Multimodal 3D medical image registration guided by shape encoder–decoder networks

Blendowski, Max; Bouteldja, Nassim; Heinrich, Mattias P.

doi:10.1007/s11548-019-02089-8

Cited by 32 publications

(20 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These approaches, generally called feature-based registration, are of particular interest when the relation between content is unknown or cannot be taken as an assumption (for example for the validation of a new probe or a new imaging modality). The method to find the matching and compute the transformation can be done with two main paradigms: transforming the image data in localizations with potential additional features using point-based registration ( [89,91] for the AutoFINDER part of ec-clem) or shape-based registration, potentially with intensity-based machine learning approaches [99]. Note that a plethora of variants exists for point-cloud registration, some of them sounding particularly promising for feature-based multimodal registration [100].…”

Section: Correlation Softwarementioning

confidence: 99%

Correlated Multimodal Imaging in Life Sciences: Expanding the Biomedical Horizon

et al. 2020

View full text Add to dashboard Cite

show abstract

Section: Correlation Softwarementioning

confidence: 99%

Correlated Multimodal Imaging in Life Sciences: Expanding the Biomedical Horizon

et al. 2020

View full text Add to dashboard Cite

show abstract

“…The experiment was trained on 60 CT volumes with a dice value of 0.914. Max et al [26] conducted multi-modality (i.e., CT and MRI) 3D whole-heart segmentation using supervised deep learning by defining a shape encoder-decoder network. The experiment was trained on 15 CT volumes with a dice value of 0.653.…”

Section: Related Workmentioning

confidence: 99%

“…Later, the deep learning approach showed promising successful performance [7,9,10]. The deep learning approach has two learning manners, including supervised learning [22][23][24][25][26], which requires a ground truth to align the loss function, and unsupervised learning [27][28][29], which learns the features without ground truth labeling.…”

Section: Introductionmentioning

confidence: 99%

“…Labeling the ground truth for cardiac segmentation requires the extensive manual annotation of images by the radiologist, which can be time-consuming and labor intensive [7][8][9][10][22][23][24][25][26]. Due to the difficulty in obtaining annotated data and the required expertise as an annotator, an unsupervised approach has been considered in this study.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Semantic Cardiac Segmentation in Chest CT Images Using K-Means Clustering and the Mathematical Morphology Method

Lee

Gil

et al. 2021

Sensors

View full text Add to dashboard Cite

Whole cardiac segmentation in chest CT images is important to identify functional abnormalities that occur in cardiovascular diseases, such as coronary artery disease (CAD) detection. However, manual efforts are time-consuming and labor intensive. Additionally, labeling the ground truth for cardiac segmentation requires the extensive manual annotation of images by the radiologist. Due to the difficulty in obtaining the annotated data and the required expertise as an annotator, an unsupervised approach is proposed. In this paper, we introduce a semantic whole-heart segmentation combining K-Means clustering as a threshold criterion of the mean-thresholding method and mathematical morphology method as a threshold shifting enhancer. The experiment was conducted on 500 subjects in two cases: (1) 56 slices per volume containing full heart scans, and (2) 30 slices per volume containing about half of the top of heart scans before the liver appears. In both cases, the results showed an average silhouette score of the K-Means method of 0.4130. Additionally, the experiment on 56 slices per volume achieved an overall accuracy (OA) and mean intersection over union (mIoU) of 34.90% and 41.26%, respectively,; while the performance for the first 30 slices per volume achieved an OA and mIoU of 55.10% and 71.46%, respectively.

show abstract

“…Since years, multiple independent research groups try to improve current IF visualization [13] and introduce more advanced guidance techniques, such as e.g., automatic crossmodal image registration [14][15][16][17][18], image-based tracking [14,19], automatic compensation of heartbeat and respiratory motion [20][21][22] to overcome these specific issues. To introduce these techniques into a product functionality is, however, not easy due to a proprietary character of commercial XR systems.…”

Section: Introductionmentioning

confidence: 99%

3D-XGuide: open-source X-ray navigation guidance system

et al. 2020

View full text Add to dashboard Cite

Purpose With the growing availability and variety of imaging modalities, new methods of intraoperative support have become available for all kinds of interventions. The basic principles of image fusion and image guidance have been widely adopted and are commercialized through a number of platforms. Although multimodal systems have been found to be useful for guiding interventional procedures, they all have their limitations. The integration of more advanced guidance techniques into the product functionality is, however, not easy due to the proprietary solutions of the vendors. Therefore, the purpose of this work is to introduce a software system for image fusion, real-time navigation, and working points documentation during transcatheter interventions performed under X-ray (XR) guidance. Methods An interactive software system for cross-modal registration and image fusion of XR fluoroscopy with CT or MRI-derived anatomic 3D models is implemented using Qt application framework and VTK visualization pipeline. DICOM data can be imported in retrospective mode. Live XR data input is realized by a video capture card application interface. Results The actual software release offers a graphical user interface with basic functionality including data import and handling, calculation of projection geometry and transformations between related coordinate systems, rigid 3D-3D registration, and template matching-based tracking and motion compensation algorithms in 2D and 3D. The link to the actual software release on GitHub including source code and executable is provided to support independent research and development in the field of intervention guidance. Conclusion The introduced system provides a common foundation for the rapid prototyping of new approaches in the field of XR fluoroscopic guidance. As a pure software solution, the developed system is potentially vendor-independent and can be easily extended to be used with the XR systems of different manufacturers.

show abstract

Multimodal 3D medical image registration guided by shape encoder–decoder networks

Cited by 32 publications

References 15 publications

Correlated Multimodal Imaging in Life Sciences: Expanding the Biomedical Horizon

Correlated Multimodal Imaging in Life Sciences: Expanding the Biomedical Horizon

Semantic Cardiac Segmentation in Chest CT Images Using K-Means Clustering and the Mathematical Morphology Method

3D-XGuide: open-source X-ray navigation guidance system

Contact Info

Product

Resources

About