Hybrid CAD scheme for lung nodule detection in PET/CT images

Atsushi Teramoto, Hiroshi Fujita, Yoya Tomita, Katsuaki Takahashi, Osamu Yamamuro, Tsuneo Tamaki, Naoki Hayashi, Shinichi Tamai, Masami Nishio, Wei Ping Chen, Toshiki Kobayashi

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

3 Citations (Scopus)

Abstract

Lung cancer is the leading cause of death among male in the world. PET/CT is useful for the detection of early lung cancer since it is an imaging technique that has functional and anatomical information. However, radiologist has to examine using the large number of images. Therefore reduction of radiologist's load is strongly desired. In this study, hybrid CAD scheme has been proposed to detect lung nodule in PET/CT images. Proposed method detects the lung nodule from both CT and PET images. As for the detection in CT images, solitary nodules are detected using Cylindrical Filter that we developed. PET images are binarized based on standard uptake value (SUV); highly uptake regions are detected. FP reduction is performed using seven characteristic features and Support Vector Machine. Finally by integrating these results, candidate regions are obtained. In the experiment, we evaluated proposed method using 50 cases of PET/CT images obtained for the cancer-screening program. We evaluated true-positive fraction (TPF) and the number of false positives / case (FPs/case). As a result, TPFs for CT and PET were 0.67 and 0.38, respectively. By integrating the both results, TPF was improved to 0.80. These results indicate that our method may be useful for the lung cancer detection using PET/CT images.

Original languageEnglish
Title of host publicationMedical Imaging 2011
Subtitle of host publicationComputer-Aided Diagnosis
Volume7963
DOIs
Publication statusPublished - 13-05-2011
EventMedical Imaging 2011: Computer-Aided Diagnosis - Lake Buena Vista, FL, United States
Duration: 15-02-201117-02-2011

Other

OtherMedical Imaging 2011: Computer-Aided Diagnosis
CountryUnited States
CityLake Buena Vista, FL
Period15-02-1117-02-11

Fingerprint

nodules
computer aided design
lungs
Lung Neoplasms
Computer aided design
Lung
Support vector machines
cancer
Screening
Early Detection of Cancer
Imaging techniques
Cause of Death
Experiments
death
imaging techniques
screening
Positron Emission Tomography Computed Tomography
Radiologists
filters
causes

All Science Journal Classification (ASJC) codes

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

Cite this

Teramoto, A., Fujita, H., Tomita, Y., Takahashi, K., Yamamuro, O., Tamaki, T., ... Kobayashi, T. (2011). Hybrid CAD scheme for lung nodule detection in PET/CT images. In Medical Imaging 2011: Computer-Aided Diagnosis (Vol. 7963). [796335] https://doi.org/10.1117/12.877104
Teramoto, Atsushi ; Fujita, Hiroshi ; Tomita, Yoya ; Takahashi, Katsuaki ; Yamamuro, Osamu ; Tamaki, Tsuneo ; Hayashi, Naoki ; Tamai, Shinichi ; Nishio, Masami ; Chen, Wei Ping ; Kobayashi, Toshiki. / Hybrid CAD scheme for lung nodule detection in PET/CT images. Medical Imaging 2011: Computer-Aided Diagnosis. Vol. 7963 2011.
@inproceedings{61096bbef5ac418296540f8aa466a18c,
title = "Hybrid CAD scheme for lung nodule detection in PET/CT images",
abstract = "Lung cancer is the leading cause of death among male in the world. PET/CT is useful for the detection of early lung cancer since it is an imaging technique that has functional and anatomical information. However, radiologist has to examine using the large number of images. Therefore reduction of radiologist's load is strongly desired. In this study, hybrid CAD scheme has been proposed to detect lung nodule in PET/CT images. Proposed method detects the lung nodule from both CT and PET images. As for the detection in CT images, solitary nodules are detected using Cylindrical Filter that we developed. PET images are binarized based on standard uptake value (SUV); highly uptake regions are detected. FP reduction is performed using seven characteristic features and Support Vector Machine. Finally by integrating these results, candidate regions are obtained. In the experiment, we evaluated proposed method using 50 cases of PET/CT images obtained for the cancer-screening program. We evaluated true-positive fraction (TPF) and the number of false positives / case (FPs/case). As a result, TPFs for CT and PET were 0.67 and 0.38, respectively. By integrating the both results, TPF was improved to 0.80. These results indicate that our method may be useful for the lung cancer detection using PET/CT images.",
author = "Atsushi Teramoto and Hiroshi Fujita and Yoya Tomita and Katsuaki Takahashi and Osamu Yamamuro and Tsuneo Tamaki and Naoki Hayashi and Shinichi Tamai and Masami Nishio and Chen, {Wei Ping} and Toshiki Kobayashi",
year = "2011",
month = "5",
day = "13",
doi = "10.1117/12.877104",
language = "English",
isbn = "9780819485052",
volume = "7963",
booktitle = "Medical Imaging 2011",

}

Teramoto, A, Fujita, H, Tomita, Y, Takahashi, K, Yamamuro, O, Tamaki, T, Hayashi, N, Tamai, S, Nishio, M, Chen, WP & Kobayashi, T 2011, Hybrid CAD scheme for lung nodule detection in PET/CT images. in Medical Imaging 2011: Computer-Aided Diagnosis. vol. 7963, 796335, Medical Imaging 2011: Computer-Aided Diagnosis, Lake Buena Vista, FL, United States, 15-02-11. https://doi.org/10.1117/12.877104

Hybrid CAD scheme for lung nodule detection in PET/CT images. / Teramoto, Atsushi; Fujita, Hiroshi; Tomita, Yoya; Takahashi, Katsuaki; Yamamuro, Osamu; Tamaki, Tsuneo; Hayashi, Naoki; Tamai, Shinichi; Nishio, Masami; Chen, Wei Ping; Kobayashi, Toshiki.

Medical Imaging 2011: Computer-Aided Diagnosis. Vol. 7963 2011. 796335.

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

TY - GEN

T1 - Hybrid CAD scheme for lung nodule detection in PET/CT images

AU - Teramoto, Atsushi

AU - Fujita, Hiroshi

AU - Tomita, Yoya

AU - Takahashi, Katsuaki

AU - Yamamuro, Osamu

AU - Tamaki, Tsuneo

AU - Hayashi, Naoki

AU - Tamai, Shinichi

AU - Nishio, Masami

AU - Chen, Wei Ping

AU - Kobayashi, Toshiki

PY - 2011/5/13

Y1 - 2011/5/13

N2 - Lung cancer is the leading cause of death among male in the world. PET/CT is useful for the detection of early lung cancer since it is an imaging technique that has functional and anatomical information. However, radiologist has to examine using the large number of images. Therefore reduction of radiologist's load is strongly desired. In this study, hybrid CAD scheme has been proposed to detect lung nodule in PET/CT images. Proposed method detects the lung nodule from both CT and PET images. As for the detection in CT images, solitary nodules are detected using Cylindrical Filter that we developed. PET images are binarized based on standard uptake value (SUV); highly uptake regions are detected. FP reduction is performed using seven characteristic features and Support Vector Machine. Finally by integrating these results, candidate regions are obtained. In the experiment, we evaluated proposed method using 50 cases of PET/CT images obtained for the cancer-screening program. We evaluated true-positive fraction (TPF) and the number of false positives / case (FPs/case). As a result, TPFs for CT and PET were 0.67 and 0.38, respectively. By integrating the both results, TPF was improved to 0.80. These results indicate that our method may be useful for the lung cancer detection using PET/CT images.

AB - Lung cancer is the leading cause of death among male in the world. PET/CT is useful for the detection of early lung cancer since it is an imaging technique that has functional and anatomical information. However, radiologist has to examine using the large number of images. Therefore reduction of radiologist's load is strongly desired. In this study, hybrid CAD scheme has been proposed to detect lung nodule in PET/CT images. Proposed method detects the lung nodule from both CT and PET images. As for the detection in CT images, solitary nodules are detected using Cylindrical Filter that we developed. PET images are binarized based on standard uptake value (SUV); highly uptake regions are detected. FP reduction is performed using seven characteristic features and Support Vector Machine. Finally by integrating these results, candidate regions are obtained. In the experiment, we evaluated proposed method using 50 cases of PET/CT images obtained for the cancer-screening program. We evaluated true-positive fraction (TPF) and the number of false positives / case (FPs/case). As a result, TPFs for CT and PET were 0.67 and 0.38, respectively. By integrating the both results, TPF was improved to 0.80. These results indicate that our method may be useful for the lung cancer detection using PET/CT images.

UR - http://www.scopus.com/inward/record.url?scp=79955757407&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79955757407&partnerID=8YFLogxK

U2 - 10.1117/12.877104

DO - 10.1117/12.877104

M3 - Conference contribution

SN - 9780819485052

VL - 7963

BT - Medical Imaging 2011

ER -

Teramoto A, Fujita H, Tomita Y, Takahashi K, Yamamuro O, Tamaki T et al. Hybrid CAD scheme for lung nodule detection in PET/CT images. In Medical Imaging 2011: Computer-Aided Diagnosis. Vol. 7963. 2011. 796335 https://doi.org/10.1117/12.877104