• AI
  • Molecular Imaging
  • CT
  • X-Ray
  • Ultrasound
  • MRI
  • Facility Management
  • Mammography

Structured query language extension supports content-based image retrieval

Article

Schemes proposed to support growing interest in content-based image retrieval have generally failed to incorporate the use of structured query language to support similarity queries. Recent research from Brazil provides this ability.

Schemes proposed to support growing interest in content-based image retrieval have generally failed to incorporate the use of structured query language to support similarity queries. Recent research from Brazil provides this ability.

Similarity queries scan databases for objects similar to the reference object. To advance the efficiency of similarity searches, several index structures have been proposed, including M-trees, slim-trees, and Dbm-trees. The Brazilian work, however, is among the first to propose extending structured query language (SQL) to support these queries.

"We propose an image handling extension to the relational database management system known as PostgreSQL, with the aim of providing a tool for the development of content-based image retrieval (CBIR) systems," said Denise Guliato, Ph.D., of the Faculdade de Computação, Universidade Federal de Uberlândia in Minas Gerais.

PostgreSQL with Image-handling Extension, or PostgreSQL-IE, makes use of powerful features available with PostgreSQL to facilitate similarity queries, according to Guliato .

In addition to supporting similarity queries, the proposed system extends SQL with new functions able to create new feature extraction procedures, new feature vectors as combinations of previously defined features, and new access methods, she said.

PostgreSQL-IE makes available a new image data type attribute called PGImage, which permits the user to model a relational scheme by storing various images of different classes in the same attribute, she said.

"This novel approach makes it possible to combine visual features of different images in the same feature vector, which is helpful in developing medical applications where each image attribute is composed of a set of images, such as CT, MRI, or mammography," Guliato said.

A paper to illustrate and validate the power of the resources available in PostgreSQL-IE presents a research system that supports content-based retrieval of mammograms, termed SISPRIM.

"This system allows the user to retrieve information from a mammographic database by combining conventional and visual data, using a friendly graphical web interface," Guliato said.

Currently, PostgreSQL-IE offers two conventional similarity operators (K-nearest neighbor and range), but work has commenced to extend PostgreSQL-IE with two new similarity search procedures based on fuzzy sets to take into account the uncertainty present in medical applications of CBIR, Guliato said.

PostgreSQL-IE is open source, portable, extendable, easy to install, applications-independent, and available for Windows or Linux platforms. The script to install the software is available online.

Related Videos
Can Fiber Optic RealShape (FORS) Technology Provide a Viable Alternative to X-Rays for Aortic Procedures?
Does Initial CCTA Provide the Best Assessment of Stable Chest Pain?
Making the Case for Intravascular Ultrasound Use in Peripheral Vascular Interventions
Can Diffusion Microstructural Imaging Provide Insights into Long Covid Beyond Conventional MRI?
Assessing the Impact of Radiology Workforce Shortages in Rural Communities
Emerging MRI and PET Research Reveals Link Between Visceral Abdominal Fat and Early Signs of Alzheimer’s Disease
Reimbursement Challenges in Radiology: An Interview with Richard Heller, MD
Nina Kottler, MD, MS
The Executive Order on AI: Promising Development for Radiology or ‘HIPAA for AI’?
Related Content
© 2024 MJH Life Sciences

All rights reserved.