Automated prostate tissue referencing for cancer detection and diagnosis

Abstract

BackgroundThe current practice of histopathology review is limited in speed and accuracy. The current diagnostic paradigm does not fully describe the complex and complicated patterns of cancer. To address these needs, we develop an automated and objective system that facilitates a comprehensive and easy information management and decision-making. We also develop a tissue similarity measure scheme to broaden our understanding of tissue characteristics.ResultsThe system includes a database of previously evaluated prostate tissue images, clinical information and a tissue retrieval process. In the system, a tissue is characterized by its morphology. The retrieval process seeks to find the closest matching cases with the tissue of interest. Moreover, we define 9 morphologic criteria by which a pathologist arrives at a histomorphologic diagnosis. Based on the 9 criteria, true tissue similarity is determined and serves as the gold standard of tissue retrieval. Here, we found a minimum of 4 and 3 matching cases, out of 5, for ~80 % and ~60 % of the queries when a match was defined as the tissue similarity score ≥5 and ≥6, respectively. We were also able to examine the relationship between tissues beyond the Gleason grading system due to the tissue similarity scoring system.ConclusionsProviding the closest matching cases and their clinical information with pathologists will help to conduct consistent and reliable diagnoses. Thus, we expect the system to facilitate quality maintenance and quality improvement of cancer pathology.Electronic supplementary materialThe online version of this article (doi:10.1186/s12859-016-1086-6) contains supplementary material, which is available to authorized users.

System Approach to the Development of Intelligent Complexes of Oncological Diagnostics

“…Number of papers are dedicated to research and development of key provisions of the methodology of high-tech information-measuring cancer diagnosis complexes [1][2][3][4][5][6][7][8][9][10][11][12].…”

Similar image search for histopathology: SMILY

“…Relevant tools in non-medical domains include "search by image" for general images 4 , visual search 5 for retail products, and other tools for faces 6 and art 7 . In medical imaging, related works include CBIR for radiology [8][9][10] and pathology [11][12][13][14][15][16][17][18] . Prior machine learning based CBIR systems have employed application-specific models, which require collecting labeled data for each application, creating a significant burden to their implementation.…”

Development of an Intellectual Educational and Diagnostic Complex for the Histological Analysis of Thyroid Tumors

“…Such statistics reveal factors that make predictions about threats to human health. Specialists will generate such statistics at different levels of the patient's examination in order to make the diagnosis more accurate.The relevance of the work is to systematize and compare already existing knowledge about thyroid with new data, which makes it possible to detect regularities in the signs, both benign neoplasms, and in developing malignant tumors [1][2][3][4][5][6][7][8][9][10][11].…”