Promising new technologies such as content-based image retrieval can enrich medical informatics. Yet this technology is not well established in PACS, in spite of the advantages. The cause for this lag is usually attributed to its inability to bridge the divide between low-level computational pixel analysis and high-level human cognitive capabilities.
Promising new technologies such as content-based image retrieval can enrich medical informatics. Yet this technology is not well established in PACS, in spite of the advantages. The cause for this lag is usually attributed to its inability to bridge the divide between low-level computational pixel analysis and high-level human cognitive capabilities.A new study from Germany could help narrow that gap. Researchers at the Aachen University of Technology propose a nomenclature and classification scheme for analysis and assessment of medical CBIR systems (J Digit Imaging 2008 Feb. 2 [Epub ahead of print]). "CBIR has come a long way, yet very few systems make it to the forefront of daily clinical, research, and educational use," said Thomas M. Deserno, Ph.D., of the department of medical informatics at Aachen. CBIR enables image access by image pattern rather than by alphanumeric-based indices.Deserno attempted to address the core features and required functionality of medical CBIR systematically, using the concept of gaps as a unifying idea to highlight potential shortcomings in CBIR systems. Since an evaluation of all systems is not practical, an ontology of gaps provides categories and classes of systems, he said.
The literature identifies two gaps in CBIR techniques:
"But we believe there are many other gaps that hinder the use of CBIR techniques in routine medical image management," Deserno said.
The paper defines an ontology of 14 gaps that address image content and features, as well as system performance and usability.There is a highly significant gap, for instance, in the level of integration of CBIR into general patient care information systems. Another gap occurs in the automation of feature extraction, Deserno said. "The paper classifies some prominent CBIR approaches in an effort to spur a more comprehensive view of the concept of gaps in medical CBIR research," he said.
Mammography Study Compares False Positives Between AI and Radiologists in DBT Screening
May 8th 2025For DBT breast cancer screening, 47 percent of radiologist-only flagged false positives involved mass presentations whereas 40 percent of AI-only flagged false positive cases involved benign calcifications, according to research presented at the recent American Roentgen Ray Society (ARRS) conference.