A clinician-defensible model is not one technique — it's a stack: calibration, interpretability, pediatric-specific data, and external validation. Each theme contributes one layer.
We design transformer architectures that are calibrated, robust, and Lipschitz-bounded — so a model's confidence has clinical meaning. LRFormer (UAI 2023) provides a theoretical guarantee for uncertainty estimation in single-forward-pass vision models, addressing the overconfidence that has kept transformers off the wards.
ViTASD (ICASSP 2023) introduced a robust ViT baseline for pediatric ASD facial diagnosis — distilling knowledge from large facial-expression datasets and adding a Gaussian-process decoder for clinical robustness.
We build and release open tools where pediatric medicine has been overlooked. AggPose (IJCAI 2022) is a large-scale infant pose-estimation dataset and a high-resolution transformer model intended to support pediatric research on early neurodevelopmental assessment. Datasets, weights, and training recipes are released under permissive licenses.
Cui C, Ma Y, Cao X, Ye W, …, Cao J, Wang Z, Zheng C.
Ma Y, Lu J, Cui C, Zhao S, Cao X, Ye W, Wang Z.
Ye W, Ma Y, Cao X, Tang K.
Cao X, Ye W, Sizikova E, Bai X, Coffee M, Zeng H, Cao J.
Cao X, Li X, Ma L, Huang Y, Feng X, Chen Z, Zeng H, Cao J.
Liang K, Cao X, Liao KD, Gao T, Ye W, Chen Z, Cao J, Nama T, Sun J.