Planning Stage

H. Li et al, Improving the precision of fluorescence lifetime measurement using a streak camera, Chinese Optics Letters 8(10), 934-936 (2010).

An etalon of known timing separation is used to calibrate the time axis of a streak camera. The time axis is fit accordingly to a 4th-order polynomial equation. This performs the same method of fitting non-linearity on the time axis and non-uniformity of pixel gain, but without assuming a sinusoidal pattern. The technique is for a one-time calibration of the time axis that can be applied for all future data, versus our real-time method. The field non-linearity is much less pronounced in their system compared to ours, where the sinusoidal sweep amplitude is at the edge of what can fit on the readout screen. This calibration method was successful in reducing the error in a fluorescence lifetime measurement (Rose Bengal, 700ps) from 10% to 2%.

Similarities:

• use of defined time separations to map the temporal nonlinearity
• calibration factors for time as well as collection time (“gain non-uniformity”)

Differences:

• 4th order polynomial fitting versus sinusoidal fit
• not a real-time approach, not suitable for drift conditions

W. Uhring et al, Very high long-term stability synchroscan streak camera, Review of Scientific Instruments 74 (5), 2646-2653 (2003).

This paper directly targets the drift sources in an Optronis streak camera. Patrick Summ of Optronis is listed as a co-author, though the paper is written by Groupe d'optique appliquée PHASE of the Centre national de la recherche scientifique (CNRS) in France. A two-pronged approach is taken. First, the sweep voltage is observed by either inductive or capactive coupling, and corrected using a phase-locked loop. This achieves higher stability, but does not account for drift factors that are not associated with the sweep voltage - for instance, the high voltage photocathode. It is for this reason that they employ a laser reference spot, to correct for the final result instead of a major contributing factor. This system applies a real-time correction of phase by following and correcting a laser spot on the screen. Only one reference spot is provided, so this does not provide any form of linearity correction.

Similarities:

• real-time correction of drift
• same basic method of correcting the COG of a reference spot on screen

Differences:

• single spot used to fix drift, with no extension to time axis correction

Y. Tsuchiya, Advances in streak camera instrumentation for the study of biological and physical processes, IEEE Journal of Quantum Electronics QE20 (12), 1516-1528 (1984).

The article is written by Hamamatsu, and describes a great deal of the specifics and limitations in streak cameras, which have not changed significantly over the years. Notably, it describes the use of an etalon to correct for the time non-linearity and gain non-uniformity, without getting into specifics. This is described for a streak camera with 5% nonlinearity.

The novelty of the invention lies in its simultaneous ability to:

1. calibrate phase drift using pulse COG feedback
2. determine sweep amplitude in real-time using multiple pulses of known temporal separation

While these two tasks have been done independently, they have not been performed simultaneously in a real-time scheme. The combination of the two fix the phase and determine the sweep amplitude, allowing for a full and accurate mapping of the non-uniform time axis independently for each frame.

1. Demonstration of the correct remapping of a fluorescence standard decay (Coumarin 6?) at different points on the screen
2. Timelapse demonstrating how the pulse COG feedback prevents phase drift
3. Temporal remapping at different points in time using the changing reference pulse positions to recalculate the scheme

Scaggs, B. M., & Haas, G. (2007). High Precision Refractive Scanner. Flanders, NJ.

One of the scenarios of the NeoScan product uses refractive surfaces mounted on galvos to perform scans. In the simple case, it amounts to essentially the same system being proposed. It also adds a case where the glass pieces are meniscus lenses, together performing an optical demagnification. This should not be necessary in our system, since the microscope system handles the magnification.

Similarities:

• galvo-mounted glass to perform programmed scan

Differences:

• application meant for laser-machining
• device used to scan a single spot, though it is technically capable of multiple foci

Poland, S. P., Coelho, S., Krstajić, N., Tyndall, D., Walker, R., Li, D. D., Henderson, R., et al. (2013). Development of a fast TCSPC FLIM-FRET imaging system. Proceedings of the SPIE (Vol. 8588, pp. 1–8).

This describes a 7×7 array of beamlets being raster scanned in a similar manner to our requirements. This is done with a conventional galvo setup, where the galvo mirrors affect the beamlets in their Fourier domain. No mention is made to complications involved with the non-uniformity of the scan for each beam member, but this should be a concern. At the very least, the system is complicated by two pairs of relay lenses required to provide the Fourier domain for each galvo scanner.

Similarities:

• application of raster deflecting beamlets to create a full scan in a microscope
• use of a galvo scheme

Differences:

• galvo mirrors used to perform the scan

1. Nipkow-disk - unsuitable for a static array of detectors (collection fiber array)
   (Bewersdorf, J., Pick, R., and Hell, S.W., 1998, Multifocal multi-photon microscopy, Opt. Lett. 23:655–657.)
2. Scanning stage - unsuitable for high-throughput scanning, biological samples
   (Nielsen, T., Fricke, M., Hellwig, D., and Andersen, P., 2001, High efficiency beam splitter for multifocal multi-photon microscopy, J. Microsc. 201:368–376.)

The novelty of the invention is the use of window-equipped galvo scanners as a means of evenly shifting a foci array for use in point-scanning microscopy.

While window galvos have been implemently before, this is the first known demonstration that applies them to a) microscopy and b) multiple foci. Typically, multiple foci schemes in microscopy are scanned with the use of traditional mirror galvos, though these do not achieve the same level of uniform foci shifting applied across an entire foci array. They also run galvos at the limits of their achievable resolution, and require sets of relay lenses and potentially specialized scan lenses (f-theta lenses) to provide a linear output.

1. Reconstructed image of a raster scan target (Convallaria?) using the window galvo system
2. Readout images at the sample stage to quantify the foci array shift vs. window tilt