods despite using a smaller model.

Structuring supervision around disease-specific relationships leads to better radiology report generation than simply scaling up model size.

Tail classes in medical image classification can be effectively shielded from head-class encroachment, leading to significant improvements in performance.

High-resolution chest X-ray perception can be achieved without inflating token counts, leading to better diagnostic accuracy and report generation.