Localization of the transverse processes in ultrasound for spinal curvature measurement

Shahrokh Kamali, Tamas Ungi, Andras Lasso, Christina Yan, Matthew Lougheed, Gabor Fichtinger

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

PURPOSE: In scoliosis monitoring, tracked ultrasound has been explored as a safer imaging alternative to traditional radiography. The use of ultrasound in spinal curvature measurement requires identification of vertebral landmarks such as transverse processes, but as bones have reduced visibility in ultrasound imaging, skeletal landmarks are typically segmented manually, which is an exceedingly laborious and long process. We propose an automatic algorithm to segment and localize the surface of bony areas in the transverse process for scoliosis in ultrasound. METHODS: The algorithm uses cascade of filters to remove low intensity pixels, smooth the image and detect bony edges. By applying first differentiation, candidate bony areas are classified. The average intensity under each area has a correlation with the possibility of a shadow, and areas with strong shadow are kept for bone segmentation. The segmented images are used to reconstruct a 3-D volume to represent the whole spinal structure around the transverse processes. RESULTS: A comparison between the manual ground truth segmentation and the automatic algorithm in 50 images showed 0.17 mm average difference. The time to process all 1, 938 images was about 37 Sec. (0.0191 Sec./Image), including reading the original sequence file. CONCLUSION: Initial experiments showed the algorithm to be sufficiently accurate and fast for segmentation transverse processes in ultrasound for spinal curvature measurement. An extensive evaluation of the method is currently underway on images from a larger patient cohort and using multiple observers in producing ground truth segmentation.

Original languageEnglish (US)
Title of host publicationMedical Imaging 2017
Subtitle of host publicationImage-Guided Procedures, Robotic Interventions, and Modeling
PublisherSPIE
Volume10135
ISBN (Electronic)9781510607156
DOIs
StatePublished - 2017
Externally publishedYes
EventMedical Imaging 2017: Image-Guided Procedures, Robotic Interventions, and Modeling - Orlando, United States
Duration: Feb 14 2017Feb 16 2017

Other

OtherMedical Imaging 2017: Image-Guided Procedures, Robotic Interventions, and Modeling
CountryUnited States
CityOrlando
Period2/14/172/16/17

Fingerprint

Spinal Curvatures
Ultrasonics
curvature
Scoliosis
landmarks
ground truth
Bone
Bone and Bones
bones
Imaging techniques
Radiography
Reading
Ultrasonography
Visibility
radiography
files
visibility
Pixels
cascades
pixels

Keywords

  • Scoliosis
  • Transverse process localization
  • Ultrasound automatic bone segmentation

ASJC Scopus subject areas

  • Atomic and Molecular Physics, and Optics
  • Electronic, Optical and Magnetic Materials
  • Biomaterials
  • Radiology Nuclear Medicine and imaging

Cite this

Kamali, S., Ungi, T., Lasso, A., Yan, C., Lougheed, M., & Fichtinger, G. (2017). Localization of the transverse processes in ultrasound for spinal curvature measurement. In Medical Imaging 2017: Image-Guided Procedures, Robotic Interventions, and Modeling (Vol. 10135). [101350I] SPIE. https://doi.org/10.1117/12.2256007

Localization of the transverse processes in ultrasound for spinal curvature measurement. / Kamali, Shahrokh; Ungi, Tamas; Lasso, Andras; Yan, Christina; Lougheed, Matthew; Fichtinger, Gabor.

Medical Imaging 2017: Image-Guided Procedures, Robotic Interventions, and Modeling. Vol. 10135 SPIE, 2017. 101350I.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Kamali, S, Ungi, T, Lasso, A, Yan, C, Lougheed, M & Fichtinger, G 2017, Localization of the transverse processes in ultrasound for spinal curvature measurement. in Medical Imaging 2017: Image-Guided Procedures, Robotic Interventions, and Modeling. vol. 10135, 101350I, SPIE, Medical Imaging 2017: Image-Guided Procedures, Robotic Interventions, and Modeling, Orlando, United States, 2/14/17. https://doi.org/10.1117/12.2256007
Kamali S, Ungi T, Lasso A, Yan C, Lougheed M, Fichtinger G. Localization of the transverse processes in ultrasound for spinal curvature measurement. In Medical Imaging 2017: Image-Guided Procedures, Robotic Interventions, and Modeling. Vol. 10135. SPIE. 2017. 101350I https://doi.org/10.1117/12.2256007
Kamali, Shahrokh ; Ungi, Tamas ; Lasso, Andras ; Yan, Christina ; Lougheed, Matthew ; Fichtinger, Gabor. / Localization of the transverse processes in ultrasound for spinal curvature measurement. Medical Imaging 2017: Image-Guided Procedures, Robotic Interventions, and Modeling. Vol. 10135 SPIE, 2017.
@inproceedings{472f70acb1a14389af00e9b34b98c7f1,
title = "Localization of the transverse processes in ultrasound for spinal curvature measurement",
abstract = "PURPOSE: In scoliosis monitoring, tracked ultrasound has been explored as a safer imaging alternative to traditional radiography. The use of ultrasound in spinal curvature measurement requires identification of vertebral landmarks such as transverse processes, but as bones have reduced visibility in ultrasound imaging, skeletal landmarks are typically segmented manually, which is an exceedingly laborious and long process. We propose an automatic algorithm to segment and localize the surface of bony areas in the transverse process for scoliosis in ultrasound. METHODS: The algorithm uses cascade of filters to remove low intensity pixels, smooth the image and detect bony edges. By applying first differentiation, candidate bony areas are classified. The average intensity under each area has a correlation with the possibility of a shadow, and areas with strong shadow are kept for bone segmentation. The segmented images are used to reconstruct a 3-D volume to represent the whole spinal structure around the transverse processes. RESULTS: A comparison between the manual ground truth segmentation and the automatic algorithm in 50 images showed 0.17 mm average difference. The time to process all 1, 938 images was about 37 Sec. (0.0191 Sec./Image), including reading the original sequence file. CONCLUSION: Initial experiments showed the algorithm to be sufficiently accurate and fast for segmentation transverse processes in ultrasound for spinal curvature measurement. An extensive evaluation of the method is currently underway on images from a larger patient cohort and using multiple observers in producing ground truth segmentation.",
keywords = "Scoliosis, Transverse process localization, Ultrasound automatic bone segmentation",
author = "Shahrokh Kamali and Tamas Ungi and Andras Lasso and Christina Yan and Matthew Lougheed and Gabor Fichtinger",
year = "2017",
doi = "10.1117/12.2256007",
language = "English (US)",
volume = "10135",
booktitle = "Medical Imaging 2017",
publisher = "SPIE",

}

TY - GEN

T1 - Localization of the transverse processes in ultrasound for spinal curvature measurement

AU - Kamali, Shahrokh

AU - Ungi, Tamas

AU - Lasso, Andras

AU - Yan, Christina

AU - Lougheed, Matthew

AU - Fichtinger, Gabor

PY - 2017

Y1 - 2017

N2 - PURPOSE: In scoliosis monitoring, tracked ultrasound has been explored as a safer imaging alternative to traditional radiography. The use of ultrasound in spinal curvature measurement requires identification of vertebral landmarks such as transverse processes, but as bones have reduced visibility in ultrasound imaging, skeletal landmarks are typically segmented manually, which is an exceedingly laborious and long process. We propose an automatic algorithm to segment and localize the surface of bony areas in the transverse process for scoliosis in ultrasound. METHODS: The algorithm uses cascade of filters to remove low intensity pixels, smooth the image and detect bony edges. By applying first differentiation, candidate bony areas are classified. The average intensity under each area has a correlation with the possibility of a shadow, and areas with strong shadow are kept for bone segmentation. The segmented images are used to reconstruct a 3-D volume to represent the whole spinal structure around the transverse processes. RESULTS: A comparison between the manual ground truth segmentation and the automatic algorithm in 50 images showed 0.17 mm average difference. The time to process all 1, 938 images was about 37 Sec. (0.0191 Sec./Image), including reading the original sequence file. CONCLUSION: Initial experiments showed the algorithm to be sufficiently accurate and fast for segmentation transverse processes in ultrasound for spinal curvature measurement. An extensive evaluation of the method is currently underway on images from a larger patient cohort and using multiple observers in producing ground truth segmentation.

AB - PURPOSE: In scoliosis monitoring, tracked ultrasound has been explored as a safer imaging alternative to traditional radiography. The use of ultrasound in spinal curvature measurement requires identification of vertebral landmarks such as transverse processes, but as bones have reduced visibility in ultrasound imaging, skeletal landmarks are typically segmented manually, which is an exceedingly laborious and long process. We propose an automatic algorithm to segment and localize the surface of bony areas in the transverse process for scoliosis in ultrasound. METHODS: The algorithm uses cascade of filters to remove low intensity pixels, smooth the image and detect bony edges. By applying first differentiation, candidate bony areas are classified. The average intensity under each area has a correlation with the possibility of a shadow, and areas with strong shadow are kept for bone segmentation. The segmented images are used to reconstruct a 3-D volume to represent the whole spinal structure around the transverse processes. RESULTS: A comparison between the manual ground truth segmentation and the automatic algorithm in 50 images showed 0.17 mm average difference. The time to process all 1, 938 images was about 37 Sec. (0.0191 Sec./Image), including reading the original sequence file. CONCLUSION: Initial experiments showed the algorithm to be sufficiently accurate and fast for segmentation transverse processes in ultrasound for spinal curvature measurement. An extensive evaluation of the method is currently underway on images from a larger patient cohort and using multiple observers in producing ground truth segmentation.

KW - Scoliosis

KW - Transverse process localization

KW - Ultrasound automatic bone segmentation

UR - http://www.scopus.com/inward/record.url?scp=85020400151&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85020400151&partnerID=8YFLogxK

U2 - 10.1117/12.2256007

DO - 10.1117/12.2256007

M3 - Conference contribution

AN - SCOPUS:85020400151

VL - 10135

BT - Medical Imaging 2017

PB - SPIE

ER -