DocumentCode :
2484467
Title :
Automated lymph node detection and classification on breast and prostate cancer SPECT-CT images
Author :
Papp, Laszlo ; Zsoter, Norbert ; Loh, Charlotte ; Ole, Baeumer ; Egeler, Bernhard ; Garai, Ildiko ; Luetzen, Ulf
Author_Institution :
Mediso Med. Imaging Syst. Ltd., Budapest, Hungary
fYear :
2011
fDate :
Aug. 30 2011-Sept. 3 2011
Firstpage :
3431
Lastpage :
3434
Abstract :
We present a novel detection and classification method to process SPECT-CT images representing breast and prostate lymph nodes. Lymph nodes are those nodes that are near the primer tumor and may become cancerous in time, hence their early detection is a key factor for the successful treatment of the patient. Prior methods focus on the visual aid to manually detect the lymph nodes which still makes the process time-consuming. Other solutions segment the lymph nodes only on CT, where the small lymph nodes may not be located accurately. Our solution processed both SPECT and CT data to provide an accurate classification of all SPECT hot spots. The method has been validated on a huge amount of medical data. Results show that our method is a very effective tool to support physicians working with related images in the field of nuclear medicine.
Keywords :
biological organs; cancer; computerised tomography; image classification; mammography; single photon emission computed tomography; tumours; SPECT-CT images; automated lymph node classification; automated lymph node detection; breast cancer; breast lymph nodes; prostate cancer; prostate lymph nodes; tumor; Biomedical imaging; Bones; Breast; Computed tomography; Image segmentation; Lymph nodes; Single photon emission computed tomography; Automation; Breast Neoplasms; Female; Humans; Lymphatic Metastasis; Male; Prostatic Neoplasms; Tomography, Emission-Computed, Single-Photon; Tomography, X-Ray Computed;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Engineering in Medicine and Biology Society, EMBC, 2011 Annual International Conference of the IEEE
Conference_Location :
Boston, MA
ISSN :
1557-170X
Print_ISBN :
978-1-4244-4121-1
Electronic_ISBN :
1557-170X
Type :
conf
DOI :
10.1109/IEMBS.2011.6090928
Filename :
6090928
Link To Document :
بازگشت