Bone density prediction via CT scans to estimate T-scores is crucial, providing a more precise assessment of bone health compared to traditional methods like X-ray bone density tests, which lack spatial resolution and the ability to detect localized changes. However, CT-based prediction faces two major challenges: the high computational complexity of transformer-based architectures, which limits their deployment in portable and clinical settings, and the imbalanced, long-tailed distribution of real-world hospital data that skews predictions. To address these issues, we introduce MedConv, a convolutional model for bone density prediction that outperforms transformer models with lower computational demands. We also adapt Bal-CE loss and post-hoc logit adjustment to improve class balance. Extensive experiments on our AustinSpine dataset shows that our approach achieves up to 21% improvement in accuracy and 20% in ROC AUC over previous state-of-the-art methods.
@article{arxiv.2502.00631,
title = {MedConv: Convolutions Beat Transformers on Long-Tailed Bone Density Prediction},
author = {Xuyin Qi and Zeyu Zhang and Huazhan Zheng and Mingxi Chen and Numan Kutaiba and Ruth Lim and Cherie Chiang and Zi En Tham and Xuan Ren and Wenxin Zhang and Lei Zhang and Hao Zhang and Wenbing Lv and Guangzhen Yao and Renda Han and Kangsheng Wang and Mingyuan Li and Hongtao Mao and Yu Li and Zhibin Liao and Yang Zhao and Minh-Son To},
journal= {arXiv preprint arXiv:2502.00631},
year = {2025}
}