Patent Application
20130006074
In the science of photoplethysmography, light is used to illuminate or trans-illuminate living tissue for the purpose of providing noninvasive measurements of blood analytes or other hemodynamic parameters or tissue properties. In this monitoring modality light is directed into living tissue and a portion of the light which is not absorbed by the tissues, or scattered in some other direction, is detected a short distance from the point at which the light entered the tissue. The detected light is converted into electronic signals that are indicative of the received light intensity exiting the tissue. These signals, one for each emitter or spectral band of light incident on the living tissue (“living tissue” being monitored by photoplethysmography is referred to in this specification as the tissue-under-test), vary with the pulsation of the blood through the tissue-under-test. These time varying signals are referred to as photoplethysmographic signals. The photoplethysmographic signals are used to calculate blood analyte levels such as arterial blood oxygen saturation and/or hemodynamic variables such as heart rate, cardiac output, or tissue perfusion. Among the blood analytes that may be measured by photoplethysmography are various types of hemoglobin, including the percentages of oxyhemoglobin, carboxyhemoglobin, methemoglobin, and reduced hemoglobin in the arterial blood. A device which detects and processes photoplethysmographic signals to measure the levels of various blood analytes and/or various hemodynamic parameters is referred to as a photoplethysmographic measurement apparatus, photoplethysmographic device, or photoplethysmographic instrument.
The first widespread commercially-used photoplethysmographic device in medicine was the pulse oximeter, a photoplethysmographic device designed to measure arterial blood oxygen saturation. In conventional pulse oximetry two different bands of light are used, with each light band possessing a unique spectral content. In early pulse oximetry a tungsten lamp was used to generate broadband light and interference filters were used to isolate two narrow spectral bands of light. Later pulse oximeters replaced the tungsten lamp source and interference filters with two light emitting diodes (LEDs) to generate the desired spectral bands. More recently photoplethysmographic instruments have been developed in which more than two light bands are utilized to allow the measurement of a larger number of blood analytes, including such blood analytes as oxyhemoglobin, carboxyhemoglobin, methemoglobin, and reduced hemoglobin.
For accurate photoplethysmographic measurement of blood analytes it is crucial that the light from all emitters enters the tissue-under-test through the same small aperture. Ideally the light from each emitter is also evenly distributed throughout the aperture. This homogeneous distribution of the light, regardless of the spectral band, emitted through a small aperture, in combination with a small detector aperture, ensures that the light from each emitter is traversing essentially the same sample of tissue-under-test. Conversely, if the light from different emitters can travel different paths through the tissue-under-test then the received pulsatile signal amplitude related to each individual emitter may vary due to the differences in the microvasculature through which the light travels. Any lack of commonality in the optical path from one emitter to another translates directly into measurement error.
The problem of generating a homogeneous distribution of light that is emitted through a small aperture into the tissue-under-test has existed since the earliest days of LED-based pulse oximetry. Given the physical separation of the LEDs in a typical pulse oximetry sensor, the light from each of the emitters could, on average, take a slightly different path though the tissue-under-test. In an attempt to minimize this effect as much as possible, all manufacturers of pulse oximeters positioned the LEDs in the sensors as closely together as possible. Often a diffuser was also added, typically placed some distance from the LEDs, in an attempt to further homogenize the light from the two sources.
The use of a diffuser was disclosed in U.S. Pat. No. 4,407,290 where the specification states that “ . . . the light emitted from LEDs 39 and 40 being preferably directed through a light diffusing disc 42 to the blood containing sample 45 to be tested, which sample may be, for example, tissue such as an ear lobe or the like.” This use of a diffuser in the sensor design was implemented in the early commercial Ohmeda Biox ear and finger sensors.
The use of a diffuser to homogenize and properly distribute the light over the output aperture was also expressed in U.S. Pat. No. 6,115,621 wherein the specification explains that “One type of oximeter sensor will add a diffusing optic to diffuse the light emitted from the light-emitting diodes (LEDs) . . . ” and then proceeds to refer back to the use of the diffuser shown in U.S. Pat. No. 4,407,290. The specification for U.S. Pat. No. 6,115,621 goes on to state “Also shown is an optional optical diffuser 80 for diffusing the light from emitter 76, which causes a further spreading or mixing of light and may enhance the amount of tissue penetrated in some instances.”
A non-photoplethysmographic technology that uses multiple spectral bands of light to measure tissue oxygen saturation is revealed in U.S. Pat. No. 6,594,513. This monitoring modality also recognized the need for homogeneous light entering the tissue-under-test and used a “mixer” which “mixes the light from multiple lasers with different wavelengths in order to enter it homogeneously into the scalp from a single aperture 18.”This particular diffuser used a “clear material, such as a stiff silicone gel, containing white, (i.e. non-absorbing and wavelength independent) scattering particles, such as titanium dioxide powder, which disperses the light entering the mixer.”
Similarly, US patent application publication number 2008/0242958 recognizes the need for proper diffusion of the light incident on the tissue-under-test. This application states that “in various embodiments a diffuser scatters the radiated light so that a tissue site is uniformly illuminated across all of the wavelengths.”
An MRI compatible photoplethysmographic sensor is disclosed in the US patent application publication number US 2010/0113902. In this sensor the light sources, or emitters, are optical fibers which deliver the light from LEDs to the sensor. The applicant states that the sensor may include “ . . . a diffuser between the fiber optic material and the animal tissue.”
In these five references the diffuser is substantially positioned against the skin some distance from the LEDs or light sources (or, in one case, filling the space from the light sources to the output aperture). This configuration has a problem in that it does not fully homogenize the light across the output aperture. (The output aperture, or exit aperture, being the surface area through which the light is transmitted from the sensor into the tissue-under-test. The aperture is also intended to block light from exiting from any other point on the sensor and entering the tissue-under-test.) Mapping the intensity of the individual light sources as a function of position on the output side of the aperture would show a non-uniform intensity profile across the output aperture and, worse yet from the standpoint of measurement accuracy, the intensity mapping would vary from one light source to another. Ideally fully homogenized light, intended for illumination of the tissue-under-test, would not exhibit any variation in output intensity as a function of position of measurement on the output surface of the output aperture, regardless of which light source was being measured.
A second problem with the method of mixing the light defined in these prior art references, and seen in the early Ohmeda Biox sensors, is in the actual design of the diffusing element. To the extent that the diffusers are made up of scattering (or light reflecting) elements randomly suspended in a clear transmitting medium, the light transmission losses are relatively large. Furthermore, the more effective this type of diffuser is at mixing the incident light, the worse the optical throughput. This loss in light intensity out of the sensor has several detrimental effects. It means either degradation in signal-to-noise level or that the light sources must emit more light to ensure sufficient optical signal strength after the light passes through the tissue-under-test. This typically means more input power to drive the LEDs or other light generating elements such as lasers or tungsten lamps. Higher input power results in shorter lifetimes of the light generating elements, more heat generation, and even shorter operating times for battery-powered devices.
US patent application publication number 2005/0228253 reveals a different method of mixing the light where a set of input fiber bundles carries light from each of a number of individual light generating elements to an N-to-1 fiber coupler where all the input fibers are coupled into a single larger output fiber. At the N-to-1 fiber coupler the individual fibers, originating from any given light generating element, are spatially distributed evenly throughout the N-fiber bundle thereby, to some extent, homogenizing the light into the single output fiber. This solution suffers from the problem of incomplete homogenization of the light as the mixing is limited by the discrete nature and size of the input fibers. Furthermore this solution is high in cost, requires a great deal of physical space, and, potentially, a great many optical fibers. And, finally, the large diameter of the single output fiber in the coupler creates a cable that is inflexible and prone to breakage during clinical use.
Whereas conventional pulse oximetry sensors have had the problem of mixing only two light sources together, more advanced photoplethysmographic devices are now using many more emitters to measure ever more blood analytes. In addition, some of these analytes have lower optical absorption thus requiring more sensitive and accurate photoplethysmographic instruments to make measurements with clinically-meaningful resolution, precision, and accuracy. A crucial element in the design of an accurate photoplethysmographic measurement apparatus is a light homogenizing apparatus that puts out well homogenized light, from multiple emitters, at a low cost, and with relatively high optical efficiency (high optical efficiency being a low loss from optical power-in to optical power-out). An additional design requirement of the light homogenizing apparatus may also be that it is physically small in size to allow it to be integral to the sensor, or sensor cable, where it could be positioned at, on, or near, the tissue-under-test.
In accordance with one embodiment a light homogenizing apparatus for a photoplethysmographic device comprises one or more light guides conducting light from at least one laser wherein the light exiting the light guides is incident on a light mixing element. The light mixing element comprises a plurality of one or more light sources incident upon a diffusing element followed by an internally reflective light guide. Accordingly, several advantages of one or more aspects are as follows: that light exiting the light mixing element is a substantially homogeneous mixture of the light that was incident on the light homogenizing apparatus; and that the apparatus provides a high optical throughput. The combination of these advantages contributes to accurate, high-resolution photoplethysmographic measurement.
a. Basic light homogenizing apparatus.
b. Basic light homogenizing apparatus. Side view with representative optical ray trace.
a. Light homogenizing apparatus with folded optical path.
b. Light homogenizing apparatus with folded optical path. Side view with representative optical ray trace.
a. Light homogenizing apparatus with folded optical path and integral diffractive element on prism.
b. Light homogenizing apparatus with folded optical path and integral diffractive element on prism. Side view with representative optical ray trace.
One embodiment of a light homogenizing apparatus for a laser-based photoplethysmographic device is shown in
In these two figures light is conducted down a set of one or more light guides 110 toward a light mixing element. The light mixing element is made up of the combination of a diffuser 130 and an internally reflective light guide 160. Alternatively, or in addition to the light guides 110, light may also be generated by one or more discrete light sources 120. Light 170 exiting the light guides 110 or light 165 being emitted by the discreet light sources 120 is incident upon the diffuser 130 portion of the light mixing element. The diffuser 130 increases the angular dispersion of light (165 and 170) entering the internally reflective light guide 160. Finally, homogenized light 180, a well mixed version of the input light 165 and 170, exits the internally reflective light guide.
The internally reflective light guide 160 is a combination of a light pipe 150 and a reflector 140 that reflects light incident upon it back into the light pipe 150. The reflector 140 substantially surrounds all surfaces of the light pipe 150 with the exception of the input and the output surfaces as shown in the side view of the apparatus in
It is this homogenized light 180 exiting the internally reflective light guide 160 portion of the light mixing element that would then be made incident upon a tissue-under-test for use in'photoplethysmographic measurement. Ideally output light 180 is a completely homogeneous mixture of light from all of the input emitters, regardless of whether those emitters are the one or more light guides 110 or the one or more discrete light sources 120 or any combination of the two.
A homogeneous mixture refers to a distribution of light intensity such that, at any point on the output surface (or the output side of the output aperture) of the light mixing element 160, the light does not vary in intensity as a function of position on that output surface. Ideally a high degree of homogeneity would be maintained whether any one individual emitter is turned on or any combination of emitters is turned on.
Functionally, at least one of the light sources is a laser source. In addition to a laser source, other types of light sources can be used concurrently in the apparatus. These could include, for example, light emitting diodes, tungsten or other filament type lamps, or gas lamps. The laser source could be a semiconductor laser, a gas laser, or nearly any other type of laser where the laser light can be directed or launched into a light guide 110 or where the laser can act as a discrete light source 120. The light guide 110 can be any of a number of different elements that conduct light from one point to another including glass or plastic optical fibers, a liquid light guides, plastic light pipes, or other such elements.
Light emitted from the light guides 110 or from the discrete light sources 120 is then incident upon a diffuser 130. While
The diffuser 130 then further disperses (i.e. the scatters and spreads in different directions) the light as is indicated by the ray trace shown in
One way in which the internally reflective light guide 160 can function is to utilize the principle of total internal reflection (TIR). In this case the internally reflective light guide 160 would not require reflector 140. This TIR method will work efficiently only if light entering the light pipe 150 is at an angle less than the NA of the light pipe 150. TIR can also be used if the optical power budget of the system can afford the loss of any light entering the internally reflective light guide 160 at an angle greater than the NA of the internally reflective light guide 160.
If the angle of portions of the light entering the internally reflective light guide 160 is greater than the angle that can be supported by TIR, a reflector 140 can be added to the light pipe 150. This allows a high dispersion angle of light to be created by the combination of the spatial irradiance pattern of the emitters 110 and 120 and the diffuser 130. High angle light rays are then reflected by the reflector 140 back into the light pipe 150. Note that internally reflective light guide 160 can be a hollow cylinder (or other hollow shape), a glass or plastic light guide, or a liquid light guide, forming the light pipe 150 surrounded by the reflector 140.
The internally reflective behavior of the internally reflective light guide 160 maximizes the homogenizing effect of this apparatus and is shown graphically by the ray trace lines in
An additional benefit of this apparatus is that it can also simultaneously provide a fixed output aperture. The reflector 140 blocks any light from escaping the light pipe 150 except where it is not surrounded by the reflector 140. By designing the sensor so that the internally reflective light guide 160 is positioned against the tissue-under-test, the reflector 140 also provides the function of creating a fixed output aperture for homogenized light 180 exiting the light homogenizing apparatus.
To further improve the optical performance of the light homogenizing apparatus one can optimize the design of the diffuser itself. In prior art diffusers used in photoplethysmographic instrumentation, a scattering medium is suspended in an otherwise optically clear substrate material to “diffuse” or scatter the light in all directions. Such diffusers, by their very nature, also scatter light back toward the light sources, thus making them inefficient with low optical throughput. An alternative solution that further enhances the optical throughput of the light homogenizing apparatus is to replace the conventional scattering-based diffuser with a patterned surface diffuser.
Patterned surface diffusers are essentially a diffractive element engineered to diffuse light in a predetermined manner. A patterned surface diffuser disperses light in an engineered manner by providing an engineered pattern or micro structure embedded in one surface of the diffuser. Some of the advantages of a patterned surface diffuser include high optical throughput and controlled dispersion of the input light.
“Controlled dispersion” refers to the ability to select the angular dispersion of light exiting the diffuser. For example, a diffuser can be engineered to cause light incident upon it, at an angle perpendicular to its diffusing surface, to be distributed evenly over a predetermined solid angle. For example, if diffuser 130 in
Using a patterned surface diffuser, and selecting the length and cross sectional profile of the light pipe 150, the designer can control the maximum number of reflections off the reflector 140. In
An additional advantage to the patterned surface diffuser is that the patterned surface diffuser 130 can be integrated into the light pipe 150; i.e. the patterned surface diffuser can be constructed on the input end of the light pipe 150. This integrated design has the potential to be less expensive than individual elements, easier to manufacture, and eliminates any possible air gap between the two elements that would create Fresnel reflections and reduce the optical throughput of the apparatus.
If the diffuser 130 is not integrated into the light pipe 150, any potential air gap between the diffuser 130 and the light pipe 150 can alternatively be eliminated by using an optically clear adhesive or index-matching material between the two components to minimize changes in index of refraction that cause Fresnel reflections as the light travels through the system.
a, an isometric view, and
As shown in
The 45 degree surface of the prism 210 may be reflective based only on the principle of total internal reflection or it may be coated with a reflector to allow any light incident upon that surface (at any angle with respect to the normal of that 45 degree angle surface) to be internally reflected. It should be noted that the prism 210 could alternatively be replaced with a mirror, or mirrored surface, placed at approximately the same position as the 45 degree angle surface of prism 210 to achieve the same optical behavior.
The prism 210, as would a mirror, allows light from the two different types of emitters 110 and 120 to be introduced into the system from two different orientations as illustrated in
Other than having a folded optical path, the design shown in
The reflector 140 may also function to provide the aperture for the homogenized output light 180. A well controlled and relatively small optical aperture for light entering the tissue-under-test is preferred as it increases the accuracy of photoplethysmographic measurements in living tissue.
The embodiment of the optical design of the light homogenizing apparatus shown in isometric view in
The increase in throughput, or optical efficiency, of this embodiment results from, at a minimum, the elimination of two changes in index of refraction with resulting Fresnel reflection losses that exist in the light paths of the previous two embodiments shown in
Laser-based light sources are now being employed in photoplethysmographic instruments to improve measurement accuracy and to further increase the number of blood analytes that can be effectively and accurately measured by this monitoring modality. But when laser light travels down a light guide, often the distribution of the laser light within the light guide is non-uniform, particularly if the light guide is a multimode optical fiber. This is referred to as a non uniform modal distribution. The use of laser light in photoplethysmography therefore further increases the need to homogenize the light before it is incident on the tissue-under-test.
In past generations of LED-based photoplethysmographic devices a high degree light mixing was not as important. Simply placing the two LED die in very close proximity to one another in the emitter assembly, and providing a single aperture for the output light, was sufficient. In the earliest pulse oximetry sensors, when LEDs were not yet used in die form and pre-packaged LEDs were used in the sensor emitter assembly, a simple diffuser, encircled by an output aperture and placed a short distance from the LEDs, provided sufficient light mixing for the relatively crude measurement of oxygen saturation.
As newer generations of photoplethysmographic devices attempt to make ever more accurate measurements of multiple blood analytes, a more homogeneous mixture of light must be launched into the tissues to ensure that the light from all light sources follows essentially the same optical path through the tissue. The apparatus of this invention supplies this required homogeneity, in part, by following the diffusing element 130 (or the 45 degree diffusing surface of prism 310) with the internally reflective light guide 160.
The previous discussion of the embodiments has been presented for the purposes of illustration and description. The description is not intended to limit the invention to the form disclosed herein. Variations and modifications commensurate with the above are considered to be within the scope of the present invention. The embodiments described herein are further intended to explain the best modes presently known of practicing the invention and to enable others skilled in the art to utilize the invention as such, or in other embodiments, and with the particular modifications required by their particular application or uses of the invention. It is intended that the appended claims be construed to include alternative embodiments to the extent permitted by the prior art.
This invention was made with government support under R44 HL073518 awarded by the National Institutes of Health. The government has certain rights in the invention.
