Subtle anomaly detection: Application to brain MRI analysis of de novo Parkinsonian patients

作者：

Highlights：

• Auto-encoders can detect anomalies in the MR brain scans of Parkinsonian patients (PD).

• PD anomalies were mainly detected in white matter, temporal lobes and Globus Pallidus internus.

• Anomalies in white matter were found in PD with cognitive impairment.

• An original evaluation method is proposed in absence of gold standard.

摘要

•Auto-encoders can detect anomalies in the MR brain scans of Parkinsonian patients (PD).•PD anomalies were mainly detected in white matter, temporal lobes and Globus Pallidus internus.•Anomalies in white matter were found in PD with cognitive impairment.•An original evaluation method is proposed in absence of gold standard.

论文关键词：Machine learning,Unsupervised learning,Variational autoencoder,Neurodegenerative disease,Parkinson's disease

论文评审过程：Received 8 February 2021, Revised 26 November 2021, Accepted 29 January 2022, Available online 16 February 2022, Version of Record 18 February 2022.

论文官网地址：https://doi.org/10.1016/j.artmed.2022.102251