• Corpus ID: 248069672

Image prediction of disease progression by style-based manifold extrapolation

  title={Image prediction of disease progression by style-based manifold extrapolation},
  author={Tianyu Han and Jakob Nikolas Kather and Federico Pedersoli and Markus Zimmermann and Sebastian Keil and Maximilian Franz Schulze-Hagen and Marc Terwoelbeck and Peter Isfort and Christoph Haarburger and Fabian Kiessling and Volkmar Schulz and Christiane Kuhl and Sven Nebelung and Daniel Truhn},
Disease-modifying management aims to prevent deterioration and progression of the disease, not just relieve symptoms. Unfortunately, the development of necessary therapies is often hampered by the failure to recognize the presymptomatic disease and limited understanding of disease development. We present a generic solution for this problem by a methodology that allows the prediction of progression risk and morphology in individuals using a latent extrapolation optimization approach. To this end… 

Figures and Tables from this paper

Medical Diffusion - Denoising Diffusion Probabilistic Models for 3D Medical Image Generation

It is shown that diffusion probabilistic models can synthesize high quality medical imaging data and can be used in a self-supervised pre-training and improve the performance of breast segmentation models when data is scarce.



Highly accurate protein structure prediction with AlphaFold

This work validated an entirely redesigned version of the neural network-based model, AlphaFold, in the challenging 14th Critical Assessment of protein Structure Prediction (CASP14)15, demonstrating accuracy competitive with experiment in a majority of cases and greatly outperforming other methods.

Continuous-Time Deep Glioma Growth Models

Neural Processes is extended, a class of conditional generative models for stochastic time series with a hierarchical multi-scale representation encoding including a spatio-temporal attention mechanism, resulting in a learned growth model that can be conditioned on an arbitrary number of observations, and that can produce a distribution of temporally consistent growth trajectories on a continuous time axis.

An algorithmic approach to reducing unexplained pain disparities in underserved populations

An algorithmic, machine-learning approach to measuring severe pain from osteoarthritis applied to X-ray images of knees suggests that reported disparities in knee pain in underserved populations can be reduced by comparison with use of standard radiographic measures of disease severity.

Breaking medical data sharing boundaries by using synthesized radiographs

This work proposes to use generative models (GMs) to produce high-resolution synthetic radiographs that do not contain any personal identification information, and integrates federated learning strategies to improve the performance of CV algorithms trained on smaller datasets.

Advancing diagnostic performance and clinical usability of neural networks via adversarial training and dual batch normalization

It is demonstrated that adversarially trained models can significantly enhance the usability of pathology detection as compared to their standard counterparts and it is elucidate that different paths for adversarial and real images are needed during training to achieve state of the art results with superior clinical interpretability.

Statistical Disease Progression Modeling in Alzheimer Disease

This article presents a class of statistical disease progression models and applies them to longitudinal cognitive scores and gives new insights to the value of biomarkers for staging patients and suggest alternative explanations for previous findings related to accelerated cognitive decline among highly educated patients and patients on symptomatic treatments.

Predicting conversion to wet age-related macular degeneration using deep learning

In individuals diagnosed with age-related macular degeneration in one eye, a deep learning model can predict progression to the ‘wet’, sight-threatening form of the disease in the second eye within a 6-month time frame, and demonstrates the potential of using AI to predict disease progression.

Improved protein structure prediction using potentials from deep learning

It is shown that a neural network can be trained to make accurate predictions of the distances between pairs of residues, which convey more information about the structure than contact predictions, and the resulting potential can be optimized by a simple gradient descent algorithm to generate structures without complex sampling procedures.

International evaluation of an AI system for breast cancer screening

A robust assessment of the AI system paves the way for clinical trials to improve the accuracy and efficiency of breast cancer screening and using a combination of AI and human inputs could help to improve screening efficiency.

Analyzing and Improving the Image Quality of StyleGAN

This work redesigns the generator normalization, revisit progressive growing, and regularize the generator to encourage good conditioning in the mapping from latent codes to images, and thereby redefines the state of the art in unconditional image modeling.