View in EDS HTML Full Text PDF Full Text

Fully-automated deep learning-based flow quantification of 2D CINE phase contrast MRI.

Saved in:
Bibliographic Details
Title: Fully-automated deep learning-based flow quantification of 2D CINE phase contrast MRI.
Authors: Pradella, Maurice1,2 maurice.pradella@northwestern.edu, Scott, Michael B.1, Omer, Muhammad3, Hill, Seth K.1, Lockhart, Lisette3, Yi, Xin3, Amir-Khalili, Alborz3, Sojoudi, Alireza3, Allen, Bradley D.1, Avery, Ryan1, Markl, Michael1
Source: European Radiology. Mar2023, Vol. 33 Issue 3, p1707-1718. 12p. 1 Black and White Photograph, 1 Diagram, 3 Charts, 4 Graphs.
Subjects: Deep learning, Magnetic resonance imaging, Pulmonary artery, Blood flow, Medical statistics
Abstract: Objectives: Time-resolved, 2D-phase-contrast MRI (2D-CINE-PC-MRI) enables in vivo blood flow analysis. However, accurate vessel contour delineation (VCD) is required to achieve reliable results. We sought to evaluate manual analysis (MA) compared to the performance of a deep learning (DL) application for fully-automated VCD and flow quantification and corrected semi-automated analysis (corSAA). Methods: We included 97 consecutive patients (age = 52.9 ± 16 years, 41 female) with 2D-CINE-PC-MRI imaging on 1.5T MRI systems at sinotubular junction (STJ), and 28/97 also received 2D-CINE-PC at main pulmonary artery (PA). A cardiovascular radiologist performed MA (reference) and corSAA (built-in tool) in commercial software for all cardiac time frames (median: 20, total contours per analysis: 2358 STJ, 680 PA). DL-analysis automatically performed VCD, followed by net flow (NF) and peak velocity (PV) quantification. Contours were compared using Dice similarity coefficients (DSC). Discrepant cases (> ± 10 mL or > ± 10 cm/s) were reviewed in detail. Results: DL was successfully applied to 97% (121/125) of the 2D-CINE-PC-MRI series (STJ: 95/97, 98%, PA: 26/28, 93%). Compared to MA, mean DSC were 0.91 ± 0.02 (DL), 0.94 ± 0.02 (corSAA) at STJ, and 0.85 ± 0.08 (DL), 0.93 ± 0.02 (corSAA) at PA; this indicated good to excellent DL-performance. Flow quantification revealed similar NF at STJ (p = 0.48) and PA (p > 0.05) between methods while PV assessment was significantly different (STJ: p < 0.001, PA: p = 0.04). A detailed review showed noisy voxels in MA and corSAA impacted PV results. Overall, DL analysis compared to human assessments was accurate in 113/121 (93.4%) cases. Conclusions: Fully-automated DL-analysis of 2D-CINE-PC-MRI provided flow quantification at STJ and PA at expert level in > 93% of cases with results being available instantaneously. Key Points: • Deep learning performed flow quantification on clinical 2D-CINE-PC series at the sinotubular junction and pulmonary artery at the expert level in > 93% of cases. • Location detection and contouring of the vessel boundaries were performed fully-automatic with results being available instantaneously compared to human assessments which approximately takes three minutes per location. • The evaluated tool indicates usability in daily practice. [ABSTRACT FROM AUTHOR]
Copyright of European Radiology is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database: Engineering Source
Full text is not displayed to guests.
FullText Links:
  – Type: pdflink
Text:
  Availability: 1
Header DbId: egs
DbLabel: Engineering Source
An: 161963595
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Fully-automated deep learning-based flow quantification of 2D CINE phase contrast MRI.
– Name: Author
  Label: Authors
  Group: Au
  Data: &lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Pradella%2C+Maurice%22&quot;&gt;Pradella, Maurice&lt;/searchLink&gt;&lt;relatesTo&gt;1,2&lt;/relatesTo&gt;&lt;i&gt; maurice.pradella@northwestern.edu&lt;/i&gt;&lt;br /&gt;&lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Scott%2C+Michael+B%2E%22&quot;&gt;Scott, Michael B.&lt;/searchLink&gt;&lt;relatesTo&gt;1&lt;/relatesTo&gt;&lt;br /&gt;&lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Omer%2C+Muhammad%22&quot;&gt;Omer, Muhammad&lt;/searchLink&gt;&lt;relatesTo&gt;3&lt;/relatesTo&gt;&lt;br /&gt;&lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Hill%2C+Seth+K%2E%22&quot;&gt;Hill, Seth K.&lt;/searchLink&gt;&lt;relatesTo&gt;1&lt;/relatesTo&gt;&lt;br /&gt;&lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Lockhart%2C+Lisette%22&quot;&gt;Lockhart, Lisette&lt;/searchLink&gt;&lt;relatesTo&gt;3&lt;/relatesTo&gt;&lt;br /&gt;&lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Yi%2C+Xin%22&quot;&gt;Yi, Xin&lt;/searchLink&gt;&lt;relatesTo&gt;3&lt;/relatesTo&gt;&lt;br /&gt;&lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Amir-Khalili%2C+Alborz%22&quot;&gt;Amir-Khalili, Alborz&lt;/searchLink&gt;&lt;relatesTo&gt;3&lt;/relatesTo&gt;&lt;br /&gt;&lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Sojoudi%2C+Alireza%22&quot;&gt;Sojoudi, Alireza&lt;/searchLink&gt;&lt;relatesTo&gt;3&lt;/relatesTo&gt;&lt;br /&gt;&lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Allen%2C+Bradley+D%2E%22&quot;&gt;Allen, Bradley D.&lt;/searchLink&gt;&lt;relatesTo&gt;1&lt;/relatesTo&gt;&lt;br /&gt;&lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Avery%2C+Ryan%22&quot;&gt;Avery, Ryan&lt;/searchLink&gt;&lt;relatesTo&gt;1&lt;/relatesTo&gt;&lt;br /&gt;&lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Markl%2C+Michael%22&quot;&gt;Markl, Michael&lt;/searchLink&gt;&lt;relatesTo&gt;1&lt;/relatesTo&gt;
– Name: TitleSource
  Label: Source
  Group: Src
  Data: &lt;searchLink fieldCode=&quot;JN&quot; term=&quot;%22European+Radiology%22&quot;&gt;European Radiology&lt;/searchLink&gt;. Mar2023, Vol. 33 Issue 3, p1707-1718. 12p. 1 Black and White Photograph, 1 Diagram, 3 Charts, 4 Graphs.
– Name: Subject
  Label: Subjects
  Group: Su
  Data: &lt;searchLink fieldCode=&quot;DE&quot; term=&quot;%22Deep+learning%22&quot;&gt;Deep learning&lt;/searchLink&gt;&lt;br /&gt;&lt;searchLink fieldCode=&quot;DE&quot; term=&quot;%22Magnetic+resonance+imaging%22&quot;&gt;Magnetic resonance imaging&lt;/searchLink&gt;&lt;br /&gt;&lt;searchLink fieldCode=&quot;DE&quot; term=&quot;%22Pulmonary+artery%22&quot;&gt;Pulmonary artery&lt;/searchLink&gt;&lt;br /&gt;&lt;searchLink fieldCode=&quot;DE&quot; term=&quot;%22Blood+flow%22&quot;&gt;Blood flow&lt;/searchLink&gt;&lt;br /&gt;&lt;searchLink fieldCode=&quot;DE&quot; term=&quot;%22Medical+statistics%22&quot;&gt;Medical statistics&lt;/searchLink&gt;
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: Objectives: Time-resolved, 2D-phase-contrast MRI (2D-CINE-PC-MRI) enables in vivo blood flow analysis. However, accurate vessel contour delineation (VCD) is required to achieve reliable results. We sought to evaluate manual analysis (MA) compared to the performance of a deep learning (DL) application for fully-automated VCD and flow quantification and corrected semi-automated analysis (corSAA). Methods: We included 97 consecutive patients (age = 52.9 &#177; 16 years, 41 female) with 2D-CINE-PC-MRI imaging on 1.5T MRI systems at sinotubular junction (STJ), and 28/97 also received 2D-CINE-PC at main pulmonary artery (PA). A cardiovascular radiologist performed MA (reference) and corSAA (built-in tool) in commercial software for all cardiac time frames (median: 20, total contours per analysis: 2358 STJ, 680 PA). DL-analysis automatically performed VCD, followed by net flow (NF) and peak velocity (PV) quantification. Contours were compared using Dice similarity coefficients (DSC). Discrepant cases (&gt; &#177; 10 mL or &gt; &#177; 10 cm/s) were reviewed in detail. Results: DL was successfully applied to 97% (121/125) of the 2D-CINE-PC-MRI series (STJ: 95/97, 98%, PA: 26/28, 93%). Compared to MA, mean DSC were 0.91 &#177; 0.02 (DL), 0.94 &#177; 0.02 (corSAA) at STJ, and 0.85 &#177; 0.08 (DL), 0.93 &#177; 0.02 (corSAA) at PA; this indicated good to excellent DL-performance. Flow quantification revealed similar NF at STJ (p = 0.48) and PA (p &gt; 0.05) between methods while PV assessment was significantly different (STJ: p &lt; 0.001, PA: p = 0.04). A detailed review showed noisy voxels in MA and corSAA impacted PV results. Overall, DL analysis compared to human assessments was accurate in 113/121 (93.4%) cases. Conclusions: Fully-automated DL-analysis of 2D-CINE-PC-MRI provided flow quantification at STJ and PA at expert level in &gt; 93% of cases with results being available instantaneously. Key Points: • Deep learning performed flow quantification on clinical 2D-CINE-PC series at the sinotubular junction and pulmonary artery at the expert level in &gt; 93% of cases. • Location detection and contouring of the vessel boundaries were performed fully-automatic with results being available instantaneously compared to human assessments which approximately takes three minutes per location. • The evaluated tool indicates usability in daily practice. [ABSTRACT FROM AUTHOR]
– Name: AbstractSuppliedCopyright
  Label:
  Group: Ab
  Data: &lt;i&gt;Copyright of European Radiology is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder&#39;s express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract.&lt;/i&gt; (Copyright applies to all Abstracts.)
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=egs&AN=161963595
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1007/s00330-022-09179-3
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 12
        StartPage: 1707
    Subjects:
      – SubjectFull: Deep learning
        Type: general
      – SubjectFull: Magnetic resonance imaging
        Type: general
      – SubjectFull: Pulmonary artery
        Type: general
      – SubjectFull: Blood flow
        Type: general
      – SubjectFull: Medical statistics
        Type: general
    Titles:
      – TitleFull: Fully-automated deep learning-based flow quantification of 2D CINE phase contrast MRI.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Pradella, Maurice
      – PersonEntity:
          Name:
            NameFull: Scott, Michael B.
      – PersonEntity:
          Name:
            NameFull: Omer, Muhammad
      – PersonEntity:
          Name:
            NameFull: Hill, Seth K.
      – PersonEntity:
          Name:
            NameFull: Lockhart, Lisette
      – PersonEntity:
          Name:
            NameFull: Yi, Xin
      – PersonEntity:
          Name:
            NameFull: Amir-Khalili, Alborz
      – PersonEntity:
          Name:
            NameFull: Sojoudi, Alireza
      – PersonEntity:
          Name:
            NameFull: Allen, Bradley D.
      – PersonEntity:
          Name:
            NameFull: Avery, Ryan
      – PersonEntity:
          Name:
            NameFull: Markl, Michael
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 03
              Text: Mar2023
              Type: published
              Y: 2023
          Identifiers:
            – Type: issn-print
              Value: 09387994
          Numbering:
            – Type: volume
              Value: 33
            – Type: issue
              Value: 3
          Titles:
            – TitleFull: European Radiology
              Type: main
ResultId 1