Deep learning-driven diagnosis of multi-type vertebra diseases based on CT images

Osteoporotic vertebral compression fractures (OVCFs) are the most common type of fragility fracture. Distinguishing between OVCFs and other types of vertebra diseases, such as old fractures (OF), Schmorl's node (SN), Kummell's disease (KD), and previous surgery (PS), is critical for subsequent surgery and treatment. Our proposed DL system can accurately diagnose four vertebra diseases (OVCF, OF, KD, and PS) and has strong potential to facilitate the accurate and rapid diagnosis of vertebral diseases.

Key points:

1) A two-stage intelligent diagnostic system using deep learning (DL) for diagnosing five vertebra diseases: OVCFs, OF, SN, KD, and PS.

2) The large-scale dataset contained 11,417 CT slices from 1,097 patients and 19,718 manually annotated vertebrae with diseases.

2) The proposed DL system achieved an average sensitivity and specificity of 0.892 and 0.989, respectively, in diagnosing four kinds of vertebra diseases (CF, OF, KD, and PS)