Automatic anatomy recognition in whole-body PET/CT images
Huiqian Wang1, Jayaram K Udupa2, Dewey Odhner2
1College of Optoelectronic Engineering, Chongqing University, Chongqing 400044, China and Medical Image Processing Group Department of Radiology, University of Pennsylvania, Philadelphia, Pennsylvania 19104.
Medical Physics
|January 10, 2016
Summary
This study adapted automatic anatomy recognition (AAR) for whole-body PET/CT scans, improving disease quantification. The new method achieves high accuracy in localizing anatomical objects, crucial for cancer staging and treatment response assessment.
Area of Science:
- Medical Imaging
- Radiology
- Artificial Intelligence in Medicine
Background:
- Whole-body Positron Emission Tomography/Computed Tomography (PET/CT) is vital for cancer management, requiring accurate disease burden quantification.
- Accurate body-wide anatomy recognition is essential for precise quantification but challenging due to limitations in CT and PET image quality.
- Existing Automatic Anatomy Recognition (AAR) methods were primarily body-region-specific and limited in scope.
Purpose of the Study:
- To adapt and evaluate a recently developed Automatic Anatomy Recognition (AAR) methodology for whole-body PET/CT images.
- To assess the object localization accuracy achievable on PET/CT compared to diagnostic CT images.
- To improve body-wide, body-region-wise, and organ-wise quantification of disease burden in PET/CT.
Main Methods:
- Advanced the AAR approach to a whole-body scope, incorporating object-specific texture properties alongside image intensity.
- Implemented an intermodality strategy, leveraging existing fuzzy models from diagnostic CT for PET/CT analysis.
- Developed and compared 15 different strategies for object recognition in PET/CT images using diagnostic CT and PET/CT datasets.
Main Results:
- Object localization accuracy on whole-body low-dose CT images was within three voxels, and on body-region-wise low-dose images, within two voxels.
- Surprisingly, localization error within three voxels was also achievable on direct body-region-wise PET images.
- Recognition performance on low-dose CT images reached levels comparable to those achieved on diagnostic CT images.
Conclusions:
- The body-region-wise AAR approach successfully extends to whole-body torso imaging with comparable object localization performance.
- Combining image texture and intensity properties yields the highest object localization accuracy.
- The proposed framework enables the use of object-specific optimal recognition strategies for enhanced accuracy in PET/CT analysis.
Related Concept Videos
01:10Computed Tomography
9.5K
Tomography refers to imaging by sections. Computed tomography (CT) is a non-invasive imaging technique that uses computers to analyze several cross-sectional X-rays to reveal minute details about structures in the body.
The technique was invented in the 1970s and is based on the principle that as X-rays pass through the body, they are absorbed or reflected at different levels. In the technique, a patient lies on a motorized platform while a computerized axial tomography (CAT) scanner rotates...
The technique was invented in the 1970s and is based on the principle that as X-rays pass through the body, they are absorbed or reflected at different levels. In the technique, a patient lies on a motorized platform while a computerized axial tomography (CAT) scanner rotates...
9.5K
01:29Positron Emission Tomography
8.0K
Positron emission tomography (PET) is a medical imaging technique involving radiopharmaceuticals — substances that emit short-lived radiation. Although the first PET scanner was introduced in 1961, it took 15 more years before radiopharmaceuticals were combined with the technique and revolutionized its potential.
One of the main requirements of a PET scan is a positron-emitting radioisotope, which is produced in a cyclotron and then attached to a substance used by the part of the body...
One of the main requirements of a PET scan is a positron-emitting radioisotope, which is produced in a cyclotron and then attached to a substance used by the part of the body...
8.0K