3D Brain MRI GAN-Based Synthesis Conditioned on Partial Volume Maps

Rusak, Filip; Cruz, Rodrigo Santa; Bourgeat, Pierrick; Fookes, Clinton; Fripp, Jürgen; Bradley, Andrew P.; Salvado, Olivier

doi:10.1007/978-3-030-59520-3_2

Cited by 8 publications

(5 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The excitation volume, as well as contrast behavior in accelerated 3D sequences 25 or related SNR changes, might be equally important. The simulation may sharpen tissue interfaces, which could be improved by using partial volume signal contributions during simulation or by signal synthesis using GANs conditioned by these maps as described in 26 . Other SNR or resolution related parameters could be easily included in the systematic test, after baseline image denoising.…”

Section: Discussionmentioning

confidence: 99%

Novel concept for systematic testing of AI models for MRI acquisition shifts with simulated data

Posselt

Gramotke

Parida³

et al. 2023

Medical Imaging 2023: Image Perception, Observer Performance, and Technology Assessment

View full text Add to dashboard Cite

Recently published guidelines recommend the validation of models for medical image processing against variations of image acquisition parameters. This work presents a novel concept for systematic tests that investigate a model's performance to MRI acquisition shifts in brain scans of multiple sclerosis patients. Contrast changes of T2-weighted FLAIR images related to changes of two scan parameters (TE, TI) are simulated using artificial data. These images were created from 114 brain MRI, normal tissue segmentation masks and expert MS lesion masks (NAMIC, OpenMS, MSSEG2, ISBI15, LC08). Contrast changes were simulated by using the FLAIR signal equation. The experiments evaluate the F1 scores of models trained on images based on different uniform and non-uniform acquisition parameters when tested on typical acquisition parameter shifts. The acquisition parameter variations of the experiments are based on guidelines for quality assurance and the analysis of protocols and images from different institutions and open datasets. The dependence of the F1 score on both parameters could be approximated by linear functions with R 2 scores of 0.94 to 0.98. Thus, linear functions can be used to model the dependence of the F1 score on the influencing factors, thereby allowing for the derivation of a "save" range of image acquisition parameters to meet a desired performance metric. The segmentation model was more sensitive to changes in TE compared to TI. The maximum lesion F1 loss when applying the models to out-of-distribution data ranged from 0.04 to 0.36 and was significant in all the experiments, even when the model was trained on data representing scans of different contrast (p<0.01). This underlines the need for testing segmentation models against acquisition shifts.

show abstract

Section: Discussionmentioning

confidence: 99%

Novel concept for systematic testing of AI models for MRI acquisition shifts with simulated data

Posselt

Gramotke

Parida³

et al. 2023

Medical Imaging 2023: Image Perception, Observer Performance, and Technology Assessment

View full text Add to dashboard Cite

show abstract

“…SPADE is designed to accept categorical segmentations. However, previous work on MRI synthesis [30] shows that partial volume maps, that associate probability of belonging to each class to each pixel, result in finer details on the output images. 3.…”

Section: Label Generatormentioning

confidence: 98%

Can segmentation models be trained with fully synthetically generated data?

Fernández¹,

Pinaya²,

Borges³

et al. 2022

Preprint

View full text Add to dashboard Cite

In order to achieve good performance and generalisability, medical image segmentation models should be trained on sizeable datasets with sufficient variability. Due to ethics and governance restrictions, and the costs associated with labelling data, scientific development is often stifled, with models trained and tested on limited data. Data augmentation is often used to artificially increase the variability in the data distribution and improve model generalisability. Recent works have explored deep generative models for image synthesis, as such an approach would enable the generation of an effectively infinite amount of varied data, addressing the generalisability and data access problems. However, many proposed solutions limit the user's control over what is generated. In this work, we propose brainSPADE, a model which combines a synthetic diffusion-based label generator with a semantic image generator. Our model can produce fully synthetic brain labels on-demand, with or without pathology of interest, and then generate a corresponding MRI image of an arbitrary guided style. Experiments show that brainSPADE synthetic data can be used to train segmentation models with performance comparable to that of models trained on real data.

show abstract

“…In Korkmaz et al (2022) instead use Transformer architectures for the task of MRI denoising. Rusak et al (2020) found that given paired Partial Volume maps and corresponding MRI scans, GANs can learn to synthesise brain imaging with accurate tissue borders from any given partial volume map. It is critical to observe that all these methods require paired imaging.…”

Section: Introductionmentioning

confidence: 99%