A novel collaborative self-supervised learning method for radiomic data

• Published in: NeuroImage (Orlando, Fla.) Vol. 277; p. 120229
• Main Authors: Li, Zhiyuan, Li, Hailong, Ralescu, Anca L., Dillman, Jonathan R., Parikh, Nehal A., He, Lili
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 15.08.2023; Elsevier Limited; Elsevier
• Subjects: Alzheimer's disease; Between-subjects design; Collaboration; Collaborative learning; Computer Simulation; Data analysis; Diagnosis; Diagnosis, Computer-Assisted; Disease diagnosis; Humans; Learning; Magnetic resonance imaging; Medical prognosis; Morphology; MRI; Performance evaluation; Radiomic data; Radiomics; Self-supervised learning; Supervised Machine Learning; Tomography; Disease diagnosis; MRI; Collaborative learning; Self-supervised learning; Radiomic data

•Deep learning and radiomics to predict cognitive deficits in very preterm infants.•Self-supervised learning to address the lack of labeled radiomic data.•Collaborative self-supervised learning to learn latent radiomic representations.•Integrate pretext tasks of radiomics reconstruction and subject...