Neuroimaging profiles of the negative affective network predict anxiety severity in patients with chronic insomnia disorder: A machine learning study

•Insomnia patients with and without anxiety disorders have common and unique brain function changes.•The specific abnormal neuroimaging profile of CID-A patients is in the right DMPFC, a central hub of the negative affective network.•A machine learning model using DMPFC-related features predicts anxiety severity in two insomnia patient cohorts.•Disconnected negative affective network may be targeted for stratified treatment of insomnia subtypes.