News Column

Patent Issued for Method and System for Medical Decision Support Using Organ Models and Learning Based Discriminative Distance Functions

September 1, 2014

By a News Reporter-Staff News Editor at Cardiovascular Week -- According to news reporting originating from Alexandria, Virginia, by NewsRx journalists, a patent by the inventors Voigt, Ingmar (Erlangen, DE); Vitanovski, Dime (Erlangen, DE); Ionasec, Razvan Ioan (Lawrenceville, NJ); Tsymbal, Alexey (Erlangen, DE); Georgescu, Bogdan (Plainsboro, NJ); Zhou, Shaohua Kevin (Plainsboro, NJ); Huber, Martin (Uttenreuth, DE); Comaniciu, Dorin (Princeton Junction, NJ), filed on January 28, 2011, was published online on August 19, 2014 (see also Siemens Aktiengesellschaft).

The assignee for this patent, patent number 8812431, is Siemens Aktiengesellschaft (Munich, DE).

Reporters obtained the following quote from the background information supplied by the inventors: "The present invention relates to model based decision support, and more particularly, to decision support using organ models estimated from medical image data and learning based discriminative distance functions.

"Valvular heart disease (VHD) is a cardiac disorder that affects a large number of patients and often requires elaborate diagnostic procedures, intervention, and long-term management. Abnormalities may occur in conjunction with other heart diseases, and can be caused by congenital defects, pulmonary hypertension, endocarditis, rheumatic fever, and carcinoid heart disease. Such conditions require constant monitoring and a complex clinical workflow, including patient evaluation, percutaneous intervention planning, valve replacement and repair, and follow-up evaluations.

"Treatment of VHD is typically expensive and conventional VHD treatment has a relatively high in-hospital death rate due to elaborate, time consuming, and potentially inaccurate diagnostic procedures and complex interventions into patients' cardiac systems. Recent advances in medical imaging technology have enabled 4D imaging with computed tomography (CT) and ultrasound. However, due to lack of efficient and convenient tools, anatomical performance assessment of the cardiac valves typically relies on manual measurements in 2D image planes derived from the 4D image acquisitions. Although such performance assessment can be error prone and time consuming, diagnosis, treatment decisions, interventional planning, and follow up evaluation typically rely on such performance assessment, which can lead to suboptimal treatment results, follow up interventions, and increased treatment costs. Moreover, clinical decisions are based on generic information from clinical guidelines and publications and personal experience of clinicians. Clinical decisions are not necessarily personalized to the specific patient, due to the potential lack of similar cases, which would provide patient histories and treatment results as references for decision support."

In addition to obtaining background information on this patent, NewsRx editors also obtained the inventors' summary information for this patent: "The present invention provides a method and system for medical decision support using learning based discriminative distance functions and virtual organ models derived from medical image data. Embodiments of the present invention derive high-level information from organ models and extracted features using learning based discriminative distance functions and utilize relative neighborhood graphs to visualize the relationships between various cases. Embodiments of the present invention are flexible and allow similarity between cases to be arbitrarily defined and learned based on any meaningful concept.

"In one embodiment of the present invention, a virtual patient-specific organ model is generated from medical image data of a patient. One or more similar organ models to the patient-specific organ model are retrieved from a plurality of previously stored organ models using a learned discriminative distance function. The learned discriminative distance function can be trained based on various representations and learning algorithms, such as equivalence constraints assigned to the training data using one of Random Forests and a boosting algorithm or the intrinsic Random Forest distance. The patient-specific organ model can be classified into one of a first class and a second class based on the retrieved one or more organ models similar to the patient-specific organ model. The patient specific organ model may be a patient-specific heart valve model generated from 4D cardiac image data.

"These and other advantages of the invention will be apparent to those of ordinary skill in the art by reference to the following detailed description and the accompanying drawings."

For more information, see this patent: Voigt, Ingmar; Vitanovski, Dime; Ionasec, Razvan Ioan; Tsymbal, Alexey; Georgescu, Bogdan; Zhou, Shaohua Kevin; Huber, Martin; Comaniciu, Dorin. Method and System for Medical Decision Support Using Organ Models and Learning Based Discriminative Distance Functions. U.S. Patent Number 8812431, filed January 28, 2011, and published online on August 19, 2014. Patent URL:

Keywords for this news article include: Cardiology, Siemens Aktiengesellschaft.

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: Cardiovascular Week

Story Tools Facebook Linkedin Twitter RSS Feed Email Alerts & Newsletters