Dual path sub-type diagnostic model based on the anatomic morphological feature

Lingling Fang, Yumeng Jiang

Abstract: Highlights•The anatomic morphological features are extracted to complete the learning of prior features, ensure the integrity of internal information of intracranial hematoma.•The dual path sub-type diagnostic model (AM-DSNet) is constructed.•The input of dual modules can extract and integrate feature information. The dense convolution module provides effective feature supplementation.•Cooperates with the hospital to maximize the simulation of the clinical diagnosis environment.•Uses the clinical data samples as the training set and test set to minimize the medical cost.