Title :
Lung image segmentation and registration for quantitative image analysis
Author :
Haneishi, Hideaki ; Ue, Hidenori ; Takita, Noriaki ; Toyama, Hinako ; Miyamoto, Tadaaki ; Yamamoto, Naoyohi ; Mori, Yutaka
Author_Institution :
Chiba Univ., Japan
Abstract :
Functional images obtained PET and/or SPECT become more useful when those images are provided with detail anatomical information obtained by X ray CT or MRI. A series of image processing including image registration and segmentation is presented for quantitative analysis of functional images. A clinical application described in the paper is to evaluate the effect of radiation therapy for lung cancer on lung functions quantitatively. Though SPECT images give information on lung functions such as ventilation and perfusion, it is difficult to identify correctly the location and amount of radioactivity distribution with only those images. To overcome this difficulty, we synthesize both anatomical image (X ray CT, or shortly XCT) and functional image (SPECT) effectively. Furthermore, we have developed a method for dividing lung in XCT image into the lobes that are anatomically meaningful. The segmented lobes in XCT image can be used for quantitative evaluation in each lobe. Change in perfusion at each lobe of lung along with the therapy is presented as an effective example.
Keywords :
cancer; computerised tomography; image registration; image segmentation; lung; medical image processing; positron emission tomography; radiation therapy; single photon emission computed tomography; X ray computerised tomography image; anatomical image; anatomical information; functional images; image processing; lung cancer; lung functions; lung image registration; lung image segmentation; perfusion; quantitative image analysis; radiation therapy; radioactivity distribution; segmented lobes; single photon emission computed tomography images; ventilation; Biomedical applications of radiation; Cancer; Computed tomography; Image analysis; Image processing; Image registration; Image segmentation; Lungs; Magnetic resonance imaging; Positron emission tomography;
Conference_Titel :
Nuclear Science Symposium Conference Record, 2001 IEEE
Print_ISBN :
0-7803-7324-3
DOI :
10.1109/NSSMIC.2001.1008596