Key result
Using a realistic CT-based torso and heart model resulted in a mean lesion localization error of 0.7 cm, compared to 3.4 cm when using a common torso and heart model.
Why the study?
Does an adjusted common torso model combined with a realistic heart model provide acceptable accuracy for noninvasive localization of heart pathologies compared to a realistic CT-based model?
Does an adjusted common torso model combined with a realistic heart model provide acceptable accuracy for noninvasive localization of heart pathologies compared to a realistic CT-based model?
A common torso model adjusted to patient dimensions combined with a realistic heart model provides acceptable accuracy for noninvasive localization of heart pathologies, potentially reducing the need for patient-specific CT scans.
Answers are grounded in this paper and cite their sources.
Improves localization accuracy with patient-specific models; leaves open whether adjusted common models suffice for clinical use.
The influence of the fidelity of an inhomogeneous torso model on the noninvasive inverse localization of selected heart pathologies was studied. It was supposed that the pathology is small enough to be represented by a single dipole. In the simulation study the accuracy of localization of an ischemic lesion was investigated. 18 ischemic lesions in the ventricles were modeled and corresponding ECGs were simulated in 7 CT-based realistic torso models. Position of each lesion represented by a single dipole was then searched by an inverse solution using different approximations of the torso and heart. If the realistic CT-based torso and heart model was used the mean error of the lesion localization was 0.7 cm. When a common torso model adjusted to patient-specific shape using 10 anthropometric parameters with a common heart model was used, the localization error reached 3.4 cm and decreased to 1.0 cm if properly formed and rotated heart model was used. In the experimental study the site of ectopic activation was noninvasively localized in a patient with premature ventricular activity. The assessed site position was in agreement with the site determined by intracardiac mapping if realistic or adjusted common torso model with lungs and properly formed and located heart model were used. From the results of both studies it can be concluded that the use of a common torso model adjusted in concordance with patient torso dimensions and with realistic heart model can lead to acceptable accuracy of the inverse localization of the searched heart pathology.
No takes yet. Share an insight, caveat, or question.
Tyšler et al. (2013) studied Ischemic lesions and premature ventricular activity (n=8). Realistic CT-based torso and heart model vs. Common torso model with common heart model was evaluated on Mean lesion localization error. Using a realistic CT-based torso and heart model resulted in a mean lesion localization error of 0.7 cm, compared to 3.4 cm when using a common torso and heart model.
Synapse has enriched 10 related papers. Consider them for comparative context: