BIOIMAGING AND OPTICS PLATFORM EPFL SV PTBIOP LASER SCANNING CONFOCAL MICROSCOPY PRACTICAL CONSIDERATIONS

Transcription

1 LASER SCANNING CONFOCAL MICROSCOPY PRACTICAL CONSIDERATIONS

2 IMPORTANT PARAMETERS Pixel dwell time Zoom and pixel number

3 PIXEL DWELL TIME How much time signal is collected at every pixel Very small values, normally in microseconds range Defined either directly or indirectly by scan frequency and image size Example: 512x512 pixels, 400 Hz - dwell time sec Important characteristic for signal intensity

4 ZOOM AND PIXEL NUMBER Pixel size is defined by zoom and pixel number To change pixel size 1. Change zoom - image size changed, number of pixels constant 2. Change number of pixel - image size constant, pixel size changed 512x x x2048

5 SCANNING MECHANISM ZOOMING/PIXEL NUMBERS Zooming: change the mirror deflection range (=Amplitude of scanner input signal) Zoom in: smaller field of view; higher energy deposition (bleaching) Pixel numbers: change the speed of the scanner Large pixel numbers lower scan speed

6 PRACTICAL CONSIDERATIONS Light efficiency Signal to Noise Ratio Bleaching Resolution Physical resolution Nyquist Sampling Optimal Pixel/Voxel Size Strategies for optimal imaging

7 LIGHT EFFICIENCY Photons emitted by a single fluorophore - How many photons can one obtain from a typical fluorophore in confocal microscopy? - How does can we influence the photon flux? - Why is the photo flux important for (Laser scanning confocal) microscopy?

8 LIGHT EFFICIENCY PHOTONS EMITTED BY A SINGLE FLUOROPHORE very fast k a S 1 k b k a = s * I k b = 1/t F t F =fluorescent lifetime [ S 1 ] [ S ] tot k S 0 a ka k b Photon output (steady state) = k b *[S 1 ]

9 LIGHT EFFICIENCY PHOTONS EMITTED BY A SINGLE FLUOROPHORE very fast S 1 k a = s * I = 3.8 * 10 8 s -1 k a k b S 0 k b = 1/t F t F = 4.5 ns k b = 2.2*10 8 s -1 nm foccused to a radius of 0.25 mm (e.g. objective NA 1.25) I = 5.1 *10 5 W/cm 2 = 1.25*10 24 photons/(cm 2 *s) Fluoresceine: e = l*mol -1 *cm -1 Cross section/molecule = 3.06 * cm 2 /molecule

10 LIGHT EFFICIENCY PHOTONS EMITTED BY A SINGLE FLUOROPHORE very fast S 1 1,0 1,0 0,8 0,8 k a k b S 0 /S tot 0,6 0,6 S 1 /S tot 0,4 0,4 k [ S ] 1 a 0,0 Stot] ka k b k a /k b [ S 0 0,2 0,2 0,0

11 LIGHT EFFICIENCY PHOTONS EMITTED BY A SINGLE FLUOROPHORE k a S 1 k b very fast k a = s * I = 3.8 * 10 8 s -1 k b = 1/t F S 0 t F = 4.5 ns k b = 2.2*10 8 s -1 Photon output: Q e *k b *[S 1 ] = 0.9 * 0.63 * 2.2*10 8 s -1 = 1.3* 10 8 photons/s 512*512 pixel / s = 3.8 ms/pixel ~ 500 photons/pixel (for fluoresceine excited with 488 nm)

12 LIGHT EFFICIENCY PHOTONS EMITTED BY A SINGLE FLUOROPHORE Single Fluorophore (Fluoresceine): 512*512 pixel / s = 3.8 µs/pixel ~ 500 photons/pixel But losses due to - Geometry (fluorophore emitting in all directions) -NA % efficiency -NA 0.3 5% efficiency - Scattering - Optical elements - Objective - mirrors, filters - pinhole - Detector efficiency 10-30% Each fluorescent molecule contributes on the order of only one photoelectron/pixel/sweep

13 SIGNAL TO NOISE RATIO OVERVIEW - What is -Signal -Noise -Background -How do the above mentioned terms influence image quality -How can we influence signal, noise and background

14 SIGNAL TO NOISE RATIO DEFINITION Signal Noise Background Photons Fluctuation in the Signal (=photonflux) Unwanted signal (=photons)

15 SIGNAL TO NOISE RATIO MAIN SOURCES OF NOISE N N N tot s d N tot total noise N d detector noise N s signal (shot) noise N ph number of photons Shot noise prevails at high light level. It defines the fundamental limit of the noise. NS N ph Detector noise prevails at low light level. Statistics is applied to photons (electrons), but not to grey values. 15

16 SIGNAL TO NOISE RATIO SIGNAL-NOISE-BACKGROUND Signal to Background Signal to Noise Contrast constant increases increases

17 SIGNAL TO NOISE RATIO Dwell time: 50 µs Dwell time: 6 µs Dwell time: 1.6 µs

18 PINHOLE SIZE Pinhole: 10 AU Pinhole: 5 AU Pinhole: 1 AU Closing the pinhole increases the z-sectioning capabilities because light from out of focus planes is suppressed, but also decreases the SNR/contrast (less photons are detected even from the focal plane).

19 NOISE REDUCTION Strategies to reduce image noise - reduce scan speed = increases pixel time - averaging line averaging frame averaging integrate - increase laser intensity

20 BLEACHING Excitation - Cone of light - Bleaching occurs above and below focal plane - z-sampling affects bleaching

21 Energy BLEACHING MECHANISMS Photon output: 1.3* 10 8 photons/s = 1.3* 10 8 cycles/s S 2 High energy density High cycle time High bleaching probability T 2 S 1 S 0 t ~ ns t ~ µs-s T 1 - Photochemistry of bleaching poorly understood. - Differs from fluorophor to fluorophor -Main Bleaching route: Triplet state - Absorb another photon - Energy exchange with triplet oxygen, which becomes highly reactive singlet oxygen

22 TRIPLET STATES Energy t T =1 µs 0.8 k IC =0.01*(1/t F S 2 T 2 S 1 /S tot S 0 /S tot S 1 t ~ ns t ~ µs-s T S k a /k b Ensemble of molecules: 60% are trapped in triplet state decrease of fluorescent intensity by 60%

23 BLEACHING Parameters affecting bleaching probability - Fluorophore - Laser intensity - Pixel Dwell time - Sampling frequency Bleaching (Photon)Noise

24 LASER SCANNING CONFOAL MICROSCOPY OUTLOOK Resolution Point Spread Function Nyquist Sampling Optical Slice thickness

25 LASER SCANNING CONFOAL MICROSCOPY POINT SPREAD FUNCTION

26 LASER SCANNING CONFOAL MICROSCOPY OPTICAL RESOLUTION LIMIT The optical resolution resel 1 can be defined as the radius of the first dark fringe in the diffraction pattern, or half the diameter of the Airy disc. 1 Robert H. Webb, Confocal optical microscopy, Rep. Prog. Phys. 59 (1996) resel resel.61 n sin 0 wide-field xy-plane.44 n sin 0.61 NA 0 confocal xy-plane 0.44 NA resel axial 1.5 n 2 NA confocal axial Example: 500 nm; n=1.518; NA Obj =1.4 wf xy-plane resel=217 nm conf. xy-plane resel=157 nm conf. axial resel=580 nm

27 rel. intensity Intensity OPTICAL RESOLUTION THEORY/PRAXIS Example: 500 nm; n=1.518; NA Obj =1.4 wf xy-plane resel=217 nm conf. xy-plane resel=157 nm conf. axial resel=580 nm FWHM resel bead: 170 nm FWHM: 234 nm FWHM distance / mm resel optical unit

28 DIGITAL SAMPLING 488 nm 325 nm 244 nm 195 nm 160 nm 6.4 mm =1280 pixel à 5 nm 6.4 mm =64 pixel à 100 nm

29 PIXEL SIZE 33 nm 65 nm 130 nm 260 nm 12 MB t=64x 3 MB t=16x 0.8 MB t=4x 0.2 MB t=1x

30 PIXEL SIZE 33 nm 65 nm 130 nm 260 nm 30

31 LASER SCANNING CONFOAL MICROSCOPY RESOLUTION Physical Resolution Parameters affecting resolution Wavelength NA of objective Refractive index Difference in lateral and axial resolution Digital resolution (Sampling) Nyquist sampling Optical zooming Number of pixels

32 SCAN MODE Scan Mode unidirectional/bidirectional Sequential Cross talk Cross excitation

33 LASER SCANNING CONFOAL MICROSCOPY BLEED-THROUGH

34 LASER SCANNING CONFOCAL MICROSCOPY SUMMARY Widefield microscopy Detector efficiency : 60%-80% Detector noise : 4 RMS electrons/pixel Pixel Dwell time : ms Opical sectioning : no Backgrond supression : no Laser scanning confocal microscopy Detector efficiency : 3%-12% Detector Noise : < 0.01 RMS electrons/pixel Pixel dwell time : µs Optical sectioning : yes Background supression : yes

35 LASER SCANNING CONFOCAL MICROSCOPY SUMMARY