Detection of CT occult aneurismal subarachnoid hemorrhage using a novel spectrophotometric analysis of cerebral spinal fluid

In North America, approximately 30,000 people annually suffer an aneurismal subarachnoid hemorrhage (SAH). Using computerized tomography (CT), the blood is generally not visible after 12 hours. Currently lumbar puncture (LP) results are equivocal for diagnosing SAH largely because of technical limitations in performing a quick and objective evaluation. Having ruptured once, an aneurysm is statistically more likely to rupture again. Therefore, for those individuals with a sentinel (or warning) hemorrhage, detection within the first 12 hours is paramount. We present a diagnostic technology based on visible spectroscopy to quickly and objectively assess low-blood volume SAH from a diagnostic spinal tap. This technology provides clinicians, with the resources necessary for assessing patients with suspected aneurismal SAH beyond the current 12-hour limitation imposed by CT scans. This aids in the improvement of patient care and results in rapid and appropriate treatment of the patient. To perform this diagnosis, we quantify bilirubin and hemoglobin in human CSF over a range of concentrations. Because the bilirubin and hemoglobin spectra overlap quantification is problematic. To solve this problem, two algorithmic approaches are presented: a statistical or a random stochastic component known as Partial Least Square (PLS) and a control theory based mathematical model. These algorithms account for the noise and distortion from blood in CSF leading to the quantification of bilirubin and methemoglobin spectroscopically. The configurations for a hardware platform is introduced, that is portable and user-friendly composed of specific components designed to have the sensitivity and specificity required. This aids in measuring bilirubin in CSF, hemorrhagic-CSF and CSF-like solutions. The prototype uses purpose built algorithms contained within the platform, such that physicians can use it in the hospital and lab as a point of care diagnostic test.

[1]  H. Abdi.  Partial Least Square Regression PLS-Regression , 2007 .

[2]  Jennifer L. Dungan,et al.  Visualizing spatially varying distribution data , 2002, Proceedings Sixth International Conference on Information Visualisation.

[3]  J. Friedman,et al.  [A Statistical View of Some Chemometrics Regression Tools]: Response , 1993 .

[4]  S. Haykin,et al.  Adaptive Filter Theory , 1986 .

[5]  Hilko van der Voet,et al.  Comparing the predictive accuracy of models using a simple randomization test , 1994 .

[6]  E. Lentjes,et al.  Iterative model for the calculation of oxyhemoglobin, methemoglobin, and bilirubin in absorbance spectra of cerebrospinal fluid. , 2001, Clinical chemistry.

[7]  James Caffery,et al.  Derivative spectrophotometric analysis of cerebrospinal fluid for the detection of a ruptured cerebral aneurysm , 2003, SPIE Optics + Photonics.

[8]  James Caffery,et al.  Analysis of cerebrospinal fluid for detection of ruptured cerebral aneurysm using spectrophotometry and signal processing techniques , 2004, SPIE BiOS.

[9]  J. Gill,et al.  Comparative studies of the diagnostic value of cerebrospinal fluid spectrophotometry and computed tomographic scanning in subarachnoid hemorrhage. , 1985, Neurosurgery.

[10]  Lennart Ljung,et al.  System Identification: Theory for the User , 1987 .

[11]  James Caffery,et al.  Development of a diagnostic system for bilirubin detection in cerebral spinal fluid , 2004, SPIE Optics + Photonics.

[12]  G. Bots,et al.  Blood-stained cerebrospinal fluid: traumatic puncture or haemorrhage? , 1981, Journal of neurology, neurosurgery, and psychiatry.

[13]  S. A. Iversen.  CSF spectrophotometry in the diagnosis of subarachnoid haemorrhage , 2002, Journal of clinical pathology.

[14]  Y. Ho,et al.  An approach to the identification and control of linear dynamic systems with unknown parameters , 1963 .

[15]  Stephen A. Book.  Statistics : basic techniques for solving applied problems , 1977 .

[16]  Svante Wold,et al.  Personal memories of the early PLS development , 2001 .

[17]  S. Wold,et al.  PLS-regression: a basic tool of chemometrics , 2001 .

[18]  P. Eykhoff.  System Identification Parameter and State Estimation , 1974 .