Use of machine learning for classification of magneto cardiograms


Ontology type: sgo:Patent     


Patent Info

DATE

2010-06-22T00:00

AUTHORS

Karsten Sternickel , Boleslaw Szymanski , Mark Embrechts

ABSTRACT

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. Thus, a device and associated method for classifying cardiography data is disclosed, comprising applying a kernel transform to sensed data acquired from sensors sensing electromagnetic heart activity, resulting in transformed data, prior to classifying the transformed data using machine learning. More... »

Related SciGraph Publications

JSON-LD is the canonical representation for SciGraph data.

TIP: You can open this SciGraph record using an external JSON-LD service: JSON-LD Playground Google SDTT

[
  {
    "@context": "https://springernature.github.io/scigraph/jsonld/sgcontext.json", 
    "about": [
      {
        "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/2746", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "type": "DefinedTerm"
      }
    ], 
    "author": [
      {
        "name": "Karsten Sternickel", 
        "type": "Person"
      }, 
      {
        "name": "Boleslaw Szymanski", 
        "type": "Person"
      }, 
      {
        "name": "Mark Embrechts", 
        "type": "Person"
      }
    ], 
    "citation": [
      {
        "id": "sg:pub.10.1007/bf02345010", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1029760276", 
          "https://doi.org/10.1007/bf02345010"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/bf02345010", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1029760276", 
          "https://doi.org/10.1007/bf02345010"
        ], 
        "type": "CreativeWork"
      }
    ], 
    "datePublished": "2010-06-22T00:00", 
    "description": "

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. Thus, a device and associated method for classifying cardiography data is disclosed, comprising applying a kernel transform to sensed data acquired from sensors sensing electromagnetic heart activity, resulting in transformed data, prior to classifying the transformed data using machine learning.

", "id": "sg:patent.US-7742806-B2", "keywords": [ "machine", "classification", "pattern recognition", "mcg", "magnetic field", "electrophysiological activity", "heart", "kernel method", "pattern", "normal one", "kernel", "map", "partial least square", "regression", "Support Vector Machine", "parameter", "method", "validation", "training data", "pre-processing", "scaling", "wavelet", "variable selection", "trained expert", "Equipment and Supply", "associated method", "classifying", "sensed data", "sensor" ], "name": "Use of machine learning for classification of magneto cardiograms", "sameAs": [ "https://app.dimensions.ai/details/patent/US-7742806-B2" ], "sdDataset": "patents", "sdDatePublished": "2019-04-18T10:05", "sdLicense": "https://scigraph.springernature.com/explorer/license/", "sdPublisher": { "name": "Springer Nature - SN SciGraph project", "type": "Organization" }, "sdSource": "s3://com-uberresearch-data-patents-target-20190320-rc/data/sn-export/402f166718b70575fb5d4ffe01f064d1/0000100128-0000352499/json_export_00076.jsonl", "type": "Patent" } ]
 

Download the RDF metadata as:  json-ld nt turtle xml License info

HOW TO GET THIS DATA PROGRAMMATICALLY:

JSON-LD is a popular format for linked data which is fully compatible with JSON.

curl -H 'Accept: application/ld+json' 'https://scigraph.springernature.com/patent.US-7742806-B2'

N-Triples is a line-based linked data format ideal for batch operations.

curl -H 'Accept: application/n-triples' 'https://scigraph.springernature.com/patent.US-7742806-B2'

Turtle is a human-readable linked data format.

curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/patent.US-7742806-B2'

RDF/XML is a standard XML format for linked data.

curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/patent.US-7742806-B2'


 

This table displays all metadata directly associated to this object as RDF triples.

61 TRIPLES      14 PREDICATES      43 URIs      37 LITERALS      2 BLANK NODES

Subject Predicate Object
1 sg:patent.US-7742806-B2 schema:about anzsrc-for:2746
2 schema:author N23c8366cf280496bb5740a457ee35850
3 schema:citation sg:pub.10.1007/bf02345010
4 schema:datePublished 2010-06-22T00:00
5 schema:description <p num="p-0001">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. Thus, a device and associated method for classifying cardiography data is disclosed, comprising applying a kernel transform to sensed data acquired from sensors sensing electromagnetic heart activity, resulting in transformed data, prior to classifying the transformed data using machine learning.</p>
6 schema:keywords Equipment and Supply
7 Support Vector Machine
8 associated method
9 classification
10 classifying
11 electrophysiological activity
12 heart
13 kernel
14 kernel method
15 machine
16 magnetic field
17 map
18 mcg
19 method
20 normal one
21 parameter
22 partial least square
23 pattern
24 pattern recognition
25 pre-processing
26 regression
27 scaling
28 sensed data
29 sensor
30 trained expert
31 training data
32 validation
33 variable selection
34 wavelet
35 schema:name Use of machine learning for classification of magneto cardiograms
36 schema:sameAs https://app.dimensions.ai/details/patent/US-7742806-B2
37 schema:sdDatePublished 2019-04-18T10:05
38 schema:sdLicense https://scigraph.springernature.com/explorer/license/
39 schema:sdPublisher N64b0d55043984e54834cb67f8513afc7
40 sgo:license sg:explorer/license/
41 sgo:sdDataset patents
42 rdf:type sgo:Patent
43 N23c8366cf280496bb5740a457ee35850 rdf:first N8c57b9cea9c64bf2ba5abad57c723df9
44 rdf:rest N240b4259326f48188424ba0d176f0dc1
45 N240b4259326f48188424ba0d176f0dc1 rdf:first N416ae7a0d1724990b6a7a45b36dc226c
46 rdf:rest N5c0a681ede3c4f53a0efea9f356abc79
47 N416ae7a0d1724990b6a7a45b36dc226c schema:name Boleslaw Szymanski
48 rdf:type schema:Person
49 N5c0a681ede3c4f53a0efea9f356abc79 rdf:first Nbe879463d8be42028fa5a6e964dc2af3
50 rdf:rest rdf:nil
51 N64b0d55043984e54834cb67f8513afc7 schema:name Springer Nature - SN SciGraph project
52 rdf:type schema:Organization
53 N8c57b9cea9c64bf2ba5abad57c723df9 schema:name Karsten Sternickel
54 rdf:type schema:Person
55 Nbe879463d8be42028fa5a6e964dc2af3 schema:name Mark Embrechts
56 rdf:type schema:Person
57 anzsrc-for:2746 schema:inDefinedTermSet anzsrc-for:
58 rdf:type schema:DefinedTerm
59 sg:pub.10.1007/bf02345010 schema:sameAs https://app.dimensions.ai/details/publication/pub.1029760276
60 https://doi.org/10.1007/bf02345010
61 rdf:type schema:CreativeWork
 




Preview window. Press ESC to close (or click here)


...