Quantifying Registration Uncertainty With Sparse Bayesian Modelling

Folgoc, Loïc Le; Delingette, Hervé; Criminisi, Antonio; Ayache, Nicholas

doi:10.1109/tmi.2016.2623608

Cited by 36 publications

(52 citation statements)

References 25 publications

(38 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although not reported here, preliminary experiments indicate that the proposed optimizer achieves comparable registration accuracies to the original demons algorithm [2] in this setting. The proposed sampler directly endows the demons algorithm with the first method to assess the uncertainty in its nonparametric deformation fields, the effective number of degrees of freedom of which is in the millions (compared to mere thousands in existing work for uncertainty estimation in registration [3][4][5][6][7][8][9][10]).…”

Section: Discussionmentioning

confidence: 99%

“…As in other work [3][4][5][6][7][8][9], the deformation regularization parameter γ can also be inferred automatically, rather than set by the user. When a non-informative gamma distribution Gam(γ|α 0 , β 0 ) with shape α 0 = 1 and rate β 0 = 0 is used as a conjugate prior for γ, this can be accomplished by simply including a fourth step in the sampler: γ (τ +1) ∼ p(γ|d (τ +1) ) = Gam( I 2 + 1, 1 2 Γd (τ +1) 2 ).…”

Section: Samplingmentioning

confidence: 99%

“…This results in fast optimizations of highly flexible, nonparametric deformation fields, taking only a few minutes on a standard desktop computer. Furthermore, the SSD criterion can be cast within a probabilistic modeling framework, making it possible to quantify registration uncertainty by approximating the relevant posterior probability distributions, using either variational [3][4][5][6] or sampling [7][8][9][10] methods.…”

Section: Introductionmentioning

confidence: 99%

“…And third, using largely the same code base as the proposed optimizer, we also derive a practical technique for Monte Carlo sampling from the registration posterior, allowing for direct visualization and quantification of uncertainty in MI-based models. In contrast to existing methods for uncertainty estimation in uni-modal registration [3][4][5][6][7][8][9][10], this sampler does not involve variational approximations that may significantly underestimate uncertainty [9]; does not require tuning of various Metropolis-Hastings proposal distribution parameters; and can readily handle full 3D nonparametric deformation models with orders-of-magnitude more degrees of freedom than the sparse models used so far.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Fast Nonparametric Mutual-Information-based Registration and Uncertainty Estimation

Agn

Leemput

2019

Uncertainty for Safe Utilization of Machine Learning in Medical Imaging and Clinical Image-Based Procedures

View full text Add to dashboard Cite

 Users may download and print one copy of any publication from the public portal for the purpose of private study or research.  You may not further distribute the material or use it for any profit-making activity or commercial gain  You may freely distribute the URL identifying the publication in the public portal If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Samplingmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Fast Nonparametric Mutual-Information-based Registration and Uncertainty Estimation

Agn

Leemput

2019

Uncertainty for Safe Utilization of Machine Learning in Medical Imaging and Clinical Image-Based Procedures

View full text Add to dashboard Cite

show abstract

“…The spread of the posterior of unknown registration parameters is then considered as a measure of the registration uncertainty. Existing methods including stochastic [13] and sampling [4,5,10] methods have been investigated to estimate the uncertainty, due to the fact that the posterior does not have a closed form and is computationally problematic to solve. A large computational effort is required to sample over high dimensional parameter spaces.…”

Section: Introductionmentioning

confidence: 99%

Efficient Laplace Approximation for Bayesian Registration Uncertainty Quantification

Wang

Wells

Golland³

et al. 2018

Lecture Notes in Computer Science

View full text Add to dashboard Cite

This paper presents a novel approach to modeling the pos terior distribution in image registration that is computationally efficient for large deformation diffeomorphic metric mapping (LDDMM). We develop a Laplace approximation of Bayesian registration models entirely in a bandlimited space that fully describes the properties of diffeomorphic transformations. In contrast to current methods, we compute the inverse Hessian at the mode of the posterior distribution of diffeomorphisms directly in the low dimensional frequency domain. This dramatically reduces the computational complexity of approximating posterior marginals in the high dimensional imaging space. Experimental results show that our method is significantly faster than the state-of-the-art diffeomorphic image registration uncertainty quantification algorithms, while producing comparable results. The efficiency of our method strengthens the feasibility in prospective clinical applications, e.g., real- time image-guided navigation for brain surgery.

show abstract

Deformation driven Seq2Seq longitudinal tumor and organs‐at‐risk prediction for radiotherapy

et al. 2021

View full text Add to dashboard Cite

Purpose Radiotherapy presents unique challenges and clinical requirements for longitudinal tumor and organ‐at‐risk (OAR) prediction during treatment. The challenges include tumor inflammation/edema and radiation‐induced changes in organ geometry, whereas the clinical requirements demand flexibility in input/output sequence timepoints to update the predictions on rolling basis and the grounding of all predictions in relationship to the pre‐treatment imaging information for response and toxicity assessment in adaptive radiotherapy. Methods To deal with the aforementioned challenges and to comply with the clinical requirements, we present a novel 3D sequence‐to‐sequence model based on Convolution Long Short‐Term Memory (ConvLSTM) that makes use of series of deformation vector fields (DVFs) between individual timepoints and reference pre‐treatment/planning CTs to predict future anatomical deformations and changes in gross tumor volume as well as critical OARs. High‐quality DVF training data are created by employing hyper‐parameter optimization on the subset of the training data with DICE coefficient and mutual information metric. We validated our model on two radiotherapy datasets: a publicly available head‐and‐neck dataset (28 patients with manually contoured pre‐, mid‐, and post‐treatment CTs), and an internal non‐small cell lung cancer dataset (63 patients with manually contoured planning CT and 6 weekly CBCTs). Results The use of DVF representation and skip connections overcomes the blurring issue of ConvLSTM prediction with the traditional image representation. The mean and standard deviation of DICE for predictions of lung GTV at weeks 4, 5, and 6 were 0.83 ± 0.09, 0.82 ± 0.08, and 0.81 ± 0.10, respectively, and for post‐treatment ipsilateral and contralateral parotids, were 0.81 ± 0.06 and 0.85 ± 0.02. Conclusion We presented a novel DVF‐based Seq2Seq model for medical images, leveraging the complete 3D imaging information of a relatively large longitudinal clinical dataset, to carry out longitudinal GTV/OAR predictions for anatomical changes in HN and lung radiotherapy patients, which has potential to improve RT outcomes.

show abstract

Quantifying Registration Uncertainty With Sparse Bayesian Modelling

Cited by 36 publications

References 25 publications

Fast Nonparametric Mutual-Information-based Registration and Uncertainty Estimation

Fast Nonparametric Mutual-Information-based Registration and Uncertainty Estimation

Efficient Laplace Approximation for Bayesian Registration Uncertainty Quantification

Deformation driven Seq2Seq longitudinal tumor and organs‐at‐risk prediction for radiotherapy

Contact Info

Product

Resources

About