Pulmonary nodule detection in PET/CT images: Improved approach using combined nodule detection and hybrid FP reduction

Atsushi Teramoto, Hiroshi Fujita, Yoya Tomita, Katsuaki Takahashi, Osamu Yamamuro, Tsuneo Tamaki

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

3 Citations (Scopus)


In this study, an automated scheme for detecting pulmonary nodules in PET/CT images has been proposed using combined detection and hybrid false-positive (FP) reduction techniques. The initial nodule candidates were detected separately from CT and PET images. FPs were then eliminated in the initial candidates by using support vector machine with characteristic values obtained from CT and PET images. In the experiment, we evaluated proposed method using 105 cases of PET/CT images that were obtained in the cancer-screening program. We evaluated true positive fraction (TPF) and FP/case. As a result, TPFs of CT and PET detections were 0.76 and 0.44, respectively. However, by integrating the both results, TPF was reached to 0.82 with 5.14 FPs/case. These results indicate that our method may be of practical use for the detection of pulmonary nodules using PET/CT images.

Original languageEnglish
Title of host publicationMedical Imaging 2012
Subtitle of host publicationComputer-Aided Diagnosis
Publication statusPublished - 2012
EventMedical Imaging 2012: Computer-Aided Diagnosis - San Diego, CA, United States
Duration: 07-02-201209-02-2012

Publication series

NameProgress in Biomedical Optics and Imaging - Proceedings of SPIE
ISSN (Print)1605-7422


OtherMedical Imaging 2012: Computer-Aided Diagnosis
Country/TerritoryUnited States
CitySan Diego, CA

All Science Journal Classification (ASJC) codes

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


Dive into the research topics of 'Pulmonary nodule detection in PET/CT images: Improved approach using combined nodule detection and hybrid FP reduction'. Together they form a unique fingerprint.

Cite this