News Column

Patent Issued for Classification of Medical Diagnostic Images

September 3, 2014



By a News Reporter-Staff News Editor at Journal of Engineering -- According to news reporting originating from Alexandria, Virginia, by VerticalNews journalists, a patent by the inventors Nielsen, Mads (Dragor, DK); De Bruijne, Marleen (Rotterdam, NL); Sorensen, Lauge (Frederiksberg, DK), filed on May 9, 2011, was published online on August 19, 2014.

The assignee for this patent, patent number 8811724, is The University of Copenhagen (Copenhagen, DK).

Reporters obtained the following quote from the background information supplied by the inventors: "Current quantitative measures of chronic obstructive pulmonary disease (COPD) are limited in several ways. Pulmonary function tests (PFTs) are the gold standard for assessment and diagnosis of COPD [23]. These are cheap and fast to acquire but are limited by insensitivity to early stages of COPD and lack of reproducibility [8]. Visual and computerized assessment in computed tomography (CT) imaging has emerged as an alternative that directly can measure the two components of COPD, namely, small airway disease and emphysema. However, it is difficult to assess disease severity and progression visually. Moreover, visual assessment is subjective, time-consuming, and suffers from intra-observer and inter-observer variability [3, 18]. The most widely used computerized measures, also referred to as densitometry or quantitative CT, are relative area of emphysema (RA) [18] and percentile density (PD) [11]. These measures consider each parenchyma voxel in the CT image independently thereby disregarding potentially valuable information, such as spatial relations between voxels of similar or differing intensities and patterns at larger scales, and they are restricted to a single threshold parameter, which makes them sensitive to noise in the CT images.

"COPD is estimated to become the fifth most burdening disease and the third leading cause of death worldwide by 2020[23], and the limitations of current quantitative measures may hinder progression in research on treatments for the disease. Further, more and more medical data is being produced, both in routine clinical practice, and in screening and epidemiological studies, increasing the demand for robust and sensitive automatic methods for quantification.

"Supervised texture classification in CT where a classifier is trained on manually annotated regions of interest (ROIs) [5, 20, 22, 27, 28, 30, 32] uses much more of the information available in the CT images compared to the densitometric measures, and the output of a trained classifier can be used for COPD quantification by fusion of individual voxel posterior probabilities [16, 20, 28]. However, this approach requires labeled data, which is usually acquired by manual annotation done by human experts. Manual annotation suffers from the same limitations as visual assessment of emphysema in CT images [3, 18], moreover, it is hard to demarcate or position ROIs in CT images, since the appearance of the disease patterns is subtle and diffuse, especially at early stages of COPD. Further, analysis is limited to current knowledge and experience of the experts, and there can be a bias towards typical cases in the annotated data set."

In addition to obtaining background information on this patent, VerticalNews editors also obtained the inventors' summary information for this patent: "The present invention provides the automated classification of a medical diagnostic image of a lung according to its deduced probability of relating to a lung of a patient who is suffering from a diffuse parenchymal lung disease such as chronic obstructive pulmonary disease (COPD), cystic fibrosis, or severe asthma, or to a class of patients characterised by the severity of such a condition, or to a class of patients characterised by a prognostic likelihood of developing such a condition or severity of condition.

"The invention includes a fully automatic, data-driven approach for texture-based quantitative analysis of chronic obstructive pulmonary disease (COPD) in pulmonary computed tomography (CT) images. Whilst the invention is not limited to this, it will be described herein for illustrative purposes in detail in such a context. The approach described in detail herein uses supervised learning, without however requiring manually annotated data. Instead, pulmonary function testing (PFT) data or other metadata is used to acquire the data labels."

For more information, see this patent: Nielsen, Mads; De Bruijne, Marleen; Sorensen, Lauge. Classification of Medical Diagnostic Images. U.S. Patent Number 8811724, filed May 9, 2011, and published online on August 19, 2014. Patent URL: http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sect2=HITOFF&d=PALL&p=1&u=%2Fnetahtml%2FPTO%2Fsrchnum.htm&r=1&f=G&l=50&s1=8811724.PN.&OS=PN/8811724RS=PN/8811724

Keywords for this news article include: Emphysema, Imaging Technology, The University of Copenhagen, Chronic Obstructive Pulmonary Disease.

Our reports deliver fact-based news of research and discoveries from around the world. Copyright 2014, NewsRx LLC


For more stories covering the world of technology, please see HispanicBusiness' Tech Channel



Source: Journal of Engineering


Story Tools






HispanicBusiness.com Facebook Linkedin Twitter RSS Feed Email Alerts & Newsletters