Degenerative adversarial neuroimage nets for brain scan simulations: Application in ageing and dementia

• Published in: Medical image analysis Vol. 75; p. 102257
• Main Authors: Ravi, Daniele, Blumberg, Stefano B., Ingala, Silvia, Barkhof, Frederik, Alexander, Daniel C., Oxtoby, Neil P.
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier B.V 01.01.2022; Elsevier BV; Elsevier
• Subjects: 4D-DANI-Net; 4D-MRI; Adversarial training; Ageing; Aging; Alzheimer Disease - diagnostic imaging; Alzheimer's disease; Benchmarks; Brain; Brain - diagnostic imaging; Customization; Deep learning; Dementia; Dementia disorders; Disease progression modelling; Evaluation; Generative models; High resolution; Humans; Image Processing, Computer-Assisted; Image quality; Image resolution; Magnetic Resonance Imaging; Medical imaging; Medical research; Neuro-image; Neurodegeneration; Neurodegenerative diseases; Neuroimaging; Quality assessment; Quality control; Synthetic-images; Time series; Training; Transfer learning; Visual perception; Weighting functions; Synthetic-images; Brain; 4D-DANI-Net; Neuro-image; Adversarial training; Disease progression modelling; Neurodegeneration; Ageing; 4D-MRI; Generative models; Dementia
• ISSN: 1361-8415; 1361-8423; 1361-8423
• DOI: 10.1016/j.media.2021.102257

•We implemented a new deep learning framework capable of synthesising realistic and accurate 4D brain MRI in ageing and Alzheimer’s disease.•We proposed a sequence of memory-efficient techniques designed to improve model stability, reduce artefacts, and improve individualization.•Synthesised T1w MRI scans contain only minor structural differences with real data, and have minimal noise/texture artefacts.•Synthesised MRI scans were diagnostically indistinguishable from real scans.•Synthetic MRI can be used for: i) data augmentation, ii) model validation and iii) understanding biological/disease mechanisms in the brain. [Display omitted] Accurate and realistic simulation of high-dimensional medical images has become an important research area relevant to many AI-enabled healthcare applications. However, current state-of-the-art approaches lack the ability to produce satisfactory high-resolution and accurate subject-specific images. In this work, we present a deep learning framework, namely 4D-Degenerative Adversarial NeuroImage Net (4D-DANI-Net), to generate high-resolution, longitudinal MRI scans that mimic subject-specific neurodegeneration in ageing and dementia. 4D-DANI-Net is a modular framework based on adversarial training and a set of novel spatiotemporal, biologically-informed constraints. To ensure efficient training and overcome memory limitations affecting such high-dimensional problems, we rely on three key technological advances: i) a new 3D training consistency mechanism called Profile Weight Functions (PWFs), ii) a 3D super-resolution module and iii) a transfer learning strategy to fine-tune the system for a given individual. To evaluate our approach, we trained the framework on 9852 T1-weighted MRI scans from 876 participants in the Alzheimer’s Disease Neuroimaging Initiative dataset and held out a separate test set of 1283 MRI scans from 170 participants for quantitative and qualitative assessment of the personalised time series of synthetic images. We performed three evaluations: i) image quality assessment; ii) quantifying the accuracy of regional brain volumes over and above benchmark models; and iii) quantifying visual perception of the synthetic images by medical experts. Overall, both quantitative and qualitative results show that 4D-DANI-Net produces realistic, low-artefact, personalised time series of synthetic T1 MRI that outperforms benchmark models.