Automatic anatomy recognition in post-tonsillectomy MR images of obese children with OSAS

Yubing Tong, Jayaram K. Udupa, Dewey Odhner, Sanghun Sin, Raanan Arens

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

1 Citation (Scopus)

Abstract

Automatic Anatomy Recognition (AAR) is a recently developed approach for the automatic whole body wide organ segmentation. We previously tested that methodology on image cases with some pathology where the organs were not distorted significantly. In this paper, we present an advancement of AAR to handle organs which may have been modified or resected by surgical intervention. We focus on MRI of the neck in pediatric Obstructive Sleep Apnea Syndrome (OSAS). The proposed method consists of an AAR step followed by support vector machine techniques to detect the presence/absence of organs. The AAR step employs a hierarchical organization of the organs for model building. For each organ, a fuzzy model over a population is built. The model of the body region is then described in terms of the fuzzy models and a host of other descriptors which include parent to offspring relationship estimated over the population. Organs are recognized following the organ hierarchy by using an optimal threshold based search. The SVM step subsequently checks for evidence of the presence of organs. Experimental results show that AAR techniques can be combined with machine learning strategies within the AAR recognition framework for good performance in recognizing missing organs, in our case missing tonsils in post-tonsillectomy images as well as in simulating tonsillectomy images. The previous recognition performance is maintained achieving an organ localization accuracy of within 1 voxel when the organ is actually not removed. To our knowledge, no methods have been reported to date for handling significantly deformed or missing organs, especially in neck MRI.

Original languageEnglish (US)
Title of host publicationProgress in Biomedical Optics and Imaging - Proceedings of SPIE
PublisherSPIE
Volume9414
ISBN (Print)9781628415049
DOIs
StatePublished - 2015
EventSPIE Medical Imaging Symposium 2015: Computer-Aided Diagnosis - Orlando, United States
Duration: Feb 22 2015Feb 25 2015

Other

OtherSPIE Medical Imaging Symposium 2015: Computer-Aided Diagnosis
CountryUnited States
CityOrlando
Period2/22/152/25/15

Fingerprint

sleep
Tonsillectomy
anatomy
respiration
Obstructive Sleep Apnea
organs
Anatomy
Magnetic resonance imaging
Pediatrics
Pathology
Neck
Support vector machines
Learning systems
Body Regions
Palatine Tonsil
Population
Sleep
machine learning
pathology
hierarchies

Keywords

  • Adenotonsillectomy
  • Automatic anatomy recognition-AAR
  • Missing organs detection
  • Multi-object segmentation

ASJC Scopus subject areas

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

Cite this

Tong, Y., Udupa, J. K., Odhner, D., Sin, S., & Arens, R. (2015). Automatic anatomy recognition in post-tonsillectomy MR images of obese children with OSAS. In Progress in Biomedical Optics and Imaging - Proceedings of SPIE (Vol. 9414). [94140Z] SPIE. https://doi.org/10.1117/12.2081912

Automatic anatomy recognition in post-tonsillectomy MR images of obese children with OSAS. / Tong, Yubing; Udupa, Jayaram K.; Odhner, Dewey; Sin, Sanghun; Arens, Raanan.

Progress in Biomedical Optics and Imaging - Proceedings of SPIE. Vol. 9414 SPIE, 2015. 94140Z.

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

Tong, Y, Udupa, JK, Odhner, D, Sin, S & Arens, R 2015, Automatic anatomy recognition in post-tonsillectomy MR images of obese children with OSAS. in Progress in Biomedical Optics and Imaging - Proceedings of SPIE. vol. 9414, 94140Z, SPIE, SPIE Medical Imaging Symposium 2015: Computer-Aided Diagnosis, Orlando, United States, 2/22/15. https://doi.org/10.1117/12.2081912
Tong Y, Udupa JK, Odhner D, Sin S, Arens R. Automatic anatomy recognition in post-tonsillectomy MR images of obese children with OSAS. In Progress in Biomedical Optics and Imaging - Proceedings of SPIE. Vol. 9414. SPIE. 2015. 94140Z https://doi.org/10.1117/12.2081912
Tong, Yubing ; Udupa, Jayaram K. ; Odhner, Dewey ; Sin, Sanghun ; Arens, Raanan. / Automatic anatomy recognition in post-tonsillectomy MR images of obese children with OSAS. Progress in Biomedical Optics and Imaging - Proceedings of SPIE. Vol. 9414 SPIE, 2015.
@inproceedings{adad1dcf30d1427fb3b61f5cce5853b7,
title = "Automatic anatomy recognition in post-tonsillectomy MR images of obese children with OSAS",
abstract = "Automatic Anatomy Recognition (AAR) is a recently developed approach for the automatic whole body wide organ segmentation. We previously tested that methodology on image cases with some pathology where the organs were not distorted significantly. In this paper, we present an advancement of AAR to handle organs which may have been modified or resected by surgical intervention. We focus on MRI of the neck in pediatric Obstructive Sleep Apnea Syndrome (OSAS). The proposed method consists of an AAR step followed by support vector machine techniques to detect the presence/absence of organs. The AAR step employs a hierarchical organization of the organs for model building. For each organ, a fuzzy model over a population is built. The model of the body region is then described in terms of the fuzzy models and a host of other descriptors which include parent to offspring relationship estimated over the population. Organs are recognized following the organ hierarchy by using an optimal threshold based search. The SVM step subsequently checks for evidence of the presence of organs. Experimental results show that AAR techniques can be combined with machine learning strategies within the AAR recognition framework for good performance in recognizing missing organs, in our case missing tonsils in post-tonsillectomy images as well as in simulating tonsillectomy images. The previous recognition performance is maintained achieving an organ localization accuracy of within 1 voxel when the organ is actually not removed. To our knowledge, no methods have been reported to date for handling significantly deformed or missing organs, especially in neck MRI.",
keywords = "Adenotonsillectomy, Automatic anatomy recognition-AAR, Missing organs detection, Multi-object segmentation",
author = "Yubing Tong and Udupa, {Jayaram K.} and Dewey Odhner and Sanghun Sin and Raanan Arens",
year = "2015",
doi = "10.1117/12.2081912",
language = "English (US)",
isbn = "9781628415049",
volume = "9414",
booktitle = "Progress in Biomedical Optics and Imaging - Proceedings of SPIE",
publisher = "SPIE",

}

TY - GEN

T1 - Automatic anatomy recognition in post-tonsillectomy MR images of obese children with OSAS

AU - Tong, Yubing

AU - Udupa, Jayaram K.

AU - Odhner, Dewey

AU - Sin, Sanghun

AU - Arens, Raanan

PY - 2015

Y1 - 2015

N2 - Automatic Anatomy Recognition (AAR) is a recently developed approach for the automatic whole body wide organ segmentation. We previously tested that methodology on image cases with some pathology where the organs were not distorted significantly. In this paper, we present an advancement of AAR to handle organs which may have been modified or resected by surgical intervention. We focus on MRI of the neck in pediatric Obstructive Sleep Apnea Syndrome (OSAS). The proposed method consists of an AAR step followed by support vector machine techniques to detect the presence/absence of organs. The AAR step employs a hierarchical organization of the organs for model building. For each organ, a fuzzy model over a population is built. The model of the body region is then described in terms of the fuzzy models and a host of other descriptors which include parent to offspring relationship estimated over the population. Organs are recognized following the organ hierarchy by using an optimal threshold based search. The SVM step subsequently checks for evidence of the presence of organs. Experimental results show that AAR techniques can be combined with machine learning strategies within the AAR recognition framework for good performance in recognizing missing organs, in our case missing tonsils in post-tonsillectomy images as well as in simulating tonsillectomy images. The previous recognition performance is maintained achieving an organ localization accuracy of within 1 voxel when the organ is actually not removed. To our knowledge, no methods have been reported to date for handling significantly deformed or missing organs, especially in neck MRI.

AB - Automatic Anatomy Recognition (AAR) is a recently developed approach for the automatic whole body wide organ segmentation. We previously tested that methodology on image cases with some pathology where the organs were not distorted significantly. In this paper, we present an advancement of AAR to handle organs which may have been modified or resected by surgical intervention. We focus on MRI of the neck in pediatric Obstructive Sleep Apnea Syndrome (OSAS). The proposed method consists of an AAR step followed by support vector machine techniques to detect the presence/absence of organs. The AAR step employs a hierarchical organization of the organs for model building. For each organ, a fuzzy model over a population is built. The model of the body region is then described in terms of the fuzzy models and a host of other descriptors which include parent to offspring relationship estimated over the population. Organs are recognized following the organ hierarchy by using an optimal threshold based search. The SVM step subsequently checks for evidence of the presence of organs. Experimental results show that AAR techniques can be combined with machine learning strategies within the AAR recognition framework for good performance in recognizing missing organs, in our case missing tonsils in post-tonsillectomy images as well as in simulating tonsillectomy images. The previous recognition performance is maintained achieving an organ localization accuracy of within 1 voxel when the organ is actually not removed. To our knowledge, no methods have been reported to date for handling significantly deformed or missing organs, especially in neck MRI.

KW - Adenotonsillectomy

KW - Automatic anatomy recognition-AAR

KW - Missing organs detection

KW - Multi-object segmentation

UR - http://www.scopus.com/inward/record.url?scp=84948765564&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84948765564&partnerID=8YFLogxK

U2 - 10.1117/12.2081912

DO - 10.1117/12.2081912

M3 - Conference contribution

SN - 9781628415049

VL - 9414

BT - Progress in Biomedical Optics and Imaging - Proceedings of SPIE

PB - SPIE

ER -