Validation and estimation of spleen volume via computer-assisted segmentation on clinically acquired CT scans

Yiyuan Yang, Yucheng Tang, Riqiang Gao, Shunxing Bao, Yuankai Huo, Matthew T. McKenna, Michael R. Savona, Richard G. Abramson, Bennett A. LandmanJournal of Medical Imaging 2021 · DOI

Abstract

<b>Purpose:</b> Deep learning is a promising technique for spleen segmentation. Our study aims to validate the reproducibility of deep learning-based spleen volume estimation by performing spleen segmentation on clinically acquired computed tomography (CT) scans from patients with myeloproliferative neoplasms. <b>Approach:</b> As approved by the institutional review board, we obtained 138 de-identified abdominal CT scans. A sum of voxel volume on an expert annotator's segmentations establishes the ground truth (estimation 1). We used our deep convolutional neural network (estimation 2) alongside traditional linear estimations (estimation 3 and 4) to estimate spleen volumes independently. Dice coefficient, Hausdorff distance, <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:mrow> <mml:msup><mml:mrow><mml:mi>R</mml:mi></mml:mrow> <mml:mrow><mml:mn>2</mml:mn></mml:mrow> </mml:msup> </mml:mrow> </mml:math> coefficient, Pearson <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:mrow><mml:mi>R</mml:mi></mml:mrow> </mml:math> coefficient, the absolute difference in volume, and the relative difference in volume were calculated for 2 to 4 against the ground truth to compare and assess methods' performances. We re-labeled on scan-rescan on a subset of 40 studies to evaluate method reproducibility. <b>Results:</b> Calculated against the ground truth, the <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:mrow> <mml:msup><mml:mrow><mml:mi>R</mml:mi></mml:mrow> <mml:mrow><mml:mn>2</mml:mn></mml:mrow> </mml:msup> </mml:mrow> </mml:math> coefficients for our method (estimation 2) and linear method (estimation 3 and 4) are 0.998, 0.954, and 0.973, respectively. The Pearson <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:mrow><mml:mi>R</mml:mi></mml:mrow> </mml:math> coefficients for the estimations against the ground truth are 0.999, 0.963, and 0.978, respectively (paired <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:mrow><mml:mi>t</mml:mi></mml:mrow> </mml:math> -tests produced <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:mrow><mml:mi>p</mml:mi> <mml:mo><</mml:mo> <mml:mn>0.05</mml:mn></mml:mrow> </mml:math> between 2 and 3, and 2 and 4). <b>Conclusion:</b> The deep convolutional neural network algorithm shows excellent potential in rendering more precise spleen volume estimations. Our computer-aided segmentation exhibits reasonable improvements in splenic volume estimation accuracy.
BibTeX
@article{yang2021validation,
  author = {Yiyuan Yang and Yucheng Tang and Riqiang Gao and Shunxing Bao and Yuankai Huo and Matthew T. McKenna and Michael R. Savona and Richard G. Abramson and Bennett A. Landman},
  title = {{Validation and estimation of spleen volume via computer-assisted segmentation on clinically acquired CT scans}},
  journal = {Journal of Medical Imaging},
  year = {2021},
  doi = {10.1117/1.jmi.8.1.014004},
}

Loading related content...

Figure