Using simulated breast lesions based on Perlin noise for evaluation of lesion segmentation
(2024) Medical Imaging 2024: Physics of Medical Imaging In Progress in Biomedical Optics and Imaging - Proceedings of SPIE 12925.- Abstract
Segmentation of diagnostic radiography images using deep learning is progressively expanding, which sets demands on the accessibility, availability, and accuracy of the software tools used. This study aimed at evaluating the performance of a segmentation model for digital breast tomosynthesis (DBT), with the use of computer-simulated breast anatomy. We have simulated breast anatomy and soft tissue breast lesions, by utilizing a model approach based on the Perlin noise algorithm. The obtained breast phantoms were projected and reconstructed into DBT slices using a publicly available open-source reconstruction method. Each lesion was then segmented using two approaches: 1. the Segment Anything Model (SAM), a publicly available AI-based... (More)
Segmentation of diagnostic radiography images using deep learning is progressively expanding, which sets demands on the accessibility, availability, and accuracy of the software tools used. This study aimed at evaluating the performance of a segmentation model for digital breast tomosynthesis (DBT), with the use of computer-simulated breast anatomy. We have simulated breast anatomy and soft tissue breast lesions, by utilizing a model approach based on the Perlin noise algorithm. The obtained breast phantoms were projected and reconstructed into DBT slices using a publicly available open-source reconstruction method. Each lesion was then segmented using two approaches: 1. the Segment Anything Model (SAM), a publicly available AI-based method for image segmentation and 2. manually by three human observers. The lesion area in each slice was compared to the ground truth area, derived from the binary mask of the lesion model. We found similar performance between SAM and manual segmentation. Both SAM and the observers performed comparably in the central slice (mean absolute relative error compared to the ground truth and standard deviation SAM: 4 ± 3 %, observers: 3 ± 3 %). Similarly, both SAM and the observers overestimated the lesion area in the peripheral reconstructed slices (mean absolute relative error and standard deviation SAM: 277 ± 190 %, observers: 295 ± 182 %). We showed that 3D voxel phantoms can be used for evaluating different segmentation methods. In preliminary comparison, tumor segmentation in simulated DBT images using SAM open-source method showed a similar performance as manual tumor segmentation.
(Less)
- author
- Tomic, Hanna LU ; Yang, Zhikai ; Tingberg, Anders LU ; Zackrisson, Sophia LU ; Moreno, Rodrigo ; Smedby, Örjan ; Dustler, Magnus LU and Bakic, Predrag LU
- organization
- publishing date
- 2024
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- subject
- keywords
- AI, Breast phantom, computer simulations and VCT, segmentation
- host publication
- Medical Imaging 2024 : Physics of Medical Imaging - Physics of Medical Imaging
- series title
- Progress in Biomedical Optics and Imaging - Proceedings of SPIE
- editor
- Fahrig, Rebecca ; Sabol, John M. and Li, Ke
- volume
- 12925
- article number
- 129251P
- publisher
- SPIE
- conference name
- Medical Imaging 2024: Physics of Medical Imaging
- conference location
- San Diego, United States
- conference dates
- 2024-02-19 - 2024-02-22
- external identifiers
-
- scopus:85193540163
- ISSN
- 1605-7422
- ISBN
- 9781510671546
- DOI
- 10.1117/12.3008840
- language
- English
- LU publication?
- yes
- id
- ccd15401-d633-49a7-b613-61735c5c6fab
- date added to LUP
- 2024-06-13 15:03:18
- date last changed
- 2024-06-13 15:03:54
@inproceedings{ccd15401-d633-49a7-b613-61735c5c6fab, abstract = {{<p>Segmentation of diagnostic radiography images using deep learning is progressively expanding, which sets demands on the accessibility, availability, and accuracy of the software tools used. This study aimed at evaluating the performance of a segmentation model for digital breast tomosynthesis (DBT), with the use of computer-simulated breast anatomy. We have simulated breast anatomy and soft tissue breast lesions, by utilizing a model approach based on the Perlin noise algorithm. The obtained breast phantoms were projected and reconstructed into DBT slices using a publicly available open-source reconstruction method. Each lesion was then segmented using two approaches: 1. the Segment Anything Model (SAM), a publicly available AI-based method for image segmentation and 2. manually by three human observers. The lesion area in each slice was compared to the ground truth area, derived from the binary mask of the lesion model. We found similar performance between SAM and manual segmentation. Both SAM and the observers performed comparably in the central slice (mean absolute relative error compared to the ground truth and standard deviation SAM: 4 ± 3 %, observers: 3 ± 3 %). Similarly, both SAM and the observers overestimated the lesion area in the peripheral reconstructed slices (mean absolute relative error and standard deviation SAM: 277 ± 190 %, observers: 295 ± 182 %). We showed that 3D voxel phantoms can be used for evaluating different segmentation methods. In preliminary comparison, tumor segmentation in simulated DBT images using SAM open-source method showed a similar performance as manual tumor segmentation.</p>}}, author = {{Tomic, Hanna and Yang, Zhikai and Tingberg, Anders and Zackrisson, Sophia and Moreno, Rodrigo and Smedby, Örjan and Dustler, Magnus and Bakic, Predrag}}, booktitle = {{Medical Imaging 2024 : Physics of Medical Imaging}}, editor = {{Fahrig, Rebecca and Sabol, John M. and Li, Ke}}, isbn = {{9781510671546}}, issn = {{1605-7422}}, keywords = {{AI; Breast phantom; computer simulations and VCT; segmentation}}, language = {{eng}}, publisher = {{SPIE}}, series = {{Progress in Biomedical Optics and Imaging - Proceedings of SPIE}}, title = {{Using simulated breast lesions based on Perlin noise for evaluation of lesion segmentation}}, url = {{http://dx.doi.org/10.1117/12.3008840}}, doi = {{10.1117/12.3008840}}, volume = {{12925}}, year = {{2024}}, }