« PrécédentContinuer »
United States Patent [w]
Nathel et al.
US006015969A [ii] Patent Number:  Date of Patent:
SPECTROSCOPIC QUANTITATION OF
LIGHT-ABSORBING SPECIES IN
 Inventors: Howard Nathel, Albany; Harry E.
Cartland; Billy W. Colston, Jr., both of Livermore; Matthew J. Everett, Pleasanton; Jeffery N. Roe, San
Ramon, all of Calif.
 Assignee: The Regents of the University of California, Oakland, Calif.
 Appl. No.: 09/008,234  Filed: Jan. 16, 1998
Related U.S. Application Data
 Continuation-in-part of application No. 08/714,745, Sep. 16, 1996, abandoned.
 Int. CI. GO IB 9/02
 U.S. CI 250/227.27; 250/339.12;
 Field of Search 250/227.27, 227.19,
250/221, 339.12, 339.09, 339.07, 340, 341.5;
356/345, 357, 360
 References Cited
U.S. PATENT DOCUMENTS
5,459,570 10/1995 Swanson et al 356/345
An oxygen concentration measurement system for blood hemoglobin comprises a multiple-wavelength lowcoherence optical light source that is coupled by single mode fibers through a splitter and combiner and focused on both a target tissue sample and a reference mirror. Reflections from both the reference mirror and from the depths of the target tissue sample are carried back and mixed to produce interference fringes in the splitter and combiner. The reference mirror is set such that the distance traversed in the reference path is the same as the distance traversed into and back from the target tissue sample at some depth in the sample that will provide light attenuation information that is dependent on the oxygen in blood hemoglobin in the target tissue sample. Two wavelengths of light are used to obtain concentrations. The method can be used to measure total hemoglobin concentration [Hb^ +Hb ] or total blood volume in tissue and in conjunction with oxygen saturation measurements from pulse oximetry can be used to absolutely quantify oxyhemoglobin [Hb02] in tissue. The apparatus and method provide a general means for absolute quantitation of an absorber dispersed in a highly scattering medium.
16 Claims, 2 Drawing Sheets
SPECTROSCOPIC QUANTITATION OF
LIGHT-ABSORBING SPECIES IN
This application is a continuation-in-part of U.S. patent application Ser. No. 08/714,745, filed Sep. 16, 1996, now abandoned.
NOTICE OF GOVERNMENT INTEREST
The United States Government has rights in this invention pursuant to Contract No. W-7405-ENG-48 between the United States Department of Energy and the University of 15 California for the operation of Lawrence Livermore National Laboratory.
BACKGROUND OF THE INVENTION
1. Field of the Invention 20 The present invention relates to quantitative spectroscopy
in turbid media or highly scattering media and more particularly to absolute measurements of various blood constituents in living tissue by non-invasive, non-harmful methods. 25
2. Description of Related Art
Near infrared radiation with wavelengths of 600-1400 nanometers passes easily through living tissue. However, these same wavelengths are variously affected by tissue 30 oxyhemoglobin concentration, e.g., on the basis of hemoglobin absorbance. The overall range is limited in wavelength, e.g., on the long wavelength side of the spectrum, longer than 1400 nanometers, by water absorption, and on the short wavelength side of the 35 spectrum, shorter than 600 nanometers, by blood absorption. Between these higher and lower limits, the light that does penetrate the tissue is highly diffuse due to scattering. Such diffusion can otherwise obscure information that could be extracted from the non-scattered light. See E. M. Sevick et 40 al., "Quantitation of Time- and Frequency-Resolved Optical Spectra for the Determination of Tissue Oxygenation," Analytical Biochemistry 195, pp. 330-351 (1991).
Optical diagnostic systems have been built to take advantage of the near infrared translucence of living tissue, but 45 these prior art systems are seriously handicapped by the photon scatter that occurs within the highly diffuse tissue. One of the earliest used optical techniques, called pulse oximetry, was only able to provide estimates of the oxygen saturation of blood, e.g., by using the phenomenon of 50 differential transmission of light caused by oxyhemoglobin and reduced hemoglobin. Saturated oxygen (Sa02) is defined as the percentage of oxygen bound to hemoglobin compared to the total hemoglobin available for reversible oxygen binding. Unfortunately, with pulse oximetry the 55 absolute concentration of free oxygen in the blood could not be discerned, because it has no NIR signatuare. Only the ratio of oxyhemoglobin to total hemoglobin can be determined through human tissue.
Quantitative spectroscopy through tissue with optical 60 radiation is facilitated by the use of scatter elimination techniques, which fix the photon path length. Measurements of the attenuation due to the material of interest in the medium is difficult without a means to discriminate the non-scattered-photons from the scattered-photons, because 65 the amount of medium involved (i.e. pathlength) is indeterminate. In useful applications, the exact path lengths must be
determined for sub-surface light penetrations of tissue that range up to several millimeters.
Time-domain and frequency-domain methods can be used for the discrimination of light that has undergone considerable scattering to selectively detect the non-scattered, first arriving photons. Scattered photons necessarily travel over longer distances and take more uncertain pathways than do ballistic or quasi-coherent photons. The non-scattered photons traverse much shorter path lengths and exit the medium in a small, forward cone. The best quantitative information is carried in the photons that are relatively non-scattered, and these arrive first at the detector from the medium. Timeresolved techniques have conventionally been used to discriminate between scattered and non-scattered light exiting tissues based on time-of-fiight. Optical coherence techniques rely on the short coherence length of a broadband low-coherence light source to provide time-of-flight information interferometrically via autocorrelations. Measurements are therefore restricted to relatively non-scattered, first arriving (i.e. ballistic) photons.
Time domain techniques, such as streak cameras, require sub-picosecond laser systems which are expensive, noncompact, and complicated. Frequency-domain techniques, however, use inexpensive optical sources, optical lowcoherence refiectometry (OLCR), and avoid the need for complicated systems. State-of-the-art refiectometers use diode light sources and fiberoptics that make for compact and modular systems that are capable of micrometer spatial resolutions and high detection sensitivities.
The relative transparency of biological tissues to near infrared (NIR) light allows the absorption properties of intact organs to be monitored non-invasively. The NIR absorption caused by hemoglobin and cytochrome oxidase can be measured and used to monitor changes in blood and tissue oxygenation. Such measurement methods were first applied to the brain of cats and subsequently to the brains of newborn infants and adults. Recently, methods for the absolute quantitation of cerebral blood flow and blood volume have been developed and applied to newborn infants and adults. The possibility of imaging of tissue oxygenation by NIR light has also been studied by various groups.
Quantitative interpretation of spectroscopic data using the Beer-Lambert law requires that the optical pathlength be known, otherwise the light intensity measurement is meaningless because the distance over which it was attenuated is unknown. At best, the prior art only approximates the pathlength. In near infrared spectroscopy (NIRS), light scattering by the tissues prevents detecting all the light that entered the tissues. The source light travels along a distribution of paths. It has, however, previously been shown that a modified Beer-Lambert law can be applied to quantify changes in chromophore concentration from the measured changes in tissue attenuation. This modified law uses the differential pathlength, which is defined as the local gradient of the attenuation versus the absorption coefficient fia of tissue. It has been shown experimentally that the differential pathlength can be approximated by measuring the mean distance L traveled across the tissue by picosecond light pulses or by measuring the phase shift of a frequency modulated light source. The differential pathlength factor which is obtained when the mean pathlength <L> is divided by the distance between light source and detector optrodes, has been shown to be approximately constant once the optrode spacing exceeds 2.5 cm.
To date, the use of differential pathlength factors have only been demonstrated to be valid for homogeneous medi