Patent Application
20230233075
The invention relates to methods and devices for carrying out fluorescence lifetime microscopy on an eye.
Numerous techniques have been applied for characterizing the human eye.
One of them is Optical Coherence Tomography (OCT), which allows to locate and characterize structures within the eye, see e.g. EP 3572765.
Fluorescence Lifetime Imaging Microscopy FLIM is one of many other techniques used for characterizing the eye, e.g. the retina, see e.g. WO 2015/072964.
The problem to be solved by the present invention is to provide higher quality fluorescence lifetime microscopy data for an eye.
This problem is solved by the method and device of the independent claims.
Accordingly, in one aspect, the invention relates to a method for carrying out fluorescence lifetime microscopy on a first structure of an eye, such as the retina. This method comprises at least the following steps:
This method is based on the understanding that it is possible to account for the influence of the second structure in the overall raw fluorescence data by providing an estimate of its fluorescence and using said estimate to correct the raw fluorescence data.
Advantageously, for good spatial resolution of the first structure, the method comprises the step of performing a first time-resolved fluorescence measurement with the probe beam focused on the first structure. The raw fluorescence data is then derived from this first measurement.
In addition, the estimated fluorescence data may then e.g. be derived from at least one second time-resolved fluorescence measurement with the probe beam focused on the second structure. This is based on the understanding that, by focusing the probe beam on the second structure, the signal will be more sensitive to the second structure than in the first measurement. Hence, the second measurements can be used to derive the estimated fluorescence data for the second structure.
The order of the first and the second time-resolved fluorescence measurement is arbitrary, i.e. the second time-resolved fluorescence measurement may take before or after the first time-resolved fluorescence measurement.
At least one of the two time-resolved fluorescence measurements advantageously includes determining at least one parameter indicative of a decay time of the fluorescence.
The raw fluorescence data can be measured for at least two locations (i.e. for several locations) of the first structure, in particular for different x- and y-locations (with x and y denoting directions perpendicular to the optical axis z of the eye and the microscope device).
In that case, the “estimated fluorescence data” may be one of the following:
a) The estimated fluorescence data may be the same for both locations. In this case, it is assumed that the second structure (such as the lens of the eye) adds substantially the same fluorescence contribution at both measurement locations.
b) The estimated fluorescence data is different between the first and the second location. In this case, it is assumed that the fluorescent properties of the second structure are sufficiently inhomogeneous to warrant a spatially resolved correction.
In case b), spatially resolved estimated fluorescence data may be obtained by performing a plurality of the “second” time-resolved fluorescence measurements with the probe beam focused on different parts of the second structure.
Also in case b), the method may comprise the step of determining the two parts of the second structure that the probe beam interacts with when being focused on the two locations of the first structure. This allows using the estimated fluorescence data attributed to these two (usually different) parts. The determination of the two parts can e.g. be carried out by means of ray tracing calculations.
In an advantageous embodiment, a device equipped for OCT (=Optical Coherence Tomography) and time-resolved fluorescence measurements is used for said fluorescence lifetime microscopy, and the method comprises the step of carrying out OCT measurements on the first and/or second structures with said device. This allows to complement the time-resolved fluorescence measurements with spatially well correlated data on the 3D structure of the eye.
Advantageously, the OCT measurements are carried out by means of a measurement beam collinear to the probe beam used for the fluorescence measurements. This provides an even better spatial correlation between the two types of measurements.
In one embodiment, OCT data obtained from the OCT measurements may be used for calculating the estimated fluorescence data. This allows to take e.g. the geometry of the second structure into account when estimating its fluorescence. For example, at least one of the following parameters of the second structure may be taken into account: its thickness along the axis of the eye, its volume, its extension perpendicular to the axis of the eye, or its position.
In an advantageous embodiment, the first structure is the retina of the eye and/or the second structure is the lens of the eye.
In another embodiment, the second structure may be the retina of the eye and/or the first structure is the lens of the eye.
The invention also relates to a microscope device for carrying out fluorescence lifetime microscopy on a first structure of an eye, comprising at least the following elements:
In a second aspect, the invention also relates to a microscope device for carrying out fluorescence lifetime microscopy of an eye. This device comprises at least the following elements:
This device allows to measure a combination of time-resolved fluorescence data and OCT data, thereby allowing to accurately relate the two types of data with each other.
Advantageously, the “fluorescence lifetime parameters” include a value indicative of at least one decay time of the fluorescence.
The device may further comprise a beam combiner arranged to collinearly combine the (fluorescence) probe beam and the (interferometric) measurement beam, i.e. to make these two beams spatially concentric and propagating into the same direction. This allows an even better spatial correlation between the two measurement methods.
The OCT measurements allow to measure the distance between the device and at least one part of the eye. In all aspects of the invention, this distance can be used for one or both of the following purposes:
a) To compensate a time offset in the fluorescence data, such as the raw fluorescence data, the estimated fluorescence data, and/or the corrected fluorescence data: This allows to accurately predict the “start time” of the fluorescence data and/or to compensate for changes in the device-eye-distance between consecutive measurements.
b) To enable a fluorescence measurement only if said distance is in a given range. This allows to disable measurements when the distance is inappropriate.
The method may further comprise the steps of measuring said distance for at least two subsequent fluorescence measurements and mutually offsetting, in time, the two fluorescence measurements as a function of the change of said distance in between the two fluorescence measurements.
The part of the eye to be used for the distance measurement is e.g. the first or the second structure of the eye, which allows to obtain the absolute “zero point” of the fluorescence data from the respective structure. If, however, only relative changes of the distance need to be tracked, any part of the eye can be used.
The invention will be better understood and objects other than those set forth above will become apparent when consideration is given to the following detailed description thereof. Such description makes reference to the annexed
Device Overview
For carrying out OCT measurement, the device comprises an optical coherence tomography interferometer 8.
Interferometer 8 has a light source 10, which, in the present embodiment, is a swept-source light source, i.e. it generates narrowband light that can be adjusted in wavelength.
The light from light source 10 passes a beam splitter 12, in particular a fiber beam splitter, and is sent into two interferometer arms 14, 16.
The first arm is the reference arm 14, which comprises one or more mirrors, in the present case two mirrors 18a, 18b, at one end. Light impinging on the mirrors 18a, 18b is sent back into beam splitter 12 and from there, at least in part, to a light detector 20.
In the shown embodiment, reference arm 14 is provided with an optical switch 19 and, as mentioned, two mirrors 18a, 18b at different positions, Switch 19 can be controlled to send light either to mirror 18a or 18b, with the two settings corresponding to different lengths of reference arm 14. Alternatively, it is e.g. possible to use a single mirror with adjustable position.
By changing the length of reference arm 14, the measurement range of interferometer 8 can be adjusted to measure the anterior or the posterior part of eye 30.
In yet another embodiment, the length of reference arm 14 is not adjustable. In this case, interferometer 8 must have a sufficient range of measurement to scan all parts of the eye that are relevant in the present context, e.g. by using a light source 10 that is sufficiently narrow-width and a detector able to sample the interference signal at sufficiently small wavelength intervals.
The second arm of interferometer 8 is the sample arm 16. It comprises collimation optics 22 for collimating the measurement light coming from beam splitter 12. The light is then fed through two scanner mirrors 24a, 24b and an objective lenses 26a, 26b for generating a measurement beam 28. Depending on the position of the scanner mirrors 24a, 24b, measurement beam 28 can be laterally offset in an x-y-plane perpendicular to the optical axis z of the device.
In the embodiment of
Measurement beam 28 enters eye 30, where it is reflected or scattered by the structures of the eye. Light cast back from such structures is returned to beam splitter 12 and from there, at least in part, to light detector 20, where it can interfere with the light from reference arm 14.
For OCT measurements, the device of
Spectral analysis, in particular a Fourier transform, of the signal from detector 20 can then be used for generating the reflection values of eye 30 along axis z for the given A-scan.
This type of OCT measurement is known to the skilled person, and it is e.g. described in EP 3572765 and the references cited therein.
The device further comprises a control unit 32, which may e.g. be provided with a microprocessor 34 and with a memory 36 as well as with a display 38. Memory 36 may hold the data as well as the program instructions required for carrying out the steps of the present method. Display 38 may e.g. be used for showing the data determined thereby and in particular for displaying any images derived by means of the techniques described herein.
The device of
Fluorescence detector 40 comprises a light source 42 generating a probe beam 44. Advantageously, light source 42 is a pulsed light source generating short light pulses, in particular having a length of no more than 100 ps.
Probe beam 44 is sent through a first dichroic mirror 45 onto a beam combiner 46, where it is made collinear with measurement beam 28 of OCT interferometer 8.
In one embodiment, beam combiner 46 may e.g. be a second dichroitic beam splitter, which reflects probe beam 44 as well as the fluorescent light from the eye but transmits measurement beam 28, or vice versa.
Advantageously, probe beam 44 is sent, together with measurement beam 28, through the scanner mirrors 24a, 24b and objective lenses 26a, 26b.
Probe beam 44 advantageously has substantially the same focal point location, in particular within +/−5 mm, advantageously within +/−1 mm, as measurement beam 28 (in this context, “focal point location” designates the location of the focal point along direction z). This provides improved spatial synchronization between the fluorescence lifetime measurements and the OCT measurements.
The wavelength of probe beam 44 is selected to generate fluorescence in eye 30. Advantageously, this wavelength is in the range of 360-500 nm, advantageously at 470 nm±20 nm (for lipofuscin excitation) or 440 nm±20 nm (for A2E excitation), even though excitation outside this range may also be possible, often with reduced efficiency. The fluorescent light may be in the range of 500-700 nm, e.g. centered at about 600-610 nm for lipofuscin fluorescence or 565-575 nm for A2E fluorescence. The wavelength of the measurement beam 28 from interferometer 8 is e.g. in the range of 980-1150 nm or 700-900 nm, which allows to spectrally separate the OCT and fluorescence measurements.
Fluorescent light generated in the structures of the eye of probe beam 44 is, in part, fed back through objective lenses 26b, 26a, mirrors 24a, 24b and beam combiner 46 to arrive at first dichroic mirror 45.
Dichroic mirror 45 is e.g. designed such that it transmits the light of probe beam 44 but reflects the fluorescent light from the eye. Hence, the fluorescent light is reflected into a light detector 48.
The signal from light detector 48 is fed into a lifetime analyzer 50, which may form part of control unit 32. Lifetime analyzer 50 is structured and adapted to measure one or more parameters of the delay process of the fluorescent light as it will be described in more detail below.
Advantageously, measurement beam 28 as well as probe beam 44 pass focus controller 52, which allows to commonly adjust the focal point location of both measurement systems (OCT and lifetime spectroscopy). In particular, focus controller 52 is adapted to vary the focal point location over an effective optical distance (i.e. distance multiplied by effective refractive index of the regions passed by the beams) of at least 30 mm, in particular of at least 40 mm, such that the focal point location can be set into the retina as well as into the lens of eye 30.
The scanning mirrors 24a, 24b form scanning optics for commonly deflecting probe beam 44 and measurement beam 28 into directions x, y perpendicular to the optical axis z of the microscope device. This allows, as mentioned, to spatially synchronize the two measurements.
Fluorescence Lifetime Measurements
Fluorescent lifetime may e.g. be measured by sending a short pulse of probe light into the eye and performing a time-resolved measurement of the fluorescent response, i.e. of the raw fluorescent data (for alternative methods, see “Notes” below). In general, the fluorescent response as a function of time t will be a sum of exponential decays. This, for example is the fluorescent response IFL(t) of the lens of the eye:
Ai are the characteristic amplitudes and z, the decay times of the n involved fluorescent processes.
Fluorescence detector 40 is adapted to determine at least part of these amplitudes and decay times.
To do so, fluorescence detector 40 may e.g. be designed to carry out time-correlated single-photon counting (TCSPC). This widely used method involves sending several pulses into the eye and recording the responses using a fast single-photon detector. After enough recorded events, a histogram of the number of events across all the recorded time points is calculated, and then the amplitudes Ai and the decay times τi of Eq. (1) are determined by curve fitting.
Advantageously, though, fluorescence detector 40 is adapted to carry out an “analog mean delay” measurement as e.g. described in in Moon et al., Optics Express 17(4), 2834-2849, US2019310198, and US2020088638. This method works with larger light intensities and is therefore faster than TCSPC.
Lens Contribution Compensation
In one embodiment, a fluorescence lifetime measurement is first performed on the lens of the eye by setting focus controller 52 to focus OCT measurement beam 28 as well as fluorescence probe beam 44 onto the lens of the eye, e.g. the center of the lens. This reduces the amount of fluorescent light returned to detector 48 from other structures of the eye.
The response from the lens takes the form of Eq. (1).
In a second step, focus controller 52 is set to focus OCT measurement beam 28 as well as fluorescence probe beam 44 onto the retina, and a scan along directions x and y (perpendicular to direction z) is carried out, which yields a response (raw fluorescent data) as a function of location x, y as follows:
Here, ÃR, i and {tilde over (τ)}R,i denote the amplitudes and decay times of a superposition of the fluorescent processes in the retina and the lens.
In order to isolate the contribution of the processes in the retina only, corrected fluorescence data IFR(x,y,t) is calculated from the raw fluorescent data ĨFR(x,y,t) as follows:
Here, IFL(t) is the fluorescent response measured from the lens, a is correction factor, and α IFL(t) is an estimate of the contribution of the lens fluorescence to the raw data. AR,i,x,y and τR, i, x, y are the amplitudes and decay times of the retina fluorescence as a function of the coordinates x, y.
Correction factor α may be as follows:
a) It may be a constant, which is e.g. estimated by the device manufacturer. For example, it may be estimated by ray tracing calculus while modeling the amount of lens fluorescence arriving on detector 48 when probe beam 44 is focused on the retina and using the approximation that the contribution of the lens fluorescence to the raw fluorescent data is substantially independent of the coordinates x, y. In this case, the “estimated fluorescence data” of the lens is the same for different locations x, y.
b) It may be selected depending on the thickness, volume, or shape of the lens or on the distance between the lens and the retina. These parameters can readily be measured by OCT.
c) It may be dependent on fluorescence inhomogeneities of the lens due to an inhomogeneous composition of the lens. In this case, the lens fluorescence IFL(t, x′, y′) is measured, in the first step above, by scanning it as a function of coordinates x′, y′ and directing probe beam 44 to different parts x′, y′ of the lens.
In cases b) and, in particular, c), the estimated of the contribution of the lens fluorescence is a function of coordinates x, y. Ray tracing calculus can be used to estimate this function. In other words, in this case the “estimated fluorescence data” of the lens is different for at least two different locations x, y. Such ray tracing may e.g. use the thickness, extension, volume or position of the second structure.
In the example above, it is assumed that the fluorescence lifetime of the retina is to be measured, and the contribution of lens fluorescence is to be eliminated. In more general terms, the fluorescence lifetime of a first structure of the eye is to be measured and the contribution of the fluorescence of a second structure of the eye is to be eliminated. For example, the structure of interest (the “first structure”) may also be the lens and the structure whose contribution is to be eliminated (the “second structure”) may be the retina. Other possible first and second structures include the cornea or the vitreous body of the eye.
In Eq. (1) above it is assumed that, when focusing on the lens, the contribution of retina fluorescence can be neglected. If this is not the case, Eq. (1) can be modified to include a contribution from the retina fluorescence. This contribution may e.g. be assumed to be the integral of IFR(x,y,t) over x and y scaled by a second correction factor α′. The second correction factor α′ can e.g. be assumed to be constant and be estimated by the device manufacturer using ray tracing calculus on typical eye model data.
Retina Thickness Compensation
When scanning the fluorescence of a structure along x- and y-, the amount of fluorescence of the structure is often proportional to the thickness tu of the structure along direction z.
Hence, it can be of interest to calculate a normalized fluorescence parameter, such as normalized fluorescent amplitudes A′R,i,x,y, by normalizing the un-corrected fluorescence parameter with said thickness, i.e.
A′
R,i,x,y
=A
R,i,x,y
/t
xy. (4)
Such a normalized fluorescence parameter may e.g. describe a composition or activity of a given part of a structure more reliably than its non-normalized counterpart. This is particularly true for the retina, for which the thickness varies as a function of x and y, but degenerative defects can better be determined from the normalized parameter.
The thickness tx,y can be readily measured by means of interferometer 8, i.e. by means of an OCT measurement.
The thickness tx,y can e.g. be the retinal pigment epithelium (RPE) thickness of the retina, but other thickness parameters may be used as well.
Motion Compensation
When carrying out several fluorescence lifetime measurements over an extended time period, such as over at least 1 ms, motion artifacts of the eye may render it difficult to correlate the time response of the measurements. For example, when the patient moves eye 30 closer to the device, the fluorescent signals will be quicker to arrive at detector 48.
This may e.g. the case during repetitive TCSPC measurements or while scanning a structure of the eye along the x- and y-coordinates.
With the present device, OCT measurements can be used to monitor the distance between the eye and the device, and the measured changes in the distance can be used to compensate a time-offset in the fluorescence data.
For example, the distance to the “first” or “second” structure as defined above can be monitored. Alternatively, the distance along z between the device and the cornea or any other part of the may be monitored.
If, for example, F1(t1) designates time response of a first fluorescent measurement and F2(t2) designates the response of a subsequent second fluorescent measurement, with time t1 and t2 being time relative to the probe light pulse, the two responses may be shifted in time if the eye has moved along z.
For example, if the eye has moved away from the device along z, between the two measurements, by a distance d, the second measurement can be offset by 2d for comparing it with the first measurement, i.e. F2(t+2d) has the same time frame as F1 (t), e.g. for superimposing event times in two TCSPC measurements.
Alternatively to (or in addition to) compensating changes in distance, the OCT measurements can be used to monitor that the eye is in a suitable range for the fluorescence measurements. In this case, a fluorescence measurement is enabled only if the distance is in a given range.
Intra-Eye Delay Compensation
The temporal resolution of fluorescence lifetime measurements may reach an order of magnitude of e.g. 10 ps. In this time, light travels a distance of 3 mm in vacuum.
Hence, the spacing between the structures of the eye may give rise to noticeably different delays in the fluorescent response from the various structures as measured by detector 48. For example, the fluorescent light from the lens may arrive at detector 48 several 10 ps before the fluorescent light from the retina.
To compensate for this, for example, Eq. (3) may be replaced with
I
FR(x,y,t)=ĨFR(x,y,t)−αIFL(t+2Dn/c) (5)
Here, D is the (average) distance between the first and the second structure, n is the (average) refractive index of the eye between the two structures, and c is the speed of light. Eq. (5) takes into account that the fluorescence from the second structure (the lens, in this case), will arrive at detector 48 earlier than the fluorescence of the first structure (the retina, in this case), namely by the time it takes for the light to travel from the second to the first structure and back, i.e. by 2Dn/c.
The distance D can be a typical distance between the two structures in the human eye. Advantageously, though, distance D is measured by means of OCT with interferometer 8.
Hence, in more general terms, the method may comprise the step of offsetting the estimated fluorescence data of the second structure may in time as a function of the distance D between the first and the second structure before it is combined with the raw fluorescence data in order to calculate the corrected fluorescence data. Advantageously, the distance D is measured by means of the OCT measurements.
Notes
As shown above, combining an OCT interferometer and a fluorescence detector in a single device provides numerous advantages. A further advantage of such a system is the fact that the space-resolved lifetime fluorescence data along x and y can be easily mapped to structures in the eye determined by OCT. Such mapping can e.g. be used to create superimposed images of structural and fluorescent features and/or to categorize structures.
On the other hand, as described above, it is possible to correct the fluorescence lifetime raw data of a first structure by estimating the fluorescence response of a second structure interfering with the experiment and by using the estimated fluorescence data of the second structure for generating corrected fluorescence data for the first structure. As shown, this technique can be advantageously combined with OCT measurements from the same device, but it can also be used with devices that do not provide OCT capability.
As it has been mentioned, fluorescent lifetime may be determined in the time domain by sending a short pulse of probe light into the eye and performing a time-resolved measurement of the fluorescent response. Alternatively, though, the measurements may be carried out in the frequency domain, see e.g. Shim et al. in Journal of the Korean Physical society, 49, pp. S647-S651. In this case, the light source is pulsed or modulated at a high frequency. The amplitude and phase shift of the fluorescent signal may be measured for different frequencies, which allows to retrieve one or more of the characteristic amplitudes Ai and decay times τi. Once these parameters are known, the methodology above can be used to compensate for the contribution of the “second structure”, such as the lens, in the raw fluorescence data.
The techniques described here can be used with any kind of OCT, in particular with time-domain OCT as well as frequency-domain OCT. Frequency-domain OCT, and in particular swept-source OCT, is, however, advantageous for its ability to obtain an A-scan quickly.
In the above examples, it is assumed that the response of fluorescence detector 48 is instantaneous and the length of the pulse of light source 42 negligible. If that is not the case, the signals at detector 48 are convolutions of the detector response, the light pulse shape, and the exponential fluorescent decays, and the functions to be fitted need to be adapted accordingly as known to the skilled person.
While there are shown and described presently preferred embodiments of the invention, it is to be distinctly understood that the invention is not limited thereto but may be otherwise variously embodied and practiced within the scope of the following claims.
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/EP2020/065940 | 6/9/2020 | WO |
