Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Leveraging a Statistical Shape Model for Efficient Generation of Annotated Training Data: A Case Study on Liver Landmarks Segmentation

Denis Krnjaca, Lorena Krames, Werner Nahm

Abstract

Anatomical landmark segmentation serves as a critical initial step for robust multimodal registration during computer-assisted interventions. Current approaches predominantly rely on deep learning, which often necessitates the extensive manual generation of annotated datasets. In this paper, we present a novel strategy for creating large annotated datasets using a statistical shape model (SSM) based on a mean shape that is manually labeled only once. We demonstrate the method's efficacy through its application to deep-learning-based anatomical landmark segmentation, specifically targeting the detection of the anterior ridge and the falciform ligament in 3D liver shapes. A specialized deep learning network was trained with 8,800 annotated liver shapes generated by the SSM. The network's performance was evaluated on 500 unseen synthetic SSM shapes, yielding a mean Intersection over Union of 91.4% (87.4% for the anterior ridge and 87.6% for the falciform ligament). Subsequently, the network was applied to clinical patient liver shapes, with qualitative evaluation indicating promising results and highlighting the generalizability of the proposed approach. Our findings suggest that the SSM-based data generation approach alleviates the labor-intensive process of manual labeling while enabling the creation of large annotated training datasets for machine learning. Although our study focuses on liver anatomy, the proposed methodology holds potential for a broad range of applications where annotated training datasets play a pivotal role in developing accurate deep-learning models.

Leveraging a Statistical Shape Model for Efficient Generation of Annotated Training Data: A Case Study on Liver Landmarks Segmentation

Abstract

Anatomical landmark segmentation serves as a critical initial step for robust multimodal registration during computer-assisted interventions. Current approaches predominantly rely on deep learning, which often necessitates the extensive manual generation of annotated datasets. In this paper, we present a novel strategy for creating large annotated datasets using a statistical shape model (SSM) based on a mean shape that is manually labeled only once. We demonstrate the method's efficacy through its application to deep-learning-based anatomical landmark segmentation, specifically targeting the detection of the anterior ridge and the falciform ligament in 3D liver shapes. A specialized deep learning network was trained with 8,800 annotated liver shapes generated by the SSM. The network's performance was evaluated on 500 unseen synthetic SSM shapes, yielding a mean Intersection over Union of 91.4% (87.4% for the anterior ridge and 87.6% for the falciform ligament). Subsequently, the network was applied to clinical patient liver shapes, with qualitative evaluation indicating promising results and highlighting the generalizability of the proposed approach. Our findings suggest that the SSM-based data generation approach alleviates the labor-intensive process of manual labeling while enabling the creation of large annotated training datasets for machine learning. Although our study focuses on liver anatomy, the proposed methodology holds potential for a broad range of applications where annotated training datasets play a pivotal role in developing accurate deep-learning models.
Paper Structure (12 sections, 4 equations, 4 figures)

This paper contains 12 sections, 4 equations, 4 figures.

Table of Contents

  1. Introduction
  2. Methodology
  3. Case Study
  4. Experiments
  5. Implementation Details
  6. Datasets
  7. Results
  8. Evaluation of Label Transfer on the Training Data Set:
  9. Quantitative and Qualitative Evaluation on SSM Shapes:
  10. Qualitative Evaluation on Shapes from MedShapeNet:
  11. Discussion and Conclusion

Figures (4)

  • Figure 1: Overview of the proposed method employing an SSM for data generation. The mean shape derived from the SSM is manually labeled once (in this example: red: anterior ridge, blue: falciform ligament). Subsequently, labels are transferred to generated shapes using vertex indices. The annotated dataset is then utilized for training a deep-learning network.
  • Figure 2: Two examples illustrating the aggregated labels annotated by the four individuals (first and third shape) and the corresponding results from the label transfer from the mean shape (second and last shape).
  • Figure 3: Evaluation of three examples from the synthetic SSM test dataset. The anterior ridge prediction is shown in blue and the falciform ligament in red. The last two rows show the mIoU over all labels as well as the IoU for the anterior ridge and the falciform ligament.
  • Figure 4: Qualitative evaluation on six examples from the MedShapeNet dataset medshapenet. The anterior ridge prediction is shown in blue and the falciform ligament in red.