Uncertainty‐aware Visualization in Medical Imaging ‐ A Survey

Gillmann, Christina; Saur, Dorothee; Wischgoll, Thomas; Scheuermann, Gerik

doi:10.1111/cgf.14333

Cited by 18 publications

(7 citation statements)

References 139 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In addition, further visualization methods, such as activation maps or salient maps as in Refs. 39 and 61, could help to understand better the internal process of the model. The activation maps are particularly important for validating and assessing prediction in medical applications 72 …”

Section: Discussionmentioning

confidence: 99%

Stroke-GFCN: ischemic stroke lesion prediction with a fully convolutional graph network

Iporre-Rivas,

Saur,

Rohr

et al. 2023

J. Med. Imag.

Self Cite

View full text Add to dashboard Cite

The interpretation of image data plays a critical role during acute brain stroke diagnosis, and promptly defining the requirement of a surgical intervention will drastically impact the patient's outcome. However, determining stroke lesions purely from images can be a daunting task. Many studies proposed automatic segmentation methods for brain stroke lesions from medical images in different modalities, though heretofore results do not satisfy the requirements to be clinically reliable. We investigate the segmentation of brain stroke lesions using a geometric deep learning model that takes advantage of the intrinsic interconnected diffusion features in a set of multimodal inputs consisting of computer tomography (CT) perfusion parameters.Approach: We propose a geometric deep learning model for the segmentation of ischemic stroke brain lesions that employs spline convolutions and unpooling/ pooling operators on graphs to excerpt graph-structured features in a fully convolutional network architecture. In addition, we seek to understand the underlying principles governing the different components of our model. Accordingly, we structure the experiments in two parts: an evaluation of different architecture hyperparameters and a comparison with state-of-the-art methods.Results: The ablation study shows that deeper layers obtain a higher Dice coefficient score (DCS) of up to 0.3654. Comparing different pooling and unpooling methods shows that the best performing unpooling method is the proportional approach, yet it often smooths the segmentation border. Unpooling achieves segmentation results more adapted to the lesion boundary corroborated with systematic lower values of Hausdorff distance. The model performs at the level of state-of-the-art models without optimized training methods, such as augmentation or patches, with a DCS of 0.4553 AE 0.0031. Conclusions:We proposed and evaluated an end-to-end trainable fully convolutional graph network architecture using spline convolutional layers for the ischemic stroke lesion prediction. We propose a model that employs graph-based operations to predict acute stroke brain lesions from CT perfusion parameters. Our results prove the feasibility of using geometric deep learning to solve segmentation problems, and our model shows a better performance than other models evaluated. The proposed model achieves improved metric values for the DCS metric, ranging from 8.61% to 69.05%, compared with other models trained under the same conditions. Next, we compare different pooling and unpooling operations in relation to their segmentation results, and we show that the model can produce segmentation outputs that adapt to irregular segmentation boundaries when using simple heuristic unpooling operations.

show abstract

Section: Discussionmentioning

confidence: 99%

Stroke-GFCN: ischemic stroke lesion prediction with a fully convolutional graph network

Iporre-Rivas,

Saur,

Rohr

et al. 2023

J. Med. Imag.

Self Cite

View full text Add to dashboard Cite

show abstract

“…It is expected that the reconstruction processes are not perfect, which is especially true when done with insufficient input data in the first place. Uncertainty visualization is a central topic in visualization both for 2D as well as in 3D data as is reflected by multiple surveys [50], [51], [52], [53], [54]. There are only a few methods that address uncertainty in volume visualization, and it is not common for the current cryo-ET reconstruction pipelines to provide this uncertainty to the user, which is an option in our case.…”

Section: Related Workmentioning

confidence: 99%

GPU Accelerated 3D Tomographic Reconstruction and Visualization From Noisy Electron Microscopy Tilt-Series

Ramirez

Rautek

Bohak

et al. 2024

IEEE Trans. Visual. Comput. Graphics

View full text Add to dashboard Cite

We present a novel framework for 3D tomographic reconstruction and visualization of tomograms from noisy electron microscopy tilt-series. Our technique takes as an input aligned tilt-series from cryogenic electron microscopy and creates denoised 3D tomograms using a proximal jointly-optimized approach that iteratively performs reconstruction and denoising, relieving the users of the need to select appropriate denoising algorithms in the pre-reconstruction or post-reconstruction steps. The whole process is accelerated by exploiting parallelism on modern GPUs, and the results can be visualized immediately after the reconstruction using volume rendering tools incorporated in the framework. We show that our technique can be used with multiple combinations of reconstruction algorithms and regularizers, thanks to the flexibility provided by proximal algorithms. Additionally, the reconstruction framework is open-source and can be easily extended with additional reconstruction and denoising methods. Furthermore, our approach enables visualization of reconstruction error throughout the iterative process within the reconstructed tomogram and on projection planes of the input tilt-series. We evaluate our approach in comparison with state-of-the-art approaches and additionally show how our error visualization can be used for reconstruction evaluation.

show abstract

“…The visualization and reduction of various sources of uncertainty in the tractography pipeline has been a more recent focus of interest [BVPtH09, BPtHV13, SVBK14, WSSS14, SHV21, GvdVS21]. These sources can broadly be categorized into measurement uncertainty, model uncertainty, parameter uncertainty, and partial voluming [SVBK14, SV19, GSWS21].…”

Section: Related Workmentioning

confidence: 99%

Model Averaging and Bootstrap Consensus‐based Uncertainty Reduction in Diffusion MRI Tractography

Gruen

Voort

Schultz

2022

Computer Graphics Forum

View full text Add to dashboard Cite

Diffusion magnetic resonance imaging (dMRI) tractography has the unique ability to reconstruct major white matter tracts non‐invasively and is, therefore, widely used in neurosurgical planning and neuroscience. In this work, we reduce two sources of uncertainty within the tractography pipeline. The first one is the model uncertainty that arises in crossing fibre tractography, from having to estimate the number of relevant fibre compartments in each voxel. We propose a mathematical framework to estimate model uncertainty, and we reduce this type of uncertainty with a model averaging approach that combines the fibre direction estimates from all candidate models, weighted by the posterior probability of the respective model. The second source of uncertainty is measurement noise. We use bootstrapping to estimate this data uncertainty, and consolidate the fibre direction estimates from all bootstraps into a consensus model. We observe that, in most voxels, a traditional model selection strategy selects different models across bootstraps. In this sense, the bootstrap consensus also reduces model uncertainty. Either approach significantly increases the accuracy of crossing fibre tractography in multiple subjects, and combining them provides an additional benefit. However, model averaging is much more efficient computationally.

show abstract

Uncertainty‐aware Visualization in Medical Imaging ‐ A Survey

Cited by 18 publications

References 139 publications

Stroke-GFCN: ischemic stroke lesion prediction with a fully convolutional graph network

Stroke-GFCN: ischemic stroke lesion prediction with a fully convolutional graph network

GPU Accelerated 3D Tomographic Reconstruction and Visualization From Noisy Electron Microscopy Tilt-Series

Model Averaging and Bootstrap Consensus‐based Uncertainty Reduction in Diffusion MRI Tractography

Contact Info

Product

Resources

About