An HPC Pipeline for Calcium Quantification of Aortic Root From Contrast-Enhanced CCT Scans
Precise assessment of calcification lesions in the Aortic Root (AR) is relevant for the success of the Transcatheter Aortic Valve Implantation (TAVI) procedure. To this end, the radiologists analyze the Cardiac Computed Tomography (CCT) scans of patients, and detect the position and extent of the ca...
Main Authors: | , , , , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2023-01-01
|
Series: | IEEE Access |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/10251934/ |
_version_ | 1797676691481427968 |
---|---|
author | Giada Minghini Armando Ugo Cavallo Andrea Miola Valentina Sisini Enrico Calore Francesca Fortini Rino Micheloni Paola Rizzo Sebastiano Fabio Schifano Francesco Vieceli Dalla Sega Cristian Zambelli |
author_facet | Giada Minghini Armando Ugo Cavallo Andrea Miola Valentina Sisini Enrico Calore Francesca Fortini Rino Micheloni Paola Rizzo Sebastiano Fabio Schifano Francesco Vieceli Dalla Sega Cristian Zambelli |
author_sort | Giada Minghini |
collection | DOAJ |
description | Precise assessment of calcification lesions in the Aortic Root (AR) is relevant for the success of the Transcatheter Aortic Valve Implantation (TAVI) procedure. To this end, the radiologists analyze the Cardiac Computed Tomography (CCT) scans of patients, and detect the position and extent of the calcium deposits. In this contribution, we develop a computationally efficient High-Performance Computing (HPC) system to detect, segment, and quantify volumes of calcium in contrast-enhanced CCTs, embedding in a three-step pipeline two 3D Convolutional Neural Networks (CNN) and a threshold adaptive filter. The first step crops the images to a bounding-box around the AR keeping the original resolution, the second builds the segmentation, and the third detects and measures the volume of the calcium lesions. Our system is trained on high-resolution contrast-CCTs routinely planned for the TAVI manually annotated by expert radiologists, and evaluated on a test-set of patients with different levels of calcifications. The accuracy achieved in segmenting the AR is approximately 92% for the test-set, while the average difference of calcium lesion volumes with respect to the radiologists measurements is about 0.49 mm3. Running on a 4X NVIDIA-V100 and an 8X NVIDIA-A100 GPU systems, we achieve a remarkable inference throughput of 17 and 70 CCT/sec respectively, and a linear scaling of computing performance. Our contribution provides an HPC system suitable for hospital premises installation and is able to aid radiologists in assessing the calcification level of patients undergoing the TAVI, making this process automated, fast and more reliable. |
first_indexed | 2024-03-11T22:33:54Z |
format | Article |
id | doaj.art-8181d3406e3543d5a61954478dc4c9cb |
institution | Directory Open Access Journal |
issn | 2169-3536 |
language | English |
last_indexed | 2024-03-11T22:33:54Z |
publishDate | 2023-01-01 |
publisher | IEEE |
record_format | Article |
series | IEEE Access |
spelling | doaj.art-8181d3406e3543d5a61954478dc4c9cb2023-09-22T23:01:28ZengIEEEIEEE Access2169-35362023-01-011110130910131910.1109/ACCESS.2023.331573410251934An HPC Pipeline for Calcium Quantification of Aortic Root From Contrast-Enhanced CCT ScansGiada Minghini0https://orcid.org/0009-0007-8298-8815Armando Ugo Cavallo1https://orcid.org/0000-0001-8390-7721Andrea Miola2https://orcid.org/0000-0003-0740-5070Valentina Sisini3Enrico Calore4https://orcid.org/0000-0002-2301-3838Francesca Fortini5https://orcid.org/0000-0003-0807-3792Rino Micheloni6https://orcid.org/0000-0002-3400-2624Paola Rizzo7https://orcid.org/0000-0001-7174-9674Sebastiano Fabio Schifano8https://orcid.org/0000-0002-0132-9196Francesco Vieceli Dalla Sega9https://orcid.org/0000-0003-3445-3983Cristian Zambelli10https://orcid.org/0000-0001-8755-0504Università degli Studi di Ferrara, Ferrara, ItalyIstituto Dermopatico dell’Immacolata (IDI) IRCCS, Rome, ItalyUniversità degli Studi di Ferrara, Ferrara, ItalyUniversità degli Studi di Ferrara, Ferrara, ItalyIstituto Nazionale di Fisica Nucleare (INFN) Ferrara, Ferrara, ItalyMaria Cecilia Hospital, GVM Care and Research, Cotignola, ItalyAvaneidi srl, Saronno, ItalyUniversità degli Studi di Ferrara, Ferrara, ItalyUniversità degli Studi di Ferrara, Ferrara, ItalyMaria Cecilia Hospital, GVM Care and Research, Cotignola, ItalyUniversità degli Studi di Ferrara, Ferrara, ItalyPrecise assessment of calcification lesions in the Aortic Root (AR) is relevant for the success of the Transcatheter Aortic Valve Implantation (TAVI) procedure. To this end, the radiologists analyze the Cardiac Computed Tomography (CCT) scans of patients, and detect the position and extent of the calcium deposits. In this contribution, we develop a computationally efficient High-Performance Computing (HPC) system to detect, segment, and quantify volumes of calcium in contrast-enhanced CCTs, embedding in a three-step pipeline two 3D Convolutional Neural Networks (CNN) and a threshold adaptive filter. The first step crops the images to a bounding-box around the AR keeping the original resolution, the second builds the segmentation, and the third detects and measures the volume of the calcium lesions. Our system is trained on high-resolution contrast-CCTs routinely planned for the TAVI manually annotated by expert radiologists, and evaluated on a test-set of patients with different levels of calcifications. The accuracy achieved in segmenting the AR is approximately 92% for the test-set, while the average difference of calcium lesion volumes with respect to the radiologists measurements is about 0.49 mm3. Running on a 4X NVIDIA-V100 and an 8X NVIDIA-A100 GPU systems, we achieve a remarkable inference throughput of 17 and 70 CCT/sec respectively, and a linear scaling of computing performance. Our contribution provides an HPC system suitable for hospital premises installation and is able to aid radiologists in assessing the calcification level of patients undergoing the TAVI, making this process automated, fast and more reliable.https://ieeexplore.ieee.org/document/10251934/Aortic root segmentationcalcifications assessmentcontrast-enhanced CCT scansGPU performance analysis |
spellingShingle | Giada Minghini Armando Ugo Cavallo Andrea Miola Valentina Sisini Enrico Calore Francesca Fortini Rino Micheloni Paola Rizzo Sebastiano Fabio Schifano Francesco Vieceli Dalla Sega Cristian Zambelli An HPC Pipeline for Calcium Quantification of Aortic Root From Contrast-Enhanced CCT Scans IEEE Access Aortic root segmentation calcifications assessment contrast-enhanced CCT scans GPU performance analysis |
title | An HPC Pipeline for Calcium Quantification of Aortic Root From Contrast-Enhanced CCT Scans |
title_full | An HPC Pipeline for Calcium Quantification of Aortic Root From Contrast-Enhanced CCT Scans |
title_fullStr | An HPC Pipeline for Calcium Quantification of Aortic Root From Contrast-Enhanced CCT Scans |
title_full_unstemmed | An HPC Pipeline for Calcium Quantification of Aortic Root From Contrast-Enhanced CCT Scans |
title_short | An HPC Pipeline for Calcium Quantification of Aortic Root From Contrast-Enhanced CCT Scans |
title_sort | hpc pipeline for calcium quantification of aortic root from contrast enhanced cct scans |
topic | Aortic root segmentation calcifications assessment contrast-enhanced CCT scans GPU performance analysis |
url | https://ieeexplore.ieee.org/document/10251934/ |
work_keys_str_mv | AT giadaminghini anhpcpipelineforcalciumquantificationofaorticrootfromcontrastenhancedcctscans AT armandougocavallo anhpcpipelineforcalciumquantificationofaorticrootfromcontrastenhancedcctscans AT andreamiola anhpcpipelineforcalciumquantificationofaorticrootfromcontrastenhancedcctscans AT valentinasisini anhpcpipelineforcalciumquantificationofaorticrootfromcontrastenhancedcctscans AT enricocalore anhpcpipelineforcalciumquantificationofaorticrootfromcontrastenhancedcctscans AT francescafortini anhpcpipelineforcalciumquantificationofaorticrootfromcontrastenhancedcctscans AT rinomicheloni anhpcpipelineforcalciumquantificationofaorticrootfromcontrastenhancedcctscans AT paolarizzo anhpcpipelineforcalciumquantificationofaorticrootfromcontrastenhancedcctscans AT sebastianofabioschifano anhpcpipelineforcalciumquantificationofaorticrootfromcontrastenhancedcctscans AT francescoviecelidallasega anhpcpipelineforcalciumquantificationofaorticrootfromcontrastenhancedcctscans AT cristianzambelli anhpcpipelineforcalciumquantificationofaorticrootfromcontrastenhancedcctscans AT giadaminghini hpcpipelineforcalciumquantificationofaorticrootfromcontrastenhancedcctscans AT armandougocavallo hpcpipelineforcalciumquantificationofaorticrootfromcontrastenhancedcctscans AT andreamiola hpcpipelineforcalciumquantificationofaorticrootfromcontrastenhancedcctscans AT valentinasisini hpcpipelineforcalciumquantificationofaorticrootfromcontrastenhancedcctscans AT enricocalore hpcpipelineforcalciumquantificationofaorticrootfromcontrastenhancedcctscans AT francescafortini hpcpipelineforcalciumquantificationofaorticrootfromcontrastenhancedcctscans AT rinomicheloni hpcpipelineforcalciumquantificationofaorticrootfromcontrastenhancedcctscans AT paolarizzo hpcpipelineforcalciumquantificationofaorticrootfromcontrastenhancedcctscans AT sebastianofabioschifano hpcpipelineforcalciumquantificationofaorticrootfromcontrastenhancedcctscans AT francescoviecelidallasega hpcpipelineforcalciumquantificationofaorticrootfromcontrastenhancedcctscans AT cristianzambelli hpcpipelineforcalciumquantificationofaorticrootfromcontrastenhancedcctscans |