Murad Megjhani1,2, Miriam Weiss3,4, Soon Bin Kwon1,2, Jenna Ford1,2, Daniel Nametz1,2, Nick Kastenholz2,3, Hart Fogel2, Angela Velazquez1, David Roh1,5, Sachin Agarwal1,5, E Sander Connolly5,6, Jan Claassen1,5, Gerrit A Schubert3,4, Soojin Park7,8,9,10. 1. Department of Neurology, Columbia University, 177 Fort Washington Ave, 8 Milstein - 300 Center, New York, NY, USA. 2. Program for Hospital and Intensive Care Informatics, Department of Neurology, Columbia University, New York, NY, USA. 3. Department of Neurosurgery, RWTH Aachen University, Aachen, Germany. 4. Department of Neurosurgery, Kantonsspital Aarau AG, Aarau, Switzerland. 5. NewYork-Presbyterian Hospital, Columbia University Irving Medical Center, New York, NY, USA. 6. Department of Neurosurgery, Columbia University, New York, NY, USA. 7. Department of Neurology, Columbia University, 177 Fort Washington Ave, 8 Milstein - 300 Center, New York, NY, USA. spark@columbia.edu. 8. Program for Hospital and Intensive Care Informatics, Department of Neurology, Columbia University, New York, NY, USA. spark@columbia.edu. 9. NewYork-Presbyterian Hospital, Columbia University Irving Medical Center, New York, NY, USA. spark@columbia.edu. 10. Department of Biomedical Informatics, Columbia University, New York, NY, USA. spark@columbia.edu.
Abstract
BACKGROUND: Dysfunctional cerebral autoregulation often precedes delayed cerebral ischemia (DCI). Currently, there are no data-driven techniques that leverage this information to predict DCI in real time. Our hypothesis is that information using continuous updated analyses of multimodal neuromonitoring and cerebral autoregulation can be deployed to predict DCI. METHODS: Time series values of intracranial pressure, brain tissue oxygenation, cerebral perfusion pressure (CPP), optimal CPP (CPPOpt), ΔCPP (CPP - CPPOpt), mean arterial pressure, and pressure reactivity index were combined and summarized as vectors. A validated temporal signal angle measurement was modified into a classification algorithm that incorporates hourly data. The time-varying temporal signal angle measurement (TTSAM) algorithm classifies DCI at varying time points by vectorizing and computing the angle between the test and reference time signals. The patient is classified as DCI+ if the error between the time-varying test vector and DCI+ reference vector is smaller than that between the time-varying test vector and DCI- reference vector. Finally, prediction at time point t is calculated as the majority voting over all the available signals. The leave-one-patient-out cross-validation technique was used to train and report the performance of the algorithms. The TTSAM and classifier performance was determined by balanced accuracy, F1 score, true positive, true negative, false positive, and false negative over time. RESULTS: One hundred thirty-one patients with aneurysmal subarachnoid hemorrhage who underwent multimodal neuromonitoring were identified from two centers (Columbia University: 52 [39.7%], Aachen University: 79 [60.3%]) and included in the analysis. Sixty-four (48.5%) patients had DCI, and DCI was diagnosed 7.2 ± 3.3 days after hemorrhage. The TTSAM algorithm achieved a balanced accuracy of 67.3% and an F1 score of 0.68 at 165 h (6.9 days) from bleed day with a true positive of 0.83, false positive of 0.16, true negative of 0.51, and false negative of 0.49. CONCLUSIONS: A TTSAM algorithm using multimodal neuromonitoring and cerebral autoregulation calculations shows promise to classify DCI in real time.
BACKGROUND: Dysfunctional cerebral autoregulation often precedes delayed cerebral ischemia (DCI). Currently, there are no data-driven techniques that leverage this information to predict DCI in real time. Our hypothesis is that information using continuous updated analyses of multimodal neuromonitoring and cerebral autoregulation can be deployed to predict DCI. METHODS: Time series values of intracranial pressure, brain tissue oxygenation, cerebral perfusion pressure (CPP), optimal CPP (CPPOpt), ΔCPP (CPP - CPPOpt), mean arterial pressure, and pressure reactivity index were combined and summarized as vectors. A validated temporal signal angle measurement was modified into a classification algorithm that incorporates hourly data. The time-varying temporal signal angle measurement (TTSAM) algorithm classifies DCI at varying time points by vectorizing and computing the angle between the test and reference time signals. The patient is classified as DCI+ if the error between the time-varying test vector and DCI+ reference vector is smaller than that between the time-varying test vector and DCI- reference vector. Finally, prediction at time point t is calculated as the majority voting over all the available signals. The leave-one-patient-out cross-validation technique was used to train and report the performance of the algorithms. The TTSAM and classifier performance was determined by balanced accuracy, F1 score, true positive, true negative, false positive, and false negative over time. RESULTS: One hundred thirty-one patients with aneurysmal subarachnoid hemorrhage who underwent multimodal neuromonitoring were identified from two centers (Columbia University: 52 [39.7%], Aachen University: 79 [60.3%]) and included in the analysis. Sixty-four (48.5%) patients had DCI, and DCI was diagnosed 7.2 ± 3.3 days after hemorrhage. The TTSAM algorithm achieved a balanced accuracy of 67.3% and an F1 score of 0.68 at 165 h (6.9 days) from bleed day with a true positive of 0.83, false positive of 0.16, true negative of 0.51, and false negative of 0.49. CONCLUSIONS: A TTSAM algorithm using multimodal neuromonitoring and cerebral autoregulation calculations shows promise to classify DCI in real time.
Authors: David Carrera; Montse Gorchs; Marisol Querol; Sònia Abilleira; Marc Ribó; Mònica Millán; Anna Ramos; Pedro Cardona; Xabier Urra; Ana Rodríguez-Campello; Luis Prats-Sánchez; Francisco Purroy; Joaquín Serena; David Cánovas; Josep Zaragoza-Brunet; Jerzy A Krupinski; Xavier Ustrell; Júlia Saura; Sonia García; Maria Àngela Mora; Xavier Jiménez; Antoni Dávalos; Natalia Pérez de la Ossa Journal: J Neurointerv Surg Date: 2018-12-22 Impact factor: 5.836
Authors: J Michael Schmidt; Daby Sow; Michael Crimmins; David Albers; Sachin Agarwal; Jan Claassen; E Sander Connolly; Mitchell S V Elkind; George Hripcsak; Stephan A Mayer Journal: Neurocrit Care Date: 2014-06 Impact factor: 3.210
Authors: Jan Claassen; Lawrence J Hirsch; Kurt T Kreiter; Evelyn Y Du; E Sander Connolly; Ronald G Emerson; Stephan A Mayer Journal: Clin Neurophysiol Date: 2004-12 Impact factor: 3.708
Authors: Murad Megjhani; Kalijah Terilli; Hans-Peter Frey; Angela G Velazquez; Kevin William Doyle; Edward Sander Connolly; David Jinou Roh; Sachin Agarwal; Jan Claassen; Noemie Elhadad; Soojin Park Journal: Front Neurol Date: 2018-03-07 Impact factor: 4.003
Authors: Karol P Budohoski; Marek Czosnyka; Peter Smielewski; Magdalena Kasprowicz; Adel Helmy; Diederik Bulters; John D Pickard; Peter J Kirkpatrick Journal: Stroke Date: 2012-11-13 Impact factor: 7.914