RECORD
Reducing the Human Effort in Developing PET-CT Registration
Computer Vision and Pattern Recognition · Submitted 25 Nov 2019
abstract compiled · 2026-09-23 · no language-model review
The authors' main claim We demonstrate the potential of our method via qualitative and quantitative experiments.
The compiled record
The paper's own sentences, arranged. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.
Open the compiled record
The abstract, sentence by sentence
We aim to reduce the tedious nature of developing and evaluating methods for aligning PET-CT scans from multiple patient visits. Current methods for registration rely on correspondences that are created manually by medical experts with 3D manipulation, or assisted alignments done by utilizing mutual information across CT scans that may not be consistent when transferred to the PET images. Instead, we propose to label multiple key points across several 2D slices, which we then fit a key curve to. This removes the need for creating manual alignments in 3D and makes the labelling process easier. We use these key curves to define an error metric for the alignments that can be computed efficiently. While our metric is non-differentiable, we further show that we can utilize it during the training of our deep model via a novel method. Specifically, instead of relying on detailed geometric labels -- e.g., manual 3D alignments -- we use synthetically generated deformations of real data. To incorporate robustness to changes that occur between visits other than geometric changes, we enforce consistency across visits in the deep network's internal representations. We demonstrate the potential of our method via qualitative and quantitative experiments.
background contribution method result interpretation main claim
| # | sentence | role | the authors' own work | states the main conclusion | strength |
|---|---|---|---|---|---|
| s0 | We aim to reduce the tedious nature of developing and evaluating methods for aligning PET-CT scans from multiple patient visits. | background 0.73 | 0.19 | 0.06 | asserted 0.57 |
| s1 | Current methods for registration rely on correspondences that are created manually by medical experts with 3D manipulation, or assisted alignments done by utilizing mutual information across CT scans that may not be consistent when transferred to the PET images. | background 0.93 | 0.05 | 0.03 | asserted 0.62 |
| s2 | Instead, we propose to label multiple key points across several 2D slices, which we then fit a key curve to. | method 0.88 | 0.82 | 0.08 | asserted 0.97 |
| s3 | This removes the need for creating manual alignments in 3D and makes the labelling process easier. | contribution 0.44 | 0.59 | 0.12 | asserted 1.00 |
| s4 | We use these key curves to define an error metric for the alignments that can be computed efficiently. | method 0.90 | 0.94 | 0.07 | asserted 1.00 |
| s5 | While our metric is non-differentiable, we further show that we can utilize it during the training of our deep model via a novel method. | method 0.71 | 0.94 | 0.08 | asserted 0.99 |
| s6 | Specifically, instead of relying on detailed geometric labels -- e.g., manual 3D alignments -- we use synthetically generated deformations of real data. | method 0.96 | 0.89 | 0.07 | asserted 0.98 |
| s7 | To incorporate robustness to changes that occur between visits other than geometric changes, we enforce consistency across visits in the deep network's internal representations. | method 0.93 | 0.91 | 0.06 | asserted 0.98 |
| s8 | We demonstrate the potential of our method via qualitative and quantitative experiments. | result 0.65 | 0.95 | 0.26 | asserted 0.65 |
What was examined, and what was not
- arXiv listing abstract v1
- full text
- figures and tables
- references and citing sentences
- relations to other papers
- P abstract
- S sections
- T body sentences
- C citations
- R cross-paper relations
The paper
Original abstract
We aim to reduce the tedious nature of developing and evaluating methods for aligning PET-CT scans from multiple patient visits. Current methods for registration rely on correspondences that are created manually by medical experts with 3D manipulation, or assisted alignments done by utilizing mutual information across CT scans that may not be consistent when transferred to the PET images. Instead, we propose to label multiple key points across several 2D slices, which we then fit a key curve to. This removes the need for creating manual alignments in 3D and makes the labelling process easier. We use these key curves to define an error metric for the alignments that can be computed efficiently. While our metric is non-differentiable, we further show that we can utilize it during the training of our deep model via a novel method. Specifically, instead of relying on detailed geometric labels -- e.g., manual 3D alignments -- we use synthetically generated deformations of real data. To incorporate robustness to changes that occur between visits other than geometric changes, we enforce consistency across visits in the deep network's internal representations. We demonstrate the potential of our method via qualitative and quantitative experiments.
Human review
Discussion
Know something about this paper? Leave a comment, or highlight any passage above to comment on it.