Academic Journal
Προβολή σε EDS

New classifier architecture and training methodologies for lung nodule detection in chest radiographs and computed tomography

Λεπτομέρειες βιβλιογραφικής εγγραφής
Τίτλος: New classifier architecture and training methodologies for lung nodule detection in chest radiographs and computed tomography
Συγγραφείς: Narayanan, Barath Narayanan
Πηγή: Graduate Theses and Dissertations
Στοιχεία εκδότη: eCommons
Έτος έκδοσης: 2017
Συλλογή: University of Dayton: eCommons
Θεματικοί όροι: Diagnostic imaging Data processing, Pattern recognition systems Data processing, Lungs Biopsy Imaging, Lungs Diseases Imaging, Electrical Engineering, Medical Imaging, Biomedical Research, Computer aided detection, Computed tomography, Chest radiographs, Lung nodules, Slice Thickness
Περιγραφή: Early detection of pulmonary lung nodules plays a significant role in the diagnosis of lung cancer. Radiologists use Computed Tomography (CT) and Chest Radiographs (CRs) to detect such nodules. In this research, we propose various pattern recognition algorithms to enhance the classification performance of the Computer Aided Detection (CAD) system for lung nodule detection in both modalities. We propose a novel optimized method of feature selection for clustering that would aid the performance of the classifier. We make use of an independent CR database for training purposes. Testing is implemented on a publicly available database created by the Standard Digital Image Database Project Team of the Scientific Committee of the Japanese Society of Radiological Technology (JRST). The JRST database comprises 154 CRs containing one radiologist confirmed nodule in each. We make use of 107 CT scans from publicly available dataset created by Lung Image Database Consortium (LIDC) for this study. We compare the performance of the cluster-classifier architecture to a single aggregate classifier architecture. Overall, with a specificity of 3 false positives per case on an average, we show a classifier performance boost of 7.7% for CRs and 5.0% for CT scans when compared to single aggregate classifier architecture. Furthermore, we study the performance of a CAD system in CT scans as a function of slice thickness. We believe this study has implication for how CT is acquired, processed and stored. We make use of CT cases acquired at a thickness of 1.25mm from the publicly available Lung Nodule Analysis 2016 (LUNA16) dataset for this research. We study the CAD performance at a native thickness of 1.25mm and various other down-sampled stages. Our study indicates that CAD performance at 2.5mm is comparable to 1.25mm and is much better than at higher thicknesses. In addition, we propose and compare three different training methodologies for utilizing non-homogenous thickness training (i.e., composed of cases with different slice ...
Τύπος εγγράφου: text
Γλώσσα: unknown
Relation: https://ecommons.udayton.edu/graduate_theses/1335; http://rave.ohiolink.edu/etdc/view?acc_num=dayton1508237793168873
Διαθεσιμότητα: https://ecommons.udayton.edu/graduate_theses/1335
http://rave.ohiolink.edu/etdc/view?acc_num=dayton1508237793168873
Rights: Copyright © 2017, author
Αριθμός Καταχώρησης: edsbas.30AC9515
Βάση Δεδομένων: BASE
FullText Text:
  Availability: 0
CustomLinks:
  – Url: https://ecommons.udayton.edu/graduate_theses/1335#
    Name: EDS - BASE (ns324271)
    Category: fullText
    Text: View record from BASE
Header DbId: edsbas
DbLabel: BASE
An: edsbas.30AC9515
RelevancyScore: 795
AccessLevel: 3
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 794.807434082031
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: New classifier architecture and training methodologies for lung nodule detection in chest radiographs and computed tomography
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Narayanan%2C+Barath+Narayanan%22">Narayanan, Barath Narayanan</searchLink>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: Graduate Theses and Dissertations
– Name: Publisher
  Label: Publisher Information
  Group: PubInfo
  Data: eCommons
– Name: DatePubCY
  Label: Publication Year
  Group: Date
  Data: 2017
– Name: Subset
  Label: Collection
  Group: HoldingsInfo
  Data: University of Dayton: eCommons
– Name: Subject
  Label: Subject Terms
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Diagnostic+imaging+Data+processing%22">Diagnostic imaging Data processing</searchLink><br /><searchLink fieldCode="DE" term="%22Pattern+recognition+systems+Data+processing%22">Pattern recognition systems Data processing</searchLink><br /><searchLink fieldCode="DE" term="%22Lungs+Biopsy+Imaging%22">Lungs Biopsy Imaging</searchLink><br /><searchLink fieldCode="DE" term="%22Lungs+Diseases+Imaging%22">Lungs Diseases Imaging</searchLink><br /><searchLink fieldCode="DE" term="%22Electrical+Engineering%22">Electrical Engineering</searchLink><br /><searchLink fieldCode="DE" term="%22Medical+Imaging%22">Medical Imaging</searchLink><br /><searchLink fieldCode="DE" term="%22Biomedical+Research%22">Biomedical Research</searchLink><br /><searchLink fieldCode="DE" term="%22Computer+aided+detection%22">Computer aided detection</searchLink><br /><searchLink fieldCode="DE" term="%22Computed+tomography%22">Computed tomography</searchLink><br /><searchLink fieldCode="DE" term="%22Chest+radiographs%22">Chest radiographs</searchLink><br /><searchLink fieldCode="DE" term="%22Lung+nodules%22">Lung nodules</searchLink><br /><searchLink fieldCode="DE" term="%22Slice+Thickness%22">Slice Thickness</searchLink>
– Name: Abstract
  Label: Description
  Group: Ab
  Data: Early detection of pulmonary lung nodules plays a significant role in the diagnosis of lung cancer. Radiologists use Computed Tomography (CT) and Chest Radiographs (CRs) to detect such nodules. In this research, we propose various pattern recognition algorithms to enhance the classification performance of the Computer Aided Detection (CAD) system for lung nodule detection in both modalities. We propose a novel optimized method of feature selection for clustering that would aid the performance of the classifier. We make use of an independent CR database for training purposes. Testing is implemented on a publicly available database created by the Standard Digital Image Database Project Team of the Scientific Committee of the Japanese Society of Radiological Technology (JRST). The JRST database comprises 154 CRs containing one radiologist confirmed nodule in each. We make use of 107 CT scans from publicly available dataset created by Lung Image Database Consortium (LIDC) for this study. We compare the performance of the cluster-classifier architecture to a single aggregate classifier architecture. Overall, with a specificity of 3 false positives per case on an average, we show a classifier performance boost of 7.7% for CRs and 5.0% for CT scans when compared to single aggregate classifier architecture. Furthermore, we study the performance of a CAD system in CT scans as a function of slice thickness. We believe this study has implication for how CT is acquired, processed and stored. We make use of CT cases acquired at a thickness of 1.25mm from the publicly available Lung Nodule Analysis 2016 (LUNA16) dataset for this research. We study the CAD performance at a native thickness of 1.25mm and various other down-sampled stages. Our study indicates that CAD performance at 2.5mm is comparable to 1.25mm and is much better than at higher thicknesses. In addition, we propose and compare three different training methodologies for utilizing non-homogenous thickness training (i.e., composed of cases with different slice ...
– Name: TypeDocument
  Label: Document Type
  Group: TypDoc
  Data: text
– Name: Language
  Label: Language
  Group: Lang
  Data: unknown
– Name: NoteTitleSource
  Label: Relation
  Group: SrcInfo
  Data: https://ecommons.udayton.edu/graduate_theses/1335; http://rave.ohiolink.edu/etdc/view?acc_num=dayton1508237793168873
– Name: URL
  Label: Availability
  Group: URL
  Data: https://ecommons.udayton.edu/graduate_theses/1335<br />http://rave.ohiolink.edu/etdc/view?acc_num=dayton1508237793168873
– Name: Copyright
  Label: Rights
  Group: Cpyrght
  Data: Copyright © 2017, author
– Name: AN
  Label: Accession Number
  Group: ID
  Data: edsbas.30AC9515
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=edsbas&AN=edsbas.30AC9515
RecordInfo BibRecord:
  BibEntity:
    Languages:
      – Text: unknown
    Subjects:
      – SubjectFull: Diagnostic imaging Data processing
        Type: general
      – SubjectFull: Pattern recognition systems Data processing
        Type: general
      – SubjectFull: Lungs Biopsy Imaging
        Type: general
      – SubjectFull: Lungs Diseases Imaging
        Type: general
      – SubjectFull: Electrical Engineering
        Type: general
      – SubjectFull: Medical Imaging
        Type: general
      – SubjectFull: Biomedical Research
        Type: general
      – SubjectFull: Computer aided detection
        Type: general
      – SubjectFull: Computed tomography
        Type: general
      – SubjectFull: Chest radiographs
        Type: general
      – SubjectFull: Lung nodules
        Type: general
      – SubjectFull: Slice Thickness
        Type: general
    Titles:
      – TitleFull: New classifier architecture and training methodologies for lung nodule detection in chest radiographs and computed tomography
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Narayanan, Barath Narayanan
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 01
              Type: published
              Y: 2017
          Identifiers:
            – Type: issn-locals
              Value: edsbas
          Titles:
            – TitleFull: Graduate Theses and Dissertations
              Type: main
ResultId 1