View in EDS

Temporal fusion and heatmap regression for precise left ventricular parameter measurement in echocardiographic parasternal long‐axis videos.

Saved in:
Bibliographic Details
Title: Temporal fusion and heatmap regression for precise left ventricular parameter measurement in echocardiographic parasternal long‐axis videos.
Authors: Chen, Yidi1 (AUTHOR), Shan, Chunjie1 (AUTHOR), Qi, Zhanru2,3,4 (AUTHOR), Shi, Zhongqing2,3,4 (AUTHOR), Guo, Guanjun2,3,4 (AUTHOR), Wang, Xiaoxian2,3,4 (AUTHOR), Chen, Hui2,3,4 (AUTHOR), Chen, Fen2,3,4 (AUTHOR), Fang, Aijuan2,3,4 (AUTHOR), Cheng, Hanlin1 (AUTHOR), Weng, Hexiang1 (AUTHOR), Luo, Shouhua1 (AUTHOR) luoshouhua@seu.edu.cn, Yao, Jing2,3,4 (AUTHOR) jingyao@nju.edu.cn, Qian, Sunnan5 (AUTHOR) 610921124@qq.com
Source: Medical Physics. May2026, Vol. 53 Issue 5, p1-20. 20p.
Subjects: Echocardiography, Temporal integration, Image segmentation, Computer vision
Abstract: Background: Left ventricular geometric parameters are critical for diagnosing and prognosticating cardiovascular diseases. Currently, most measurement techniques rely on two‐dimensional transthoracic echocardiography (TTE), where an end‐diastolic (ED) frame from the parasternal long‐axis (PLAX) view is selected, and key points on the interventricular septum (IVS), left ventricular internal dimension (LVID), and left ventricular posterior wall (LVPW) are identified. However, using a single frame often fails to capture the entire structure of the IVS and LVPW, especially when complex anatomical details or blurred edges are present, leading to positional shifts or loss of key points and, hence, considerable measurement errors. Purpose: In this study, we propose an automatic method for measuring left ventricular structural parameters based on echocardiographic PLAX‐view videos. The approach focuses on the ED frame along with the immediately preceding and following frames. Methods: We developed an ultrasound video analysis model that integrates temporally distributed and incomplete structural information to reconstruct the complete anatomies of the IVS and LVPW. The model combines a segmentation branch for precise boundary localization with a heatmap regression branch for chamber centerline and LVID measurement line estimation, enforcing perpendicular anatomical constraints. The dataset comprised 400 PLAX echocardiographic videos from 400 distinct patients, acquired at 56 fps. The data were divided into training and validation sets in a ratio of 8:2. The proposed model was compared with U‐Net, U‐Net++, DeepLabV3, SegFormer, and TransUNet for segmentation, and HRNet and ViTPose for keypoint detection. Evaluation metrics included mIoU, Dice similarity coefficient (DSC), Hausdorff distance (HD), and average precision (AP50$AP_{50}$, AP75$AP_{75}$, mAP). Statistical significance was assessed using paired t‐tests with a significance threshold of p<0.05$p < 0.05$, and multiple comparisons were corrected using the Benjamini–Hochberg (BH) procedure. Results: Our results demonstrate robust performance improvements over existing benchmarks. In the segmentation task, our method achieved a mean intersection over union (mIoU) of 83.22% (DSC 0.856, HD 10.174). Statistical analysis demonstrated that this performance is significantly superior to classic models like U‐Net (padj<0.01$p_{adj} < 0.01$), showing a positive small‐to‐medium effect size (d=0.326$d = 0.326$). In the keypoint detection task, our approach achieved an mAP of 0.698 (AP50${\rm AP}_{50}$ = 0.965), significantly outperforming the DeepLabV3 baseline (padj<0.001$p_{adj} < 0.001$) with a positive medium‐to‐large effect size (d=0.641$d = 0.641$). Moreover, against strong baselines such as ViTPose, our method maintained a statistically significant advantage (padj<0.05$p_{adj} < 0.05$) with a positive small effect size (d=0.198$d = 0.198$). Conclusions: These outcomes demonstrate the method's robust performance in accurately delineating structural boundaries and reducing measurement errors. [ABSTRACT FROM AUTHOR]
Copyright of Medical Physics is the property of Wiley-Blackwell 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
FullText Text:
  Availability: 0
Header DbId: egs
DbLabel: Engineering Source
An: 194450980
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Temporal fusion and heatmap regression for precise left ventricular parameter measurement in echocardiographic parasternal long‐axis videos.
– Name: Author
  Label: Authors
  Group: Au
  Data: &lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Chen%2C+Yidi%22&quot;&gt;Chen, Yidi&lt;/searchLink&gt;&lt;relatesTo&gt;1&lt;/relatesTo&gt; (AUTHOR)&lt;br /&gt;&lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Shan%2C+Chunjie%22&quot;&gt;Shan, Chunjie&lt;/searchLink&gt;&lt;relatesTo&gt;1&lt;/relatesTo&gt; (AUTHOR)&lt;br /&gt;&lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Qi%2C+Zhanru%22&quot;&gt;Qi, Zhanru&lt;/searchLink&gt;&lt;relatesTo&gt;2,3,4&lt;/relatesTo&gt; (AUTHOR)&lt;br /&gt;&lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Shi%2C+Zhongqing%22&quot;&gt;Shi, Zhongqing&lt;/searchLink&gt;&lt;relatesTo&gt;2,3,4&lt;/relatesTo&gt; (AUTHOR)&lt;br /&gt;&lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Guo%2C+Guanjun%22&quot;&gt;Guo, Guanjun&lt;/searchLink&gt;&lt;relatesTo&gt;2,3,4&lt;/relatesTo&gt; (AUTHOR)&lt;br /&gt;&lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Wang%2C+Xiaoxian%22&quot;&gt;Wang, Xiaoxian&lt;/searchLink&gt;&lt;relatesTo&gt;2,3,4&lt;/relatesTo&gt; (AUTHOR)&lt;br /&gt;&lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Chen%2C+Hui%22&quot;&gt;Chen, Hui&lt;/searchLink&gt;&lt;relatesTo&gt;2,3,4&lt;/relatesTo&gt; (AUTHOR)&lt;br /&gt;&lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Chen%2C+Fen%22&quot;&gt;Chen, Fen&lt;/searchLink&gt;&lt;relatesTo&gt;2,3,4&lt;/relatesTo&gt; (AUTHOR)&lt;br /&gt;&lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Fang%2C+Aijuan%22&quot;&gt;Fang, Aijuan&lt;/searchLink&gt;&lt;relatesTo&gt;2,3,4&lt;/relatesTo&gt; (AUTHOR)&lt;br /&gt;&lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Cheng%2C+Hanlin%22&quot;&gt;Cheng, Hanlin&lt;/searchLink&gt;&lt;relatesTo&gt;1&lt;/relatesTo&gt; (AUTHOR)&lt;br /&gt;&lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Weng%2C+Hexiang%22&quot;&gt;Weng, Hexiang&lt;/searchLink&gt;&lt;relatesTo&gt;1&lt;/relatesTo&gt; (AUTHOR)&lt;br /&gt;&lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Luo%2C+Shouhua%22&quot;&gt;Luo, Shouhua&lt;/searchLink&gt;&lt;relatesTo&gt;1&lt;/relatesTo&gt; (AUTHOR)&lt;i&gt; luoshouhua@seu.edu.cn&lt;/i&gt;&lt;br /&gt;&lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Yao%2C+Jing%22&quot;&gt;Yao, Jing&lt;/searchLink&gt;&lt;relatesTo&gt;2,3,4&lt;/relatesTo&gt; (AUTHOR)&lt;i&gt; jingyao@nju.edu.cn&lt;/i&gt;&lt;br /&gt;&lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Qian%2C+Sunnan%22&quot;&gt;Qian, Sunnan&lt;/searchLink&gt;&lt;relatesTo&gt;5&lt;/relatesTo&gt; (AUTHOR)&lt;i&gt; 610921124@qq.com&lt;/i&gt;
– Name: TitleSource
  Label: Source
  Group: Src
  Data: &lt;searchLink fieldCode=&quot;JN&quot; term=&quot;%22Medical+Physics%22&quot;&gt;Medical Physics&lt;/searchLink&gt;. May2026, Vol. 53 Issue 5, p1-20. 20p.
– Name: Subject
  Label: Subjects
  Group: Su
  Data: &lt;searchLink fieldCode=&quot;DE&quot; term=&quot;%22Echocardiography%22&quot;&gt;Echocardiography&lt;/searchLink&gt;&lt;br /&gt;&lt;searchLink fieldCode=&quot;DE&quot; term=&quot;%22Temporal+integration%22&quot;&gt;Temporal integration&lt;/searchLink&gt;&lt;br /&gt;&lt;searchLink fieldCode=&quot;DE&quot; term=&quot;%22Image+segmentation%22&quot;&gt;Image segmentation&lt;/searchLink&gt;&lt;br /&gt;&lt;searchLink fieldCode=&quot;DE&quot; term=&quot;%22Computer+vision%22&quot;&gt;Computer vision&lt;/searchLink&gt;
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: Background: Left ventricular geometric parameters are critical for diagnosing and prognosticating cardiovascular diseases. Currently, most measurement techniques rely on two‐dimensional transthoracic echocardiography (TTE), where an end‐diastolic (ED) frame from the parasternal long‐axis (PLAX) view is selected, and key points on the interventricular septum (IVS), left ventricular internal dimension (LVID), and left ventricular posterior wall (LVPW) are identified. However, using a single frame often fails to capture the entire structure of the IVS and LVPW, especially when complex anatomical details or blurred edges are present, leading to positional shifts or loss of key points and, hence, considerable measurement errors. Purpose: In this study, we propose an automatic method for measuring left ventricular structural parameters based on echocardiographic PLAX‐view videos. The approach focuses on the ED frame along with the immediately preceding and following frames. Methods: We developed an ultrasound video analysis model that integrates temporally distributed and incomplete structural information to reconstruct the complete anatomies of the IVS and LVPW. The model combines a segmentation branch for precise boundary localization with a heatmap regression branch for chamber centerline and LVID measurement line estimation, enforcing perpendicular anatomical constraints. The dataset comprised 400 PLAX echocardiographic videos from 400 distinct patients, acquired at 56 fps. The data were divided into training and validation sets in a ratio of 8:2. The proposed model was compared with U‐Net, U‐Net++, DeepLabV3, SegFormer, and TransUNet for segmentation, and HRNet and ViTPose for keypoint detection. Evaluation metrics included mIoU, Dice similarity coefficient (DSC), Hausdorff distance (HD), and average precision (AP50$AP_{50}$, AP75$AP_{75}$, mAP). Statistical significance was assessed using paired t‐tests with a significance threshold of p&lt;0.05$p &lt; 0.05$, and multiple comparisons were corrected using the Benjamini–Hochberg (BH) procedure. Results: Our results demonstrate robust performance improvements over existing benchmarks. In the segmentation task, our method achieved a mean intersection over union (mIoU) of 83.22% (DSC 0.856, HD 10.174). Statistical analysis demonstrated that this performance is significantly superior to classic models like U‐Net (padj&lt;0.01$p_{adj} &lt; 0.01$), showing a positive small‐to‐medium effect size (d=0.326$d = 0.326$). In the keypoint detection task, our approach achieved an mAP of 0.698 (AP50${\rm AP}_{50}$ = 0.965), significantly outperforming the DeepLabV3 baseline (padj&lt;0.001$p_{adj} &lt; 0.001$) with a positive medium‐to‐large effect size (d=0.641$d = 0.641$). Moreover, against strong baselines such as ViTPose, our method maintained a statistically significant advantage (padj&lt;0.05$p_{adj} &lt; 0.05$) with a positive small effect size (d=0.198$d = 0.198$). Conclusions: These outcomes demonstrate the method&#39;s robust performance in accurately delineating structural boundaries and reducing measurement errors. [ABSTRACT FROM AUTHOR]
– Name: AbstractSuppliedCopyright
  Label:
  Group: Ab
  Data: &lt;i&gt;Copyright of Medical Physics is the property of Wiley-Blackwell 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=194450980
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1002/mp.70491
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 20
        StartPage: 1
    Subjects:
      – SubjectFull: Echocardiography
        Type: general
      – SubjectFull: Temporal integration
        Type: general
      – SubjectFull: Image segmentation
        Type: general
      – SubjectFull: Computer vision
        Type: general
    Titles:
      – TitleFull: Temporal fusion and heatmap regression for precise left ventricular parameter measurement in echocardiographic parasternal long‐axis videos.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Chen, Yidi
      – PersonEntity:
          Name:
            NameFull: Shan, Chunjie
      – PersonEntity:
          Name:
            NameFull: Qi, Zhanru
      – PersonEntity:
          Name:
            NameFull: Shi, Zhongqing
      – PersonEntity:
          Name:
            NameFull: Guo, Guanjun
      – PersonEntity:
          Name:
            NameFull: Wang, Xiaoxian
      – PersonEntity:
          Name:
            NameFull: Chen, Hui
      – PersonEntity:
          Name:
            NameFull: Chen, Fen
      – PersonEntity:
          Name:
            NameFull: Fang, Aijuan
      – PersonEntity:
          Name:
            NameFull: Cheng, Hanlin
      – PersonEntity:
          Name:
            NameFull: Weng, Hexiang
      – PersonEntity:
          Name:
            NameFull: Luo, Shouhua
      – PersonEntity:
          Name:
            NameFull: Yao, Jing
      – PersonEntity:
          Name:
            NameFull: Qian, Sunnan
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 05
              Text: May2026
              Type: published
              Y: 2026
          Identifiers:
            – Type: issn-print
              Value: 00942405
          Numbering:
            – Type: volume
              Value: 53
            – Type: issue
              Value: 5
          Titles:
            – TitleFull: Medical Physics
              Type: main
ResultId 1