A multidisciplinary project at Umeå University aimed at developing state-of-the-art control and estimation techniques for studying the human cerebrospinal fluid (CSF) system, with application to the clinical diagnosis of hydrocephalus.

**Anders Eklund,**Associate Professor of Biomedical engineering, UmU, UmU Hospital,**Ian Manchester,**Post-doctoral Fellow in Control Systems, UmU,**Khalid Ambarki,**Post-doctoral Fellow in Biomedical Engineering, UmU,**Kennet Andesson,**PhD student in Biomedical Engineering, UmU, UmU Hospital,**Nina Andersson,**PhD student in Biomedical Engineering, UmU, UmU Hospital,**Anton Shiriaev,**Professor of Control Systems, UmU, NTNO.**Jan Malm,**Associate Professor of Neurology, UmU, UmU Hosptial,**Katrine Riklund,**Professor of Radiology, UmU, UmU Hosptial,**Noam Alperin,**Associate Professor of Medical Physics, UIC.

This work forms part of a large project, lead by Jan Malm, studying the causes, diagnosis, and treatment of hydrocephalus.

The cerebrospinal fluid provides physical support for the brain, and is believed to absorb and carry away toxic metabolic byproducts. It is generated and introduced to the brain cavity at an approximately constant rate, and it leaves the brain cavity by being reabsorbed into the bloodstream in the dural sinuses.

Circulation of the cerebrospinal fluid (light blue)

An obstruction to the normal flow can lead to a condition known as* hydrocephalus*, the symptoms of which can include gait imbalance, urinary dysfunction, and dementia. It is well established that these symptoms can be reduced or eliminated by implanting a shunt, a small valve with an attached drainage tube, into the patient’s skull, thereby altering the system and increasing the CSF flow.

Neurologists and neurosurgeons use a number of different diagnostic tests to decide whether or not the patient is likely to benefit from a shunt surgery. One important procedure is to estimate the *outflow resistance* of the CSF system using an *infusion test.* Artificial CSF is injected with a particular flow-rate pattern into the CSF system at the base of the spinal column - at the lumbar region - and the resulting CSF pressure variations are recorded. From this data, and a mathematical model of the fluid dynamics, an estimate of the outflow resistance can be obtained. Higher than average resistance values are believed to be an indicator of hydrocephalus.

Such tests are currently performed regularly at Umeå University Hospital, and many other hospitals around the world. However, all current procedures involve recording a lengthy data sets, and performing an off-line analysis of the data. Minimizing time of infusion tests is important for both patient comfort and efficient use of hospital resources, so a key aim of this project is provide real-time estimates to the clinical staff, so that the procedure can be stopped when a good estimate has been achieved.

The CSF system is an infinite-dimensional distributed parameter system, however it has been shown experimentally to be well approximated by a finite-dimensional lumped-parameter model, introduced by Anthony Marmarou, in which the intracranial pressure and other variables are assumed to be spatially invariant. The model takes the form of a first-order nonlinear differential equation.

Some remarks:

- Although the system is nonlinear, it takes the form of a
*Bernoulli equation,*and can be exactly linearized by way of a nonlinear change of variables, and linear systems techniques can be applied. - There exists physiological disturbances which can be very large in magnitude and have a highly uncertain spectrum, making them difficult to model using traditional frameworks, such as white noise filtered through a linear system. In fact, dealing with the disturbances is so far the most challenging aspect of this project.
- The artificial CSF infusion rate can be freely chosen. It should be chosen to maximize the information about the unknown parameters in the measured data. This is an
*optimal experiment design*problem.

There is much current research activity in studying the pulsatile dynamics by way of magnetic resonance imaging (MRI). In particular, the technique of phase-contrast MRI, which allows one to capture detailed images of flow velocities at approximately 30 frames per second.

There are two primary aims of this research:

- Basic science: to increase human understanding of the dynamical properties of the cerebrospinal fluid system,
- Diagnostic systems: to develop non-invasive techniques for diagnosis of conditions such as hydrocephalus.

There is a lot of work to be done in this field, and there is ample opportunity to create exciting projects for students.

For projects focusing on control and estimation aspects, please contact Ian or Anton. For projects focusing on biomedical engineering, please contact Anders. Suitable backgrounds of study would be engineering, mathematics, physics, or related fields.

K. S. Andersson, I. R. Manchester, N. Andersson, A. S. Shiriaev, J. Malm, A. Eklund,

Assessment of Cerebrospinal Fluid Outflow Conductance Using an Adaptive Observer - Experimental and Clinical Evaluation,

to appear in *Physiological Measurement*.

I. R. Manchester, K. S. Andersson, A. Eklund, A. S. Shiriaev, N. Andersson,

A Nonlinear Observer for On-line Estimation of the Cerebrospinal Fluid Outflow Resistance,

to appear in *Automatica*.

A. Eklund, P. Smielewski, I. Chambers, N. Alperin, J. Malm, M. Czosnyka, A. Marmarou,

Assessment of cerebrospinal fluid outflow resistance (Review Article)

*Medical and Biological Engineering and Computing, *45:719-735, 2007.

J. Malm, A. Eklund

Idiopathic normal pressure hydrocephalus.

*Practical Neurology*, 6:14–27, 2006.

N. Andersson, J. Malm, T. Backlund, A. Eklund

Assessment of cerebrospinal fluid outflow conductance using constant-pressure infusion-a method with real time estimation of reliability.

*Physiological Measurement* 26:1137–48, 2005.

**Conference:**

I. R. Manchester, K. S. Andersson, A. Eklund, A. S. Shiriaev,

Experimental Testing of a Method for On-Line Identification of the Cerebrospinal Fluid System,

*Proceedings of the 29th International Conference of the IEEE Engineering in Medicine and Biology Society*, Lyon, France, August 2007.

I. R. Manchester, K. S. Andersson, A. Eklund, A. S. Shiriaev,

Identifiability of the Parameters of a Nonlinear Model of the Cerebrospinal Fluid System,

*Proceedings of the 7th IFAC Symposium on Nonlinear Control Systems*, Pretoria, South Africa, August, 2007.

I. R. Manchester, K. S. Andersson, A. Eklund, A. S. Shiriaev,

Towards the Simultaneous On-line Estimation of the Compliance and Outflow Resistance of the Cerebrospinal Fluid System,

*Proceedings of the 2007 European Control Conference*, Kos, Greece, July 2007.

A. Shiriaev, N. Andersson, P.X. Miranda La Hera, M. Lilliehook, A. Eklund,

Nonlinear Observer Design for an On-line Estimation of the Cerebrospinal Fluid Outflow Resistance

*Proceedings of the 44th IEEE Conference on Decision and Control, 2005 and 2005 European Control Conference. CDC-ECC '05. *

Last updated 15th of August, 2007.