Patents by Inventor Karsten Sternickel
Karsten Sternickel has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).
-
Patent number: 9655564Abstract: The use of machine learning for pattern recognition in magnetocardiography (MCG) that measures magnetic fields emitted by the electrophysiological activity of the heart is disclosed herein. Direct kernel methods are used to separate abnormal MCG heart patterns from normal ones. For unsupervised learning, Direct Kernel based Self-Organizing Maps are introduced. For supervised learning Direct Kernel Partial Least Squares and (Direct) Kernel Ridge Regression are used. These results are then compared with classical Support Vector Machines and Kernel Partial Least Squares. The hyper-parameters for these methods are tuned on a validation subset of the training data before testing. Also investigated is the most effective pre-processing, using local, vertical, horizontal and two-dimensional (global) Mahanalobis scaling, wavelet transforms, and variable selection by filtering.Type: GrantFiled: October 30, 2015Date of Patent: May 23, 2017Assignee: Cardio Mag Imaging, Inc.Inventors: Karsten Sternickel, Boleslaw Szymanski, Mark Embrechts
-
Publication number: 20160066860Abstract: The use of machine learning for pattern recognition in magnetocardiography (MCG) that measures magnetic fields emitted by the electrophysiological activity of the heart is disclosed herein. Direct kernel methods are used to separate abnormal MCG heart patterns from normal ones. For unsupervised learning, Direct Kernel based Self-Organizing Maps are introduced. For supervised learning Direct Kernel Partial Least Squares and (Direct) Kernel Ridge Regression are used. These results are then compared with classical Support Vector Machines and Kernel Partial Least Squares. The hyper-parameters for these methods are tuned on a validation subset of the training data before testing. Also investigated is the most effective pre-processing, using local, vertical, horizontal and two-dimensional (global) Mahanalobis scaling, wavelet transforms, and variable selection by filtering.Type: ApplicationFiled: October 30, 2015Publication date: March 10, 2016Inventors: Karsten Sternickel, Boleslaw Szymanski, Mark Embrechts
-
Patent number: 9173614Abstract: The use of machine learning for pattern recognition in magnetocardiography (MCG) that measures magnetic fields emitted by the electrophysiological activity of the heart is disclosed herein. Direct kernel methods are used to separate abnormal MCG heart patterns from normal ones. For unsupervised learning, Direct Kernel based Self-Organizing Maps are introduced. For supervised learning Direct Kernel Partial Least Squares and (Direct) Kernel Ridge Regression are used. These results are then compared with classical Support Vector Machines and Kernel Partial Least Squares. The hyper-parameters for these methods are tuned on a validation subset of the training data before testing. Also investigated is the most effective pre-processing, using local, vertical, horizontal and two-dimensional (global) Mahanalobis scaling, wavelet transforms, and variable selection by filtering.Type: GrantFiled: May 28, 2014Date of Patent: November 3, 2015Assignee: CardioMag Imaging, Inc.Inventors: Karsten Sternickel, Boleslaw Szymanski, Mark Embrechts
-
Publication number: 20140343396Abstract: The use of machine learning for pattern recognition in magnetocardiography (MCG) that measures magnetic fields emitted by the electrophysiological activity of the heart is disclosed herein. Direct kernel methods are used to separate abnormal MCG heart patterns from normal ones. For unsupervised learning, Direct Kernel based Self-Organizing Maps are introduced. For supervised learning Direct Kernel Partial Least Squares and (Direct) Kernel Ridge Regression are used. These results are then compared with classical Support Vector Machines and Kernel Partial Least Squares. The hyper-parameters for these methods are tuned on a validation subset of the training data before testing. Also investigated is the most effective pre-processing, using local, vertical, horizontal and two-dimensional (global) Mahanalobis scaling, wavelet transforms, and variable selection by filtering.Type: ApplicationFiled: May 28, 2014Publication date: November 20, 2014Applicant: CardioMag Imaging, Inc.Inventors: Karsten Sternickel, Boleslaw Szymanski, Mark Embrechts
-
Patent number: 8744557Abstract: The use of machine learning for pattern recognition in magnetocardiography (MCG) that measures magnetic fields emitted by the electrophysiological activity of the heart is disclosed herein. Direct kernel methods are used to separate abnormal MCG heart patterns from normal ones. For unsupervised learning, Direct Kernel based Self-Organizing Maps are introduced. For supervised learning Direct Kernel Partial Least Squares and (Direct) Kernel Ridge Regression are used. These results are then compared with classical Support Vector Machines and Kernel Partial Least Squares. The hyper-parameters for these methods are tuned on a validation subset of the training data before testing. Also investigated is the most effective pre-processing, using local, vertical, horizontal and two-dimensional (global) Mahanalobis scaling, wavelet transforms, and variable selection by filtering.Type: GrantFiled: February 20, 2013Date of Patent: June 3, 2014Assignee: Cardio Mag Imaging, Inc.Inventors: Karsten Sternickel, Boleslaw Szymanski, Mark Embrechts
-
Patent number: 8527435Abstract: A novel Levenberg-Marquardt like second-order algorithm for tuning the Parzen window ? in a Radial Basis Function (Gaussian) kernel. Each attribute has its own sigma parameter. The values of the optimized ? are then used as a gauge for variable selection. Kernel Partial Least Squares (K-PLS) model is applied to several benchmark data sets to estimate effectiveness of second-order sigma tuning procedure for an RBF kernel. The variable subset selection method based on these sigma values is then compared with different feature selection procedures such as random forests and sensitivity analysis. The sigma-tuned RBF kernel model outperforms K-PLS and SVM models with a single sigma value. K-PLS models also compare favorably with Least Squares Support Vector Machines (LS-SVM), epsilon-insensitive Support Vector Regression and traditional PLS.Type: GrantFiled: July 13, 2011Date of Patent: September 3, 2013Assignee: CardioMag Imaging, Inc.Inventors: Long Han, Mark Embrechts, Boleslaw Szymanski, Karsten Sternickel, Alexander Ross
-
Patent number: 8391963Abstract: The use of machine learning for pattern recognition in magnetocardiography (MCG) that measures magnetic fields emitted by the electrophysiological activity of the heart is disclosed herein. Direct kernel methods are used to separate abnormal MCG heart patterns from normal ones. For unsupervised learning, Direct Kernel based Self-Organizing Maps are introduced. For supervised learning Direct Kernel Partial Least Squares and (Direct) Kernel Ridge Regression are used. These results are then compared with classical Support Vector Machines and Kernel Partial Least Squares. The hyper-parameters for these methods are tuned on a validation subset of the training data before testing. Also investigated is the most effective pre-processing, using local, vertical, horizontal and two-dimensional (global) Mahanalobis scaling, wavelet transforms, and variable selection by filtering.Type: GrantFiled: June 18, 2010Date of Patent: March 5, 2013Assignee: CardioMag Imaging, Inc.Inventors: Karsten Sternickel, Boleslaw Szymanski, Mark Embrechts
-
Publication number: 20110047105Abstract: The use of machine learning for pattern recognition in magnetocardiography (MCG) that measures magnetic fields emitted by the electrophysiological activity of the heart is disclosed herein. Direct kernel methods are used to separate abnormal MCG heart patterns from normal ones. For unsupervised learning, Direct Kernel based Self-Organizing Maps are introduced. For supervised learning Direct Kernel Partial Least Squares and (Direct) Kernel Ridge Regression are used. These results are then compared with classical Support Vector Machines and Kernel Partial Least Squares. The hyper-parameters for these methods are tuned on a validation subset of the training data before testing. Also investigated is the most effective pre-processing, using local, vertical, horizontal and two-dimensional (global) Mahanalobis scaling, wavelet transforms, and variable selection by filtering.Type: ApplicationFiled: June 18, 2010Publication date: February 24, 2011Applicant: CARDIO MAG IMAGING, INC.Inventors: Karsten Sternickel, Mark J. Embrechts, Boleslaw Szymanski
-
Patent number: 7742806Abstract: The use of machine learning for pattern recognition in magnetocardiography (MCG) that measures magnetic fields emitted by the electrophysiological activity of the heart is disclosed herein. Direct kernel methods are used to separate abnormal MCG heart patterns from normal ones. For unsupervised learning, Direct Kernel based Self-Organizing Maps are introduced. For supervised learning Direct Kernel Partial Least Squares and (Direct) Kernel Ridge Regression are used. These results are then compared with classical Support Vector Machines and Kernel Partial Least Squares. The hyper-parameters for these methods are tuned on a validation subset of the training data before testing. Also investigated is the most effective pre-processing, using local, vertical, horizontal and two-dimensional (global) Mahanalobis scaling, wavelet transforms, and variable selection by filtering. The results, similar for all three methods, were encouraging, exceeding the quality of classification achieved by the trained experts.Type: GrantFiled: July 1, 2004Date of Patent: June 22, 2010Assignee: CardioMag Imaging, Inc.Inventors: Karsten Sternickel, Boleslaw Szymanski, Mark Embrechts
-
Publication number: 20070167846Abstract: The use of machine learning for pattern recognition in magnetocardiography (MCG) that measures magnetic fields emitted by the electrophysiological activity of the heart is disclosed herein. Direct kernel methods are used to separate abnormal MCG heart patterns from normal ones. For unsupervised learning, Direct Kernel based Self-Organizing Maps are introduced. For supervised learning Direct Kernel Partial Least Squares and (Direct) Kernel Ridge Regression are used. These results are then compared with classical Support Vector Machines and Kernel Partial Least Squares. The hyper-parameters for these methods are tuned on a validation subset of the training data before testing. Also investigated is the most effective pre-processing, using local, vertical, horizontal and two-dimensional (global) Mahanalobis scaling, wavelet transforms, and variable selection by filtering. The results, similar for all three methods, were encouraging, exceeding the quality of classification achieved by the trained experts.Type: ApplicationFiled: July 1, 2004Publication date: July 19, 2007Applicant: CardioMag Imaging, Inc.Inventors: Karsten Sternickel, Boleslaw Szymanski, Mark Embrechts