WO1995030167A1

WO1995030167A1 - Scanning confocal optical microscope

Info

Publication number: WO1995030167A1
Application number: PCT/GB1995/000458
Authority: WO
Inventors: John Graham White; William Bradshaw Amos
Original assignee: Medical Research Council
Priority date: 1994-04-30
Filing date: 1995-03-03
Publication date: 1995-11-09
Also published as: GB9408688D0

Abstract

A scanning confocal optical microscope wherein a light source (S) directs a collimated beam of light on to a scanner (S3) and thence through a beam splitter (B) on to an orthogonal scanner (S1) so that a specimen (SP) is scanned with a confined area of light in two orthogonal directions, and light returned from the specimen is descanned by the scanner (S1) and via the beam splitter caused to oscillate along a slitted aperture (SA), from which light passes to a third scanner (S2) which scans the line-oscillating light to be scanned perpendicularly to the length of the line, thereby to cause the light to be incident on an area of a target (CT) corresponding to the scanned area of the specimen.

Description

TITLE: SCAHHIHG COHFOCAL OPTICAL MICROSCOPE

Field of the Invention

This invention relates to a scanning confocal optical microscope.

Background to the Invention

In 1957, U.S. Patent No. 3,013,467 (Minsky) disclosed a confocal microscope. In the patent, the sense of the word "confocal" is that a small spot of light is focussed on the specimen and the detection of emitted light is confined to the same spot. Minsky described a major advantage of this mode of microscopy: namely its ability to improve the discrimination of depth in the specimen. This effect is at a maximum when the volume corresponding to the focus of illumination and the photometric volume are congruent and as small as possible, ie at the minimum size set by the laws of optical diffraction.

However, useful forms of microscope have subsequently been described, in which either the region of illumination or the region of detection, or both, have been extended. For example, an approximation to a point of focus of illumination has been combined with a detection volume of similar or larger size, made continuously variable by means of an iris diaphragm in front of the detector; or as described below, the region of illumination and detection has taken the form of a line or bar. These later forms of confocal microscope, for many applications, have advantages which offset their lesser performance in comparison with the Minsky design.

In order to illustrate the prior art, reference is made to Figures 1, 2 and 3 of the accompanying drawings. These diagrams are symbolic rather than constitutionally representational. For example, scanning beam deflectors are shown as oscillating mirrors even where the preferred form may be an acousto-optical deflector, rotating polygonal mirror or other device. Again, where a parallel beam of light is shown as generated from a source by passage through a small circular pinhole placed at the focus of a lens (ie a collimator) it is to be understood that a laser, with or without known beam-expanding means, may be substituted. Also, where the diagrams show a lens focussing light onto a circular pinhole placed before a detector, the preferred embodiment may in fact be an extended light path without any such focussing lens and with the detector pinhole replaced by an adjustable iris with an aperture of millimetre dimensions.

According to this scheme of illustration, Figure 1 shows the prior art of U.S. Patent No. 5,032,720 (White), published in 1990, in which light from a source at 3 passes through a pinhole at A1 and a lens L1 , in such a way that it is collimated and proceeds as a parallel beam to a beam splitter B which takes the form either of a simple beam splitter without wavelength selectivity or a chromatic beamsplitter (commonly though incorrectly called a dichroic beam divider) . Light from the beam splitter B is conveyed via scanning deflectors 31 and 32 to an objective lens OBJ, which focusses the beam to a diffraction-limited spot on the specimen SP. S1 causes the spot to scan the specimen in a direction orthogonal to the scan direction produced by the action of S2. The combined scanning action of S1 and S2 produces a raster scan of the focussed light spot on the specimen SP. Light emitted by the specimen SP as a result of reflection, fluorescence of other processes passes back through the scanning system of S2 and S1 and is thereby descanned to form a stationary beam passing back towards the beam splitter B. Part of this beam is transmitted by B and is focussed by lens L2 on circular pinhole aperture A2 which is placed at a conjugate focus to pinhole A1 and also to the focussed spot in the specimen. The light which is able to pass through pinhole A2 is detected by a unitary detector such as a photomultiplier tube P. The photocurrent output from the photomultiplier P is processed by electronic means to generate an image which may be displayed on a television monitor.

The prior art described so far has the disadvantage that the scan rate is limited to that obtainable by electromechanical mirror deflectors. Acousto-optic deflection devices are faster, but cannot be used because they are dispersive. If, for example, in the White arrangement, an acousto-optic device were to be used as the deflector S1 and a fluorescent specimen was present, the light emitted by the specimen, being of a different wavelength from the incident light, would follow a path through 31 which was not the reverse of the original path and is likely to fail to pass through the detector aperture.

Figure 2 shows other prior art, due to Draaijer and Houpt (Draaijer, A. & Houpt, P.M. (1988) Scanning j_0_, 139-145) which overcomes the first of these problems. The main difference from the prior art of White is that a second beamsplitter B2 is placed between the scanning deflectors S1 and S2, such that the return beam from the specimen passed by B2 through lens L3 is descanned by scanning deflector S2 but not by scanning deflector S1. The focussed spot is therefore not stationary, but oscillates up and down a straight line which lies within a modified aperture 3A in the form of a slit. The width of the slit at SA is chosen so that light emanating from regions of the specimen outside the focussed spot is substantially prevented from reaching the photomultiplier P2, so effecting a degree of depth discrimination similar to that in a strictly confocal system with circular pinhole apertures or their equivalent. With this arrangement, the preferred form of scanning deflector S1 is an acousto- optical deflector, providing a fast line-scan, while scanning deflector S2, which produces a slower page-scan, remains an oscillating mirror. With a fluorescent specimen, the problem of dispersion does not arise, since scanning deflector S2 is a non-dispersive and a chromatic ' beamsplitter 32 is used, so that the emitted beam is directed to the detector (photomultiplier P2) without passing through 51. Draaijer and Houpt also disclose a non-chromatic beamsplitter B1, a focusssing means L2, a circular aperture A2 and a photomultiplier P1 which function analogously to the corresponding parts of White's apparatus in Figure 1. These components are used for reflection imaging, where the incident laser light and the light reflected by the specimen are of the same wavelength and monochromatic, so there is no dispersion in the acousto-optic device 31. This prior art allows faster scanning than that of White, even with fluorescence images, but the detector is inefficient under certain conditions, as will now be explained.

The sensitivity of a confocal imaging system is ultimately limited by the characteristics of the photodetectors used. Photomultipliers have fast response times but relatively low quantum efficiencies for detection (around 10% in the blue) and relatively high dark currents. The response in the red end of the spectrum drops down to a tenth of this value, even with cathode materials optimised for red response. A photomultiplier is not an integrating device, but integration can be achieved digitally by use of special purpose electronics or a high-speed computer. Charge-coupled detectors (CCD) have quite different characteristics. They are usually fabricated as linear or two dimensional arrays of pixel elements and are therefore imaging devices. CCDs have very high quantum efficiencies in the red end of the spectrum (up to 80% with a back- thinned chip) . Dark currents can be very low if the CCD is cooled and an image can be integrated on a chip before it is read out. However, the act of read-out generates noise. These devices therefore operate best by integrating an image on the chip and reading- out once per image capture. Used in this mode, the CCD can offer more than an order of magnitude increase in sensitivity (quantum efficiency) over photomultipliers.

Awamura, Ode and Yonezawa have described a confocal microscope similar to that of Draaijer and Houpt, except that instead of the photomultiplier tube, a linear array of charge-coupled elements (linear CCD) is placed in the position of the slit SA of Figure 2. This description was published in the Proceedings of 5PIE, The International Society for Optical Engineering (1987) Volume 765 pp.53- 60. This answers one of the objections to the use of a photomultiplier, ie low sensitivity at higher wavelengths, but has another disadvantage: namely that the use of a linear CCD requires that the device be read out at every horizontal line of the frame (ie typically around 500 times), thereby incurring a penalty of 500 times more readout noise than would be obtained from a two- dimensional array which would only have to be read out once per frame.

This difficulty has led to the proposal of yet another form of confocal microscope (Figure 3) which allows known, two-dimensional CCD array cameras to be used as the detectors. The figure represents an invention of White, Amos ύ Fordham (patent pending), but is in certain respects similar to that described by Brakenhoff and Visscher (Brakenhoff, G.J. f Visscher, K (1990) Trans. Roy. microsc. Soc. _, 247-250.

Thus, light from a source at 3 is formed into a beam with a cross-section in the form of a narrow bar, shown symbolically in Figure 3 as light passing through a slit aperture SA1. This light is passed via a moving-mirror type of deflector S1, which produces a scan in a single direction, to the objective lens OBJ and thence to the specimen at SP. Lens L1 and the objective lens OBJ represent a lens system which produces a line focus, ie a maximally thin bar of light on the specimen, as represented by the arrow at SP. The bar of light is caused by scanning deflector S1 to scan in a direction perpendisular to the long axis of the light bar. The beam splitter B, which may be of the simple type or with chromatic properties, directs some of the light, emitted from the specimen as the result of reflection or fluorescence, as a descanned beam to focussing optics L2 which form an image of the specimen on slit aperture 5A2. In this aperture is cast a stationary line of illumination, and the image of the specimen at this level oscillates back and forth in a direction pe pendicular to that of the bright stationary line. Focussing means, represented by L3 and L4, image the aperture SA2 on to a camera target CT or, alternatively, on the retina of the human eye. The image of the bright line is represented by a dashed arrow on the surface of the target CT. By oscillation of the mirror 32 in synchronous antiphase with S1, the bright line on the target CT is caused to sweep out a rectangular area which is equivalent to an image of the scanned area of the specimen. The slit aperture at 3A2 is varied in width to provide control over the confocal performance of the system, allowing the best compromise to be reached between signal strength (slit open) and depth-discrimination (slit closed). It is easy to run such a system at video rates or higher, since the scanning deflectors have to oscillate only at the framing rate, rather than at the much higher line scan frequency.

The prior art of Figure 3 has the advantage that a target such as a CCD camera can be used. However, the departure from spot illumination to line illumination gives poorer depth discrimination. Evidence for this comes from detailed comparative measurements by the present inventors (Amos, W.B. and White, J.G. Direct view confocal imaging systems using a slit aperture. Chapter in "The Handbook of Confocal Microscopy" edited by J. Pawley, Plenum Press (in press)). This results in inferior out-of-focus rejection performance for this class of confocal imaging system. These considerations gave rise to the present invention. The invention

According to the invention, there is provided a scanning confocal optical microscope comprising a light source, first and second scanners for scanning light from the source in first and second orthogonal directions across a specimen on which the light is focussed, a beam splitter positioned between the first and second scanners, whereby light returned from the specimen is descanned in only one of the first and the second directions and deflected by the beam splitter on to a slitted aperture on which the returned light is focussed, whereby the returned light scans or oscillates along the aperture in the other of the first and second directions, a third scanner for receiving light from the slitted aperture, and focussing means for imaging the light from the third scanner on a target.

Most preferably, the third scanner operates to scan the light received through the slit in the said one of the first and second directions, thereby to enable the returned light to be focussed on a target area corresponding to the scanned area on the specimen.

Preferably, the light is descanned by the second scanner and not the first scanner, so that the first scanner can be an acousto-optic scanner or other solid state device which scans in the first direction at high speed. Thus, light from the source oscillates back and forth at high speed across the specimen, with a slower orthogonal scan on the specimen produced by the second scanner, which is preferably a mechanically operating scanner. The third scanner is preferably also a mechanically operating scanner, so that both these scanners are able to handle fluorescent or like dispersive emissions from the specimen, and light already descanned in the second direction by the scanner is able to pass through the slitted region of the slitted aperture. Description of K«bodi»ent

Further features of the invention will be apparent from the following description, referring to the accompanying drawings, in which:-

Figures 1, 2 and 3, already described, illustrate the prior art; and

Figure 4 illustrates the present invention in similar symbolic manner which is not constructionally representative.

In the confocal scanning microscope in accordance with the invention, as shown in Figure 4, spot scanning as in the prior art of Figure 2 is combined with the imaging properties of the prior art shown in 3.

Thus, referring to Figure 4, the beam entering the microscope is shown as originating from source 3, collimated by circular pinhole aperture A and focussing system L1. It should be understood that, in practice, an expanded or unexpanded beam from a laser would be the preferred form. This beam is caused to undergo an angular oscillation in a single direction (shown as horizontal in ^'the drawing by a scanning deflector S3, which is drawn as an oscillating mirror, but would, in the preferred form, be an acousto-optic scanner or other solid-state device. The beam is then passed, either by transmission as shown, or by reflection, through a beam splitter 3 of simple or chromatic type, and thence to another scanning deflector 31 and by focussing means such as the lens OBJ to a spot focus on the soecimen SP. The illuminated spot would in the ideal case be so small as to be diff action-limited. The action of scanning deflector 33 is to cause a rapid oscillating motion of the light spot back and forth along one direction, indicated by the longer two-headed arrow on the specimen SP. The action of scanning deflector 31 is to produce a slower oscillating motion in a direction perpendicular to this, as indicated by the shorter two- headed arrow.

Light emanating from the specimen SP, as a result of reflection, fluorescence, fluorescence consequent on 2- photon excitation, Raman scattering or other processes, passes back via lens OBJ and scanning deflection S1 but does not substantially return to S3 because of the beam- splitting action of beam splitter 3, which diverts, by reflection as shown, or by transmission with an alternative orientation, a part of this signal light which then passes through focussing means L2 to a slit-shaped aperture SA of variable width. Within the slitted aperture SA, the emitted light is brought to a spot focus, which oscillates up and down the length of the slit as indicated by the two-headed arrow on SA. In different regions of this trajectory, the intensity of the light spot varies according to local variations in the specimen.

Focussing means L3 and L4 and scanning deflector 32 produce an image of the stationary trajectory in slotted aperture SA on a camera target at CT, which could alternatively consist of the retina of a human eye. This image, shown by the two-headed arrow on CT, may be made to appear as a continuous line, either by known electronic integrating means or by the persistence of vision in the case of the human eye. The line is caused to oscillate in a direction perpendicular to its length by the scanning action of the scanning deflector S2, which is preferably a moving-mirror deflector oscillating in synchronous antiphase with deflector S1. In this way, a rectangular area is^' swept out on the target which corresponds to the area of the specimen scanned by the spot.

The invention extends to the case in which the spot is replaced by a short linear focus, or series of spots or segments of such a linear focus, distributed along the line of fast scan in the specimen and preferably moving with equal velocities, though not necessarily so. Such an array of moving spots or lines could be produced from a single input beam by known art applied to the electronic control of an acousto-optical or other solid state device, or by reflective means more complicated than the single mirror shown in the diagram, or by lens arrays or rotating or oscillating masks or other known means. It is to be appreciated that in spite of the use of an acousto-optic device at 33, the action of the arrangement would not be marred by the use of polychromatic light, for example arising from a multiline laser or other incandescent or arc lamp source, since the dispersion of foci along the line would not result in any smearing of signal, each region of the line of scan being imaged separately on one pixel of the camera target.

The invention is compatible with known means of varying the scanning mode, for instance a modification in which the deflector 31 is held stationary while a time-dependent series of line scans is built up on the target CT by the continued operation of S3 and 32, and a modification in which the deflectors are similarly operated, but the focussing means OBJ is adjusted progressively to generate an optical section of the specimen SP parallel to the optical axis.

The invention also encompasses forms of microscope in which the scan lines on the specimen, each containing a single or multiple spot or a line or multiple linear segments of the line focus, are themselves multiplied. At any one time, the specimen is then exposed to a two-dimensional array of illuminated spots or lines. In this case the aperture SA is a screen perforated by multiple parallel slits, preferably adjustable in width en masse.

The invention also encompasses forms in which the size of the spots or line foci on the specimen are able to be varied by electronic acousto-optical means or by known conventional optical means. This might be desirable, for example, to avoid the use of unnecessarily high intensity consequent on focussing the beam to a size smaller than the specimen region corresponding to a single pixel.

The invention also extends to an apparatus in which the above described mode of operation is combined, either simultaneously or with rapid interchange, with the mode of operation of the prior an, including particularly that shown in Figure 3.

Claims

Clai s

1. A scanning confocal optical microscope comprising a light source, first and second scanners for scanning light from the source in first and second orthogonal directions across a specimen on which the light is focussed, a beam splitter positioned between the first and second scanners, whereby light returned from the specimen is descanned in only one of the first and the second directions and deflected by the beam splitter on to a slitted aperture on which the returned light is focussed, whereby the returned light scans or oscillates along the aperture in the other of the first and second directions, a third scanner for receiving light from the slitted aperture, and focussing means for imaging the light from the third scanner on a target.

2. A microscope according to claim 1, wherein the third scanner is operable to scan the light received through the slit in the said one of the first and second directions, thereby to enable the returned light to be focussed on a target area corresponding to the scanned area on the specimen.

3. A microscope according to claim 1 or claim 2, wherein the returned light is descanned by the second scanner only and the first scanner is a solid state, high speed scanner.

^• 4. A microscope according to any of claims 1 to 3, wherein the second and third scanners are relatively slow speed mechanically operating scanners.

5. A microscope according to claim 4, wherein the third scanner is operable in synchronous antiphase with the second scanner.

6. A microscope according to any of claims 1 to 5, wherein the light source is a laser.

7. A microscope according to any of claims 1 to 6, wherein the slitted aperture is variable in width.

8. A microscope according to any of claims 1 to 7, wherein the target is a camera having electronic integrating means whereby the image oscillating in the direction of the length of the slitted aperture appears as a continuous line, and the third scanner is operable to oscillate said line in a direction perpendicular to its length.

9. A microscope according to any of claims 1 to 8, wherein the specimen is scanned with an array of light spots or linear segments of light and the slitted aperture is a screen perforated by multiple parallel slits.

10. A microscope according to any of claims 1 to 9, wherein the area of the specimen on which light is incident is variable in size.