Browse Prior Art Database

INTERACTIVE REAL-TIME ANATOMY QUERY SYSTEM FOR RAPID PROTOTYPING

IP.com Disclosure Number: IPCOM000238876D
Publication Date: 2014-Sep-23
Document File: 5 page(s) / 178K

Publishing Venue

The IP.com Prior Art Database

Abstract

The invention proposes a technique for creating a simple yet powerful google like search bar for three dimensional (3D) medical image scans. Such a search bar is used to visually/interactively query and navigate patient anatomical regions to localize any region of interest based on its neighboring spatial and contextual relationship with its surrounding organs/regions. The technique further includes creating a high level English language driven query system that abstracts out every dependencies on imaging parameters. The parameters include, image field of view (FOV), coverage, spacing to automatically compute a personalized anatomical atlas per patient and match spacing with a patient invariant query system that is easily used by anyone with knowledge of human anatomy, such as, radiologists/anatomists. The query thus generated is used to segment semantically similar regions across patients.

This text was extracted from a Microsoft Word document.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 52% of the total text.

INTERACTIVE REAL-TIME ANATOMY QUERY SYSTEM FOR RAPID PROTOTYPING

BRIEF ABSTRACT

The invention proposes a technique for creating a simple yet powerful google like search bar for three dimensional (3D) medical image scans. Such a search bar is used to visually/interactively query and navigate patient anatomical regions to localize any region of interest based on its neighboring spatial and contextual relationship with its surrounding organs/regions. The technique further includes creating a high level English language driven query system that abstracts out every dependencies on imaging parameters. The parameters include, image field of view (FOV), coverage, spacing to automatically compute a personalized anatomical atlas per patient and match spacing with a patient invariant query system that is easily used by anyone with knowledge of human anatomy, such as, radiologists/anatomists. The query thus generated is used to segment semantically similar regions across patients.

KEYWORDS

Real-time prototyping, image localization, query, algorithm

DETAILED DESCRIPTION

All image analysis algorithms in medical image analysis domains work on image x-y-z grid and have very little knowledge of anatomy if any. Also, all the image analysis algorithms are not easily re-usable for other algorithm development. As a result, prototyping delays are achieved which increases clinical application research cycle time.

In a conventional technique semantic image tagging is used to enable automatic detection and localization of anatomical structures within three dimensional (3D) computed tomography (CT) scans. The technique includes a library to recognize various organs within CT images. The technique employs visual recognition that focuses on the analysis of patient scans using machine learning techniques for automatic detection and segmentation of healthy anatomy, as well as anomalies. Further, the technique includes an algorithm that predicts with high accuracy which web search results a user will click for repeat queries. Large-scale face image search is also incorporated into multimedia search to index and show images from a search query.

However, the above mentioned conventional technique does not provide an interactive real-time query system for three dimensional (3D) medical image scans.

Therefore, there is a need for an efficient technique that provides interactive real-time query system for rapid prototyping anatomical localization.

The invention proposes a technique for creating a simple yet powerful google like search bar for three dimensional (3D) medical image scans. Such a search bar is used to visually/interactively query and navigate patient anatomical regions to localize any region of interest based on its neighboring spatial and contextual relationship with its surrounding organs/regions. The technique further includes creating a high level English language driven query system that abstracts out every dependencies on imaging parameters. The parameters in...