Moucheng Xu
@mouchengxu.bsky.social
Building AI surgical robots at Medtronic | Medical Image Computing | Deep Learning | PhD from UCL, UK | https://moucheng2017.github.io/
Preprint: arxiv.org/abs/2502.05713
4D VQ-GAN: Synthesising Medical Scans at Any Time Point for Personalised Disease Progression Modelling of Idiopathic Pulmonary Fibrosis
Understanding the progression trajectories of diseases is crucial for early diagnosis and effective treatment planning. This is especially vital for life-threatening conditions such as Idiopathic Pulm...
arxiv.org
February 23, 2025 at 4:53 PM
Preprint: arxiv.org/abs/2502.05713
3) Works well with sparsely sampled scans at irregular time points and needs no prior assumptions about disease progression patterns;
4) Combines a 3D VQ-GAN and a novel temporal model based on neural ODE Solver.
4) Combines a 3D VQ-GAN and a novel temporal model based on neural ODE Solver.
February 23, 2025 at 4:53 PM
3) Works well with sparsely sampled scans at irregular time points and needs no prior assumptions about disease progression patterns;
4) Combines a 3D VQ-GAN and a novel temporal model based on neural ODE Solver.
4) Combines a 3D VQ-GAN and a novel temporal model based on neural ODE Solver.
1) Paves a new way for personalized disease progression modeling by providing virtual progression trajectories;
2) Has the potential for real clinical use; we verified that our generated scans have a comparable C-index of survival analysis to that of real scans;
2) Has the potential for real clinical use; we verified that our generated scans have a comparable C-index of survival analysis to that of real scans;
February 23, 2025 at 4:53 PM
1) Paves a new way for personalized disease progression modeling by providing virtual progression trajectories;
2) Has the potential for real clinical use; we verified that our generated scans have a comparable C-index of survival analysis to that of real scans;
2) Has the potential for real clinical use; we verified that our generated scans have a comparable C-index of survival analysis to that of real scans;
2) An Event-Based Disease Progression Model and Its Application to Familial Alzheimer's Disease (IPMI 2011)
3) Auto-Encoding Variational Bayes (ICLR 2014)
4) Spatial Transformer Networks (NeurIPS 2015)
5) Self-Labelling via Simulatenous Clustering and Representation Learning (ICLR 2020)
3) Auto-Encoding Variational Bayes (ICLR 2014)
4) Spatial Transformer Networks (NeurIPS 2015)
5) Self-Labelling via Simulatenous Clustering and Representation Learning (ICLR 2020)
February 9, 2025 at 7:40 PM
2) An Event-Based Disease Progression Model and Its Application to Familial Alzheimer's Disease (IPMI 2011)
3) Auto-Encoding Variational Bayes (ICLR 2014)
4) Spatial Transformer Networks (NeurIPS 2015)
5) Self-Labelling via Simulatenous Clustering and Representation Learning (ICLR 2020)
3) Auto-Encoding Variational Bayes (ICLR 2014)
4) Spatial Transformer Networks (NeurIPS 2015)
5) Self-Labelling via Simulatenous Clustering and Representation Learning (ICLR 2020)
I finished my PhD in medical image computing in 2023. At the time, my favorite papers, in order of their publication dates, were:
1) Unsupervised classifiers, mutual information and ‘phantom targets’ (NeurIPS 1991)
1) Unsupervised classifiers, mutual information and ‘phantom targets’ (NeurIPS 1991)
February 9, 2025 at 7:40 PM
I finished my PhD in medical image computing in 2023. At the time, my favorite papers, in order of their publication dates, were:
1) Unsupervised classifiers, mutual information and ‘phantom targets’ (NeurIPS 1991)
1) Unsupervised classifiers, mutual information and ‘phantom targets’ (NeurIPS 1991)