Patient-specific prediction of SEEG electrode bending for stereotactic neurosurgical planning
Tóm tắt
Electrode bending observed after stereotactic interventions is typically not accounted for in either computer-assisted planning algorithms, where straight trajectories are assumed, or in quality assessment, where only metrics related to entry and target points are reported. Our aim is to provide a fully automated and validated pipeline for the prediction of stereo-electroencephalography (SEEG) electrode bending. We transform electrodes of 86 cases into a common space and compare features-based and image-based neural networks on their ability to regress local displacement (
$$\mathbf{lu} $$
) or electrode bending (
$$\hat{\mathbf{eb }}$$
). Electrodes were stratified into six groups based on brain structures at the entry and target point. Models, both with and without Monte Carlo (MC) dropout, were trained and validated using tenfold cross-validation. mage-based models outperformed features-based models for all groups, and models that predicted
$$\mathbf{lu} $$
performed better than for
$$\hat{\mathbf{eb }}$$
. Image-based model prediction with MC dropout resulted in lower mean squared error (MSE) with improvements up to 12.9% (
$$\mathbf{lu} $$
) and 39.9% (
$$\hat{\mathbf{eb }}$$
), compared to no dropout. Using an image of brain tissue types (cortex, white and deep grey matter) resulted in similar, and sometimes better performance, compared to using a T1-weighted MRI when predicting
$$\mathbf{lu} $$
. When inferring trajectories of image-based models (brain tissue types), 86.9% of trajectories had an MSE
$$\le 1$$
mm. An image-based approach regressing local displacement with an image of brain tissue types resulted in more accurate electrode bending predictions compared to other approaches, inputs, and outputs. Future work will investigate the integration of electrode bending into planning and quality assessment algorithms.
Tài liệu tham khảo
Cardoso MJ, Modat M, Wolz R, Melbourne A, Cash D, Rueckert D, Ourselin S (2015) Geodesic information flows: spatially-variant graphs and their application to segmentation and fusion. IEEE Trans Med Imaging 34(9):1976–1988
Chassoux F, Navarro V, Catenoix H, Valton L, Vignal JP (2018) Planning and management of SEEG. Clin Neurophysiol 48:25–37
Dorfer C, Minchev G, Czech T, Stefanits H, Feucht M, Pataraia E, Baumgartner C, Kronreif G, Wolfsberger S (2017) A novel miniature robotic device for frameless implantation of depth electrodes in refractory epilepsy. J Neurosurg 126(5):1622–1628
Fonov VS, Evans AC, McKinstry RC, Almli CR, Collins DL (2009) Unbiased nonlinear average age-appropriate brain templates from birth to adulthood. NeuroImage 47:s102
Gal Y, Ghahramani Z (2016) Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning. Proceedings of the 33rd International Conference on Machine Learning, JMLR 48
Granados A, Lucena O, Vakharia V, Miserocchi A, McEvoy AW, Vos SB, Rodionov R, Duncan JS, Sparks R, Ourselin S (2020) Towards Uncertainty Quantification for Electrode Bending Prediction in Stereotactic Neurosurgery. IEEE 17th International Symposium on Biomedical Imaging (ISBI) pp. 674–677
Granados A, Mancini M, Vos SB, Lucena O, Vakharia V, Rodionov R, Miserocchi A, McEvoy AW, Duncan JS, Sparks R, Ourselin S (2018) A Machine Learning Approach to Predict Instrument Bending in Stereotactic Neurosurgery. International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI) pp. 238–246
Granados A, Rodionov R, Vakharia V, McEvoy AW, Miserocchi A, O’Keeffe AG, Duncan JS, Sparks R, Ourselin S (2020) Automated computation and analysis of accuracy metrics in stereoencephalography. J of Neurosci Methods 340:108710
Granados A, Vakharia V, Rodionov R, Schweiger M, Vos SB, Keeffe AGO, Li K, Wu C, Miserocchi A, McEvoy AW, Clarkson MJ, Duncan JS, Sparks R, Ourselin S (2018) Automatic segmentation of stereoelectroencephalography (SEEG) electrodes post-implantation considering bending. Int J Comput Assist Radiol Surg 13(6):935–946
Herff C, Krusienski DJ, Kubben P (2020) The potential of stereotactic-EEG for brain-computer interfaces: current progress and future directions. Front Neurosci, Neuroprosthetics 14(123):1–8
Iglesias JE, Liu CY, Thompson P, Tu Z (2011) Robust brain extraction across datasets and comparison with publicly available methods. IEEE Trans Med Imaging 30(9):1617–1634
Lacour SP, Courtine G, Guck J (2016) Materials and technologies for soft implantable neuroprostheses. Nat Rev Mater 1(10):1–14
Li K, Vakharia VN, Sparks R, Rodionov R, Vos SB, McEvoy AW, Miserocchi A, Wang M, Ourselin S, Duncan JS (2019) Stereoelectroencephalography electrode placement: detection of blood vessel conflicts. Epilepsia 60:19421948
Minotti L, Montavont A, Scholly J, Tyvaert L, Taussig D (2018) Indications and limits of stereoelectroencephalography (SEEG). Clin Neurophysiol 48:15–24
Modat M, Cash DM, Daga P, Winston GP, Duncan JS, Ourselin S (2014) Global image registration using a symmetric block-matching approach. J of Med Imaging 1(2):024003
Nowell M, Rodionov R, Diehl B, Wehner T, Zombori G, Kinghorn J, Ourselin S, Duncan J, Miserocchi A, McEvoy A (2014) A novel method for implementation of frameless stereo EEG in epilepsy surgery. Oper Neurosurg 10(4):525–534
Vakharia VN, Sparks R, Miserocchi A, Vos SB, Keeffe AO, Rodionov R, McEvoy AW, Ourselin S, Duncan JS (2019) Computer-assisted planning for stereoelectroencephalography (SEEG). Neurotherapeutics 16:11831197
Vakharia VN, Sparks R, Keeffe AGO, Rodionov R, Miserocchi A, McEvoy A, Ourselin S, Duncan J (2017) Accuracy of intracranial electrode placement for stereoelectroencephalography: a systematic review and meta-analysis. Epilepsia 58(6):921–932