Machine Learning For Classification Of Magneto Cardiograms


Ontology type: sgo:Patent     


Patent Info

DATE

2005-01-13T00:00

AUTHORS

STERNICKEL, KARSTEN , EMBRECHTS, MARK, J. , SZYMANSKI, BOLESLAW, K.

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

  • 2000-10. Better decisions through science. in SCIENTIFIC AMERICAN
  • 2000-06. Statistical Learning Theory: A Primer in INTERNATIONAL JOURNAL OF COMPUTER VISION
  • 1999-06. Least Squares Support Vector Machine Classifiers in NEURAL PROCESSING LETTERS
  • 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": "STERNICKEL, KARSTEN", 
            "type": "Person"
          }, 
          {
            "name": "EMBRECHTS, MARK, J.", 
            "type": "Person"
          }, 
          {
            "name": "SZYMANSKI, BOLESLAW, K.", 
            "type": "Person"
          }
        ], 
        "citation": [
          {
            "id": "https://doi.org/10.1016/s0893-6080(05)80056-5", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1014360644"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1016/s0893-6080(05)80056-5", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1014360644"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1023/a:1008110632619", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1018456915", 
              "https://doi.org/10.1023/a:1008110632619"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1016/s0169-7439(97)00010-5", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1022390679"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1023/a:1018628609742", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1025353016", 
              "https://doi.org/10.1023/a:1018628609742"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1016/s0169-7439(97)00027-0", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1040170952"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1038/scientificamerican1000-82", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1056601272", 
              "https://doi.org/10.1038/scientificamerican1000-82"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1109/72.846738", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1061219427"
            ], 
            "type": "CreativeWork"
          }
        ], 
        "datePublished": "2005-01-13T00:00", 
        "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.WO-2005002313-A2", "keywords": [ "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", "machine" ], "name": "MACHINE LEARNING FOR CLASSIFICATION OF MAGNETO CARDIOGRAMS", "sameAs": [ "https://app.dimensions.ai/details/patent/WO-2005002313-A2" ], "sdDataset": "patents", "sdDatePublished": "2019-04-18T10:30", "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_03485.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.WO-2005002313-A2'

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

    curl -H 'Accept: application/n-triples' 'https://scigraph.springernature.com/patent.WO-2005002313-A2'

    Turtle is a human-readable linked data format.

    curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/patent.WO-2005002313-A2'

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

    curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/patent.WO-2005002313-A2'


     

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

    81 TRIPLES      14 PREDICATES      49 URIs      37 LITERALS      2 BLANK NODES

    Subject Predicate Object
    1 sg:patent.WO-2005002313-A2 schema:about anzsrc-for:2746
    2 schema:author N18f447dffb3f4df18ebfa40f6624708e
    3 schema:citation sg:pub.10.1023/a:1008110632619
    4 sg:pub.10.1023/a:1018628609742
    5 sg:pub.10.1038/scientificamerican1000-82
    6 https://doi.org/10.1016/s0169-7439(97)00010-5
    7 https://doi.org/10.1016/s0169-7439(97)00027-0
    8 https://doi.org/10.1016/s0893-6080(05)80056-5
    9 https://doi.org/10.1109/72.846738
    10 schema:datePublished 2005-01-13T00:00
    11 schema:description <p>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>
    12 schema:keywords Equipment and Supply
    13 Support Vector Machine
    14 associated method
    15 classification
    16 classifying
    17 electrophysiological activity
    18 heart
    19 kernel
    20 kernel method
    21 machine
    22 magnetic field
    23 map
    24 mcg
    25 method
    26 normal one
    27 parameter
    28 partial least square
    29 pattern
    30 pattern recognition
    31 pre-processing
    32 regression
    33 scaling
    34 sensed data
    35 sensor
    36 trained expert
    37 training data
    38 validation
    39 variable selection
    40 wavelet
    41 schema:name MACHINE LEARNING FOR CLASSIFICATION OF MAGNETO CARDIOGRAMS
    42 schema:sameAs https://app.dimensions.ai/details/patent/WO-2005002313-A2
    43 schema:sdDatePublished 2019-04-18T10:30
    44 schema:sdLicense https://scigraph.springernature.com/explorer/license/
    45 schema:sdPublisher N3aca26eb9a274b629d7977313d6d814f
    46 sgo:license sg:explorer/license/
    47 sgo:sdDataset patents
    48 rdf:type sgo:Patent
    49 N0fd79ff4e7c9429792351dd2df415503 rdf:first N9adb1e8e2db84657a25c98a79ecd3017
    50 rdf:rest N3fe701137f7e43a3a8f28ef982c956bb
    51 N18f447dffb3f4df18ebfa40f6624708e rdf:first N2c692ed8fc8a4a5abc640d2c61217743
    52 rdf:rest N0fd79ff4e7c9429792351dd2df415503
    53 N2c692ed8fc8a4a5abc640d2c61217743 schema:name STERNICKEL, KARSTEN
    54 rdf:type schema:Person
    55 N3aca26eb9a274b629d7977313d6d814f schema:name Springer Nature - SN SciGraph project
    56 rdf:type schema:Organization
    57 N3fe701137f7e43a3a8f28ef982c956bb rdf:first N47296c3ddb8e411c98fd02cf176b0c99
    58 rdf:rest rdf:nil
    59 N47296c3ddb8e411c98fd02cf176b0c99 schema:name SZYMANSKI, BOLESLAW, K.
    60 rdf:type schema:Person
    61 N9adb1e8e2db84657a25c98a79ecd3017 schema:name EMBRECHTS, MARK, J.
    62 rdf:type schema:Person
    63 anzsrc-for:2746 schema:inDefinedTermSet anzsrc-for:
    64 rdf:type schema:DefinedTerm
    65 sg:pub.10.1023/a:1008110632619 schema:sameAs https://app.dimensions.ai/details/publication/pub.1018456915
    66 https://doi.org/10.1023/a:1008110632619
    67 rdf:type schema:CreativeWork
    68 sg:pub.10.1023/a:1018628609742 schema:sameAs https://app.dimensions.ai/details/publication/pub.1025353016
    69 https://doi.org/10.1023/a:1018628609742
    70 rdf:type schema:CreativeWork
    71 sg:pub.10.1038/scientificamerican1000-82 schema:sameAs https://app.dimensions.ai/details/publication/pub.1056601272
    72 https://doi.org/10.1038/scientificamerican1000-82
    73 rdf:type schema:CreativeWork
    74 https://doi.org/10.1016/s0169-7439(97)00010-5 schema:sameAs https://app.dimensions.ai/details/publication/pub.1022390679
    75 rdf:type schema:CreativeWork
    76 https://doi.org/10.1016/s0169-7439(97)00027-0 schema:sameAs https://app.dimensions.ai/details/publication/pub.1040170952
    77 rdf:type schema:CreativeWork
    78 https://doi.org/10.1016/s0893-6080(05)80056-5 schema:sameAs https://app.dimensions.ai/details/publication/pub.1014360644
    79 rdf:type schema:CreativeWork
    80 https://doi.org/10.1109/72.846738 schema:sameAs https://app.dimensions.ai/details/publication/pub.1061219427
    81 rdf:type schema:CreativeWork
     




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


    ...