TY - GEN
T1 - Confidence-based training for clinical data uncertainty in image-based prediction of cardiac ablation targets
AU - Cabrera-Lozoya, Rocío
AU - Margeta, Jan
AU - Le Folgoc, Loïc
AU - Komatsu, Yuki
AU - Berte, Benjamin
AU - Relan, Jatin
AU - Cochet, Hubert
AU - Haïssaguerre, Michel
AU - Jaïs, Pierre
AU - Ayache, Nicholas
AU - Sermesant, Maxime
N1 - Publisher Copyright:
© Springer International Publishing Switzerland 2014.
PY - 2014/1/1
Y1 - 2014/1/1
N2 - Ventricular radio-frequency ablation (RFA) can have a critical impact on preventing sudden cardiac arrest but is challenging due to a highly complex arrhythmogenic substrate. This work aims to identify local image characteristics capable of predicting the presence of local abnormal ventricular activities (LAVA). This can allow, pre-operatively and non-invasively, to improve and accelerate the procedure. To achieve this, intensity and texture-based local image features are computed and random forests are used for classification. However using machinelearning approaches on such complex multimodal data can prove difficult due to the inherent errors in the training set. In this manuscript we present a detailed analysis of these error sources due in particular to catheter motion and the data fusion process. We derived a principled analysis of confidence impact on classification. Moreover, we demonstrate how formal integration of these uncertainties in the training process improves the algorithm’s performance, opening up possibilities for noninvasive image-based prediction of RFA targets.
AB - Ventricular radio-frequency ablation (RFA) can have a critical impact on preventing sudden cardiac arrest but is challenging due to a highly complex arrhythmogenic substrate. This work aims to identify local image characteristics capable of predicting the presence of local abnormal ventricular activities (LAVA). This can allow, pre-operatively and non-invasively, to improve and accelerate the procedure. To achieve this, intensity and texture-based local image features are computed and random forests are used for classification. However using machinelearning approaches on such complex multimodal data can prove difficult due to the inherent errors in the training set. In this manuscript we present a detailed analysis of these error sources due in particular to catheter motion and the data fusion process. We derived a principled analysis of confidence impact on classification. Moreover, we demonstrate how formal integration of these uncertainties in the training process improves the algorithm’s performance, opening up possibilities for noninvasive image-based prediction of RFA targets.
UR - https://www.scopus.com/pages/publications/84917706467
U2 - 10.1007/978-3-319-13972-2_14
DO - 10.1007/978-3-319-13972-2_14
M3 - Conference contribution
AN - SCOPUS:84917706467
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 148
EP - 159
BT - Medical Computer Vision
A2 - Müller, Henning
A2 - Menze, Bjoern
A2 - Zhang, Shaoting
A2 - Cai, Weidong (Tom)
A2 - Menze, Bjoern
A2 - Langs, Georg
A2 - Metaxas, Dimitris
A2 - Langs, Georg
A2 - Müller, Henning
A2 - Kelm, Michael
A2 - Montillo, Albert
A2 - Cai, Weidong (Tom)
PB - Springer Verlag
T2 - International Workshop on Medical Computer Vision: Algorithms for Big Data was held in conjunction with 17th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI-bigMCV 2014
Y2 - 18 September 2014 through 18 September 2014
ER -