Classification and diagnosis of cervical cancer with stacked autoencoder and softmax classification

作者：

Highlights：

• The stacked autoencoder model we proposed in the study was used for the first time on cervical data.

• Classification accuracy of 97.25% was attained using stacked autoencoder – softmax model.

• Results were better than existing approaches.

• Developing a clinical decision support system to aid practitioners.

摘要

•The stacked autoencoder model we proposed in the study was used for the first time on cervical data.•Classification accuracy of 97.25% was attained using stacked autoencoder – softmax model.•Results were better than existing approaches.•Developing a clinical decision support system to aid practitioners.

论文关键词：Cervical cancer,Machine learning,Deep learning,Stacked autoencoder

论文评审过程：Received 4 April 2018, Revised 28 August 2018, Accepted 28 August 2018, Available online 29 August 2018, Version of Record 4 September 2018.

论文官网地址：https://doi.org/10.1016/j.eswa.2018.08.050