Limits...
A reference dataset for verifying numerical electrophysiological heart models.

Koch H, Bousseljot RD, Kosch O, Jahnke C, Paetsch I, Fleck E, Schnackenburg B - Biomed Eng Online (2011)

Bottom Line: The latter were recorded simultaneously from the same individuals a few hours after the MRI sessions.A training dataset is made publicly available; datasets for blind testing will remain undisclosed.While the MRI data may provide a common input that can be applied to different numerical heart models, the verification and comparison of different models can be performed by comparing the measured biosignals with forward calculated signals from the models.

View Article: PubMed Central - HTML - PubMed

Affiliation: Physikalisch-Technische Bundesanstalt, Abbestr, 2-12, 10587 Berlin, Germany. hans.koch@ptb.de

ABSTRACT

Background: The evaluation, verification and comparison of different numerical heart models are difficult without a commonly available database that could be utilized as a reference. Our aim was to compile an exemplary dataset.

Methods: The following methods were employed: Magnetic Resonance Imaging (MRI) of heart and torso, Body Surface Potential Maps (BSPM) and MagnetoCardioGraphy (MCG) maps. The latter were recorded simultaneously from the same individuals a few hours after the MRI sessions.

Results: A training dataset is made publicly available; datasets for blind testing will remain undisclosed.

Conclusions: While the MRI data may provide a common input that can be applied to different numerical heart models, the verification and comparison of different models can be performed by comparing the measured biosignals with forward calculated signals from the models.

Show MeSH
MR image stack of the torso. A subset of the survey series (301). IM_0090 and IM_0108 show the marker pills used as spatial reference.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC3037925&req=5

Figure 1: MR image stack of the torso. A subset of the survey series (301). IM_0090 and IM_0108 show the marker pills used as spatial reference.

Mentions: The image series provided for the dataset consist at first of a survey series (series 301). A subset is shown in Figure 1. For this, a breath hold multiple 2D (M2D) single shot balanced Steady State Free Precession (bSSFP) sequence was chosen with the following parameters: TR/TE/flip: 3.1 ms, 1.2 ms, 50 deg; spatial resolution: 3 × 3 × 6 mm3; number of slices: 80 (covering 480 mm in feet-head direction); scan time: 21 s. In addition, three further series have been added that contain the dynamic geometry information of the heart. Series 601 (M2D-bSSFP Cine) provides the full heart cycle in several short axis cuts (completely covering the left ventricle); sequence parameters: TR/TE/flip: 3.4 ms, 1.7 ms, 60 deg; spatial resolution: 1.8 × 1.8 × 8 mm3; 50 heart phases. The last two series (701 and 801) are free breathing single-phase navigator-gated 3D-bSSFP transversal datasets at diastole and systole, respectively, with the following parameters: T2 preparation (TE = 50 ms) and fat saturation pre-pulses, TR/TE/flip: 4.7 ms, 2.3 ms, 100 deg; measured spatial resolution: 1.5 × 1.5 × 3 mm3; reconstructed to 1.5 × 1.5 × 1.5 mm3; number of slices: 180 (covering 270 mm in feet/head direction). The acquisition duration per heart beat for the diastolic dataset was 110 ms (scan time: 257 heart beats) and for the systolic dataset 55 ms (scan time: 514 heart beats).


A reference dataset for verifying numerical electrophysiological heart models.

Koch H, Bousseljot RD, Kosch O, Jahnke C, Paetsch I, Fleck E, Schnackenburg B - Biomed Eng Online (2011)

MR image stack of the torso. A subset of the survey series (301). IM_0090 and IM_0108 show the marker pills used as spatial reference.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC3037925&req=5

Figure 1: MR image stack of the torso. A subset of the survey series (301). IM_0090 and IM_0108 show the marker pills used as spatial reference.
Mentions: The image series provided for the dataset consist at first of a survey series (series 301). A subset is shown in Figure 1. For this, a breath hold multiple 2D (M2D) single shot balanced Steady State Free Precession (bSSFP) sequence was chosen with the following parameters: TR/TE/flip: 3.1 ms, 1.2 ms, 50 deg; spatial resolution: 3 × 3 × 6 mm3; number of slices: 80 (covering 480 mm in feet-head direction); scan time: 21 s. In addition, three further series have been added that contain the dynamic geometry information of the heart. Series 601 (M2D-bSSFP Cine) provides the full heart cycle in several short axis cuts (completely covering the left ventricle); sequence parameters: TR/TE/flip: 3.4 ms, 1.7 ms, 60 deg; spatial resolution: 1.8 × 1.8 × 8 mm3; 50 heart phases. The last two series (701 and 801) are free breathing single-phase navigator-gated 3D-bSSFP transversal datasets at diastole and systole, respectively, with the following parameters: T2 preparation (TE = 50 ms) and fat saturation pre-pulses, TR/TE/flip: 4.7 ms, 2.3 ms, 100 deg; measured spatial resolution: 1.5 × 1.5 × 3 mm3; reconstructed to 1.5 × 1.5 × 1.5 mm3; number of slices: 180 (covering 270 mm in feet/head direction). The acquisition duration per heart beat for the diastolic dataset was 110 ms (scan time: 257 heart beats) and for the systolic dataset 55 ms (scan time: 514 heart beats).

Bottom Line: The latter were recorded simultaneously from the same individuals a few hours after the MRI sessions.A training dataset is made publicly available; datasets for blind testing will remain undisclosed.While the MRI data may provide a common input that can be applied to different numerical heart models, the verification and comparison of different models can be performed by comparing the measured biosignals with forward calculated signals from the models.

View Article: PubMed Central - HTML - PubMed

Affiliation: Physikalisch-Technische Bundesanstalt, Abbestr, 2-12, 10587 Berlin, Germany. hans.koch@ptb.de

ABSTRACT

Background: The evaluation, verification and comparison of different numerical heart models are difficult without a commonly available database that could be utilized as a reference. Our aim was to compile an exemplary dataset.

Methods: The following methods were employed: Magnetic Resonance Imaging (MRI) of heart and torso, Body Surface Potential Maps (BSPM) and MagnetoCardioGraphy (MCG) maps. The latter were recorded simultaneously from the same individuals a few hours after the MRI sessions.

Results: A training dataset is made publicly available; datasets for blind testing will remain undisclosed.

Conclusions: While the MRI data may provide a common input that can be applied to different numerical heart models, the verification and comparison of different models can be performed by comparing the measured biosignals with forward calculated signals from the models.

Show MeSH