BrainLab

Close menu

BrainLab

Toggle menu

2023-2025

AUTOMATIC: Analysis of the relationship between the autonomic nervous system and cerebral autoregulation using the machine learning approach

Project leader: Agnieszka Uryga, PhD

SONATA 18 programme, National Science Centre, Poland, grant UMO-2022/47/D/ST7/00229

Global cerebral circulation is under the control of metabolic factors, cerebral autoregulation, and a mechanism responding to control by the autonomic nervous system. Cerebral vasculature is known to be heavily innervated by both sympathetic and parasympathetic neurones, strongly suggesting a role in the maintenance of brain perfusion. One of the most extensively analysed markers of the autonomic nervous system is baroreceptor sensitivity (BRS), due to their non-invasiveness and ease of application in clinical settings. However, the relationship between BRS and cerebral autoregulation is still unclear. Recent studies have shown that the interrelationship between cerebral autoregulation and the autonomic nervous system is heterogeneous and varies from patient to patient in traumatic brain injury. Even studies on healthy volunteers are inconclusive. It has been suggested on the one hand that there is an inverse correlation between cerebral autoregulation and autonomic response variables – when the activity of baroreflex increase cerebral autoregulation worsens. On the other hand, it has been shown that cerebral autoregulation was attenuated after baroreceptor suppression, implying a direct relationship.

It is believed that cerebral autoregulation and BRS are complementary to each other and provide a more complex picture of cerebral blood flow regulation. As biomedical signals and time-series parameters derived from signals are mostly non-stationary (their statistical properties change over time) there is a high need to use more advanced methods to track and analyse of correlation between signals. In the standard approach, the value of the signal is averaged in a moving time window and then a mean value from the recording is captured. To estimate the strength and direction of monotonic association between two metrics in the standard approach the Spearman or Pearson correlation coefficient is applied using a simple mean value in a total group of subjects. This approach may lead to the leakage of crucial information about temporal correlation, which may be dynamic and vary in time, where the characteristics of this fluctuation may obtain additional prognostic information. Moreover, current medicine is aimed at an individualised approach to patient treatment and condition assessment therefore “one patient-one curve” approach tends to be a better option than the “one patient-one value”.

To bridge this gap, we propose a project called AUTOMATIC – Analysis of the relationship between the AUTOnoMic nervous system and cerebral AutoregulaTIon using the maChine learning approach. The main hypothesis in our project is that the dynamic of temporal association between cerebral autoregulation and the autonomic nervous system is crucial and needs to be investigated using more advanced approaches to data analysis, exploiting the information contained in the fluctuations of this association. This kind of method is well-established in econometry to track changes in stock indices and exchange rates, but we believed that they could be successfully adaptive for biomedical applications. We plan to use those methods to characterise a temporal profile of cerebral autoregulation-autonomic nervous system association based on data and signals (cerebral blood flow velocity (CBFV), electrocardiogram (ECG), arterial blood pressure (ABP)) captured during non-invasive hemodynamic measurements in healthy volunteers. As plenty of indices allow characterising cerebral autoregulation and the autonomic nervous system along with standard metrics and scales used to describe the condition of patients with intracranial pathologies, this association cannot be interpreted in isolation from the rest of the biomedical signals and the patient’s condition assessment. Therefore, we plan to use a machine-learning approach to feature selection.

In final, our goal is to build machine-learning-based diagnostic models in patients with intracranial pathologies. A robust understanding of how cerebral autoregulation-autonomic response variables are interconnected may improve the ability to predict patients with traumatic brain injury or subarachnoid haemorrhage at risk for cerebral autoregulation failure and autonomic nervous system dysfunction.

Publications:

  1. Uryga, A., Mataczyński, C., Pelah, A., Burzyńska, M., Robba, C., Czosnyka, M., & CENTER-TBI high-resolution sub-study participants and investigators (2024) Exploration of Simultaneous Transients Between Cerebral Hemodynamics and the Autonomic Nervous System Using Windowed Time-lagged Cross-correlation Matrices: A CENTER-TBI Study. Acta Neurochirurgica, 166 art. 504. https://doi.org/10.1007/s00701-024-06375-6
  2. Pelah, A., Najdek, M., Czosnyka, M., & Uryga, A. (2024) Relationship Between the Amplitudes of Cerebral Blood Flow Velocity and Intracranial Pressure Using Linear and Non-linear Approach. Journal of Clinical Monitoring and Computing, pre-online.  https://doi.org/10.1007/s10877-024-01243-1
  3. Sanfilippo, F., Uryga, A., Santonocito, C., Patroniti, N. […] & Robba, C. (2024) Effects of Very Early Hyperoxemia on Neurologic Outcome After Out-of-hospital Cardiac Arrest: A Secondary Analysis of the TTM-2 Trial. Resuscitation, 207 art. 110460. https://doi.org/10.1016/j.resuscitation.2024.110460
  4. Burzyńska, M., Woźniak, J., Urbański, P., Kędziora, J., Załuski, R., Goździk, W., & Uryga, A. (2024) Heart Rate Variability and Cerebral Autoregulation in Patients with Traumatic Brain Injury with Paroxysmal Sympathetic Hyperactivity. Neurocritical Care, pre-online. https://doi.org/10.1007/s12028-024-02149-1
  5. Uryga, A., Najda, M., Berent, I., Mataczyński, C., Urbański, P., Kasprowicz, M., & Buchner, T. (2024) The Impact of Controlled Breathing on Autonomic Nervous System Modulation: Analysis Using Phase-Rectified Signal Averaging, Entropy and Heart Rate Variability. Physiological Measurement, 45 art. 095004. https://doi.org/10.1088/1361-6579/ad7778
  6. Uryga, A., Czosnyka, M., Robba, C., Nasr, N., & Kasprowicz, M. (2024) The Time Constant of the Cerebral Arterial Bed: Exploring Age-Related Implications. Journal of Clinical Monitoring and Computing, 38, 1227-1236. https://doi.org/10.1007/s10877-024-01142-5
  7. Sanfilippo, F., Uryga, A., Ball, L., Battaglini, D., Iavarone, I. G., Smielewski, P., Beqiri, E., Czosnyka, M., Patroniti, N., & Robba, C. (2024) The Effect of Recruitment Maneuvers on Cerebrovascular Dynamics and Right Ventricular Function in Patients with Acute Brain Injury: a Single-Center Prospective Study. Neurocritical Care, 41, 38-48. https://doi.org/10.1007/s12028-024-01939-x
  8. Uryga, A., Kasprowicz, M., Budohoski, K., Nasr, N., & Czosnyka, M. (2024) Predictive Value of Cerebrovascular Time Constant for Delayed Cerebral Ischemia After Aneurysmal Subarachnoid Hemorrhage. Journal of Cerebral Blood Flow & Metabolism, 44(7), 1208-1217. https://doi.org/10.1177/0271678X241228512
  9. Mesina, A., Uryga, A., Giardina, A., Ciliberti, P., Battaglini, D., Patroniti, N., Czosnyka, M., Monnet, X., Cecconi, M., & Robba, C. (2024) The Effect of Passive Leg Raising Test on Intracranial Pressure and Cerebral Autoregulation in Brain Injured Patients: a Physiological Observational Study. Critical Care, 28 art. 23. https://doi.org/10.1186/s13054-023-04785-z
  10. Burzyńska, M., Uryga, A., Załuski, R., Goździk, A., Adamik, B., Robba, C., & Goździk, W. (2023) Cerebrospinal Fluid and Serum Biomarker Insights in Aneurysmal Subarachnoid Haemorrhage: Navigating the Brain-Heart Interrelationship for Improved Patient Outcomes. Biomedicines, 11(10), 2835. https://doi.org/10.3390/biomedicines11102835

Publications in conference proceedings:

  1. Uryga, A., Najdek, M., Najda, M., Mataczyński, C., & Buchner, T. (2024) Nonlinear Method to Assess Autonomic Modulation During Controlled Breathing. In: 2024 13th Conference of the European Study Group on Cardiovascular Oscillations (ESGCO), IEEE, Danvers, MA.
  2. Burzyńska, M., Uryga, M., Woźniak, J., Załuski, R., & Goździk, W. (2023) The Role of Biomarkers in the Development of the Delayed Cerebral Ischemia in Patients After Subarachnoid Haemorrhage from a Ruptured Aneurysm. In: Kübler, A., Goździk, W. (eds) XXXIV Konferencja Postępy w Anestezjologii i Intensywnej Terapii (p. 20), InspireCongress, Wroclaw. 
  3. Uryga, A., Mataczyński, C., Kasprowicz, M., & Burzyńska, M. (2023) Influence of Intracranial Pressure on Arterial Baroreceptor Reflex and Its Prognostic Significance in Patients with Traumatic Brain Injury. In: Kübler, A., Goździk, W. (eds) XXXIV Konferencja Postępy w Anestezjologii i Intensywnej Terapii (p. 21), InspireCongress, Wroclaw. 

International conference presentations:

  1. Uryga, A., Mataczyński, C., Najdek, M., Najda, M., & Buchner, T. Nonlinear method to assess autonomic modulation during controlled breathing. 13th International conference of the European Study Group on Cardiovascular Oscillations (ESGCO), 23–25.10.2024, Zaragoza, Spain. 
  2. Uryga, A., Kazimierska, A., Mataczyński, C., Czosnyka, M., Kasprowicz, M, & CENTER-TBI high-resolution sub-study participants and investigators. The relationship between autonomic nervous system activity and ICP pulse morphology in traumatic brain injury patients. 2024 Meeting of the International NeuroTrauma Society (INTS 2024), 2–5.09.2024, Cambridge, UK. 
  3. Mataczyński, C., Uryga, A., Pelah, A. I., Burzyńska, M., Robba, C., Czosnyka, M., & CENTER-TBI high-resolution sub-study participants and investigators. A machine learning approach to reveal temporal patterns between neuromonitoring signals and autonomic nervous system. 2024 Meeting of the International NeuroTrauma Society (INTS 2024), 2–5.09.2024, Cambridge, UK. 
  4. Uryga, A. Analysis of the relationship between the autonomic nervous system and cerebral autoregulation using machine learning approach. 12th International Meeting of the Cerebrovascular Research Network (CARNet 2023), 25–27.10.2023, Taipei, Taiwan.

National conference presentations:

  1. Najdek, M., Mataczyński, C., Burzyńska, M., & Uryga, A. The relationship between neuroparameters and autonomic nervous system metrics in traumatic brain injury patients using canonical correlation analysis. 12th Aspects of Neuroscience, 25–27.10.2024, Warsaw, Poland.
  2. Gut, A., Kazimierska, A., Kasprowicz, M., & Uryga, A. Analysis of the relationship between cerebral autoregulation and arterial baroreceptor sensitivity. 12th Aspects of Neuroscience, 25–27.10.2024, Warsaw, Poland.
  3. Uryga, A., Najda, M., Mataczyński, C., & Burzyńska, M. Explainable artificial intelligence in predicting paroxysmal sympathetic hyperactivity after traumatic brain injury. XIII Sympozjum Współczesna myśl techniczna w naukach medycznych i biologicznych, 27–28.09.2024, Wroclaw, Poland.
  4. Burzyńska, M., Uryga, A., & Urbański, P. Paroxysmal sympathetic hyperactivity in traumatic brain injury patients. XXI Międzynarodowy Zjazd Polskiego Towarzystwa Anestezjologii i Intensywnej Terapii, 12–14.09.2024, Gdansk, Poland.
  5. Uryga, A., Urbański, P., & Burzyńska, M. Autonomic nervous system and cerebral autoregulation disorders after traumatic brain injury. XXI Międzynarodowy Zjazd Polskiego Towarzystwa Anestezjologii i Intensywnej Terapii, 12–14.09.2024, Gdansk, Poland.
  6. Domczewska, W., Najda, M., & Uryga, A. Time-trend patterns in baroreflex sensitivity and intracranial pressure after traumatic brain injury. 70th International Medical Congress of Silesia (SIMC) 2024, 15–17.05.2024, Katowice, Poland.
  7. Pietroń, D., Najda, M., & Uryga, A. Reliability of the heart rate variability for assessing the autonomic nervous system in healthy volunteers. 70th International Medical Congress of Silesia (SIMC) 2024, 15–17.05.2024, Katowice, Poland.
  8. Berent, I., Najda, M., & Uryga, A. Machine-learning approach to mortality prediction after brain injury using volatility of neurosignals. 70th International Medical Congress of Silesia (SIMC) 2024, 15–17.05.2024, Katowice, Poland.
  9. Najdek, M., Najda, M., & Uryga, A. Joint symbolic analysis of blood flow velocity during controlled breathing in healthy volunteers. 70th International Medical Congress of Silesia (SIMC) 2024, 15–17.05.2024, Katowice, Poland.
  10. Uryga, A., Najda, M., Berent, I., Mataczyński, C., & Urbański, P. Exploring the relationship between autonomic nervous system activity assessed by phase-rectified signal averaging and cerebral autoregulation during controlled breathing. Jubileuszowa 30. Konferencja Asocjacji Elektrokardiologii Nieinwazyjnej i Telemedycyny Polskiego Towarzystwa Kardiologicznego, 10–13.04.2024, Zakopane, Poland.
  11. Uryga, A., & Kasprowicz M. Relationship between autonomic control of the cardiovascular system and cerebral hemodynamics. Jubileuszowa 30. Konferencja Asocjacji Elektrokardiologii Nieinwazyjnej i Telemedycyny Polskiego Towarzystwa Kardiologicznego, 10–13.04.2024, Zakopane, Poland.
  12. Berent, I., Najda, M., & Uryga, A. Analiza doboru hiperparametrów algorytmu „Phase rectified signal avergaing” (PRSA) na przykładzie kontrolowanego oddechu. XVI Interdyscyplinarna Konferencja Naukowa TYGIEL 2024, 21–24.03.2024, Lublin, Poland [PL].
  13. Berent, I., Uryga, A., & Kasprowicz, M. Użycie metody „phase-rectified signal averaging” do opisu zmian w układzie autonomicznym podczas hiperkapnii. X Ogólnopolskie Sympozjum Biomedyczne ESKULAP, 02.12.2023, Lublin, Poland [PL].
  14. Pietroń, D., Uryga, A., & Kasprowicz, M. Porównanie zmienności rytmu serca określonej na podstawie sygnału ciśnienia tętniczego krwi i elektrokardiografii u zdrowych ochotników. X Ogólnopolskie Sympozjum Biomedyczne ESKULAP, 02.12.2023, Lublin, Poland.
  15. Burzyńska, M., Uryga, A., Woźniak, J., Załuski, R., & Goździk, W. Rola biomarkerów w rozwoju opóźnionej strefy niedokrwiennej u pacjentów po krwotoku podpajęczynówkowym z pękniętego tętniaka–ujęcie statystyczne i z wykorzystaniem uczenia maszynowego. XXXIV Konferencja Postępy w Anestezjologii i Intensywnej Terapii, 24.06.2023, Wrocław, Poland [PL].
  16. Uryga, A., Mataczyński, C., Kasprowicz, M., & Burzyńska, M. Wpływ ciśnienia wewnątrzczaszkowego na odruch z baroreceptorów tętniczych i jego prognostyczne znaczenie u chorych z urazem czaszkowo-mózgowym. XXXIV Konferencja Postępy w Anestezjologii i Intensywnej Terapii, 24.06.2023, Wrocław, Poland [PL].