[Content-based image-retrieval system - development, usefulness and perspectives of diagnostic assistant robot]

Gan To Kagaku Ryoho. 2012 Jul;39(7):1039-43.
[Article in Japanese]

Abstract

In recent years, diagnostic imaging modalities have proliferated from standard X-ray to CT, MRI and PET, and the working environments of radiologists have changed greatly with the popular spread of the PACS system. Radiologists are now facing enormous duties due to the dramatic increase in the volume of images from various modalities, and the shortage of radiologists in Japan has reached near-crisis levels. Furthermore, it is difficult to gain the knowledge needed to interpret diagnostic imaging and modalities under the growing, increasingly diverse and complex modalities and methods, for general physicians and trainees. On the other hand, there are some computer-aided diagnosis and detection systems that support radiologists. Here, we introduce a new diagnostic assistant robot that automatically retrieves cases on record that are similar to new cases, helps in making diagnoses, and can create CT reports semi-automatically, using an existing past CT database of pulmonary nodules with a structured report.

Publication types

  • English Abstract

MeSH terms

  • Databases, Factual
  • Diagnosis, Computer-Assisted / methods*
  • Image Processing, Computer-Assisted / methods*
  • Organizational Case Studies
  • Robotics*