Lesion feature enhancement and boundary-aware fusion for pulmonary embolism recognition on CT images

1

An automatic recognition method for pulmonary embolism in CT images combines lesion feature enhancement and boundary-related structural cue fusion.
2

The proposed framework includes a Pulmonary Embolus Feature Enhancement Module and a Vascular Boundary Aware Fusion Module for improved recognition.
3

The method achieved an Accuracy of 0.956, Precision of 0.961, Recall of 0.951, and AUC of 0.971 in internal evaluations on 523 CT cases.
4

External validation showed Accuracy values of 0.779 and 0.672 on the FUMPE and RSNA datasets, indicating potential cross-domain generalization.
5

The study addresses challenges in pulmonary embolism recognition, aiming to enhance model performance and support computer-aided screening.

1