Exemplar Darknet19 feature generation technique for automated kidney stone detection with coronal CT images

作者：

Highlights：

• ExDark19 which is a new deep feature generation model is proposed in this paper.

• Public dataset was used to show classification ability of the ExDark19.

• ExDark19 achieved the accuracy of 99.71% (hold-out validation) and 99.22% (10-fold cross validation) on the used dataset.

摘要

•ExDark19 which is a new deep feature generation model is proposed in this paper.•Public dataset was used to show classification ability of the ExDark19.•ExDark19 achieved the accuracy of 99.71% (hold-out validation) and 99.22% (10-fold cross validation) on the used dataset.

论文关键词：ExDark19,INCA,Kidney stone detection,Pre-trained model,Transfer learning,Biomedical image classification

论文评审过程：Received 19 July 2021, Revised 7 February 2022, Accepted 2 March 2022, Available online 5 March 2022, Version of Record 8 March 2022.

论文官网地址：https://doi.org/10.1016/j.artmed.2022.102274