A Control and Acquisition System for a Two-Photon Fluorescence Microscope

Author(s):
Iain G. Cormack - Institut de Ciencies Fotoniques, Mediterranean Technology Park

Industry:

Products:
LabVIEW, Data Acquisition

The Challenge:
To provide an advanced, user-friendly, control system for a laser scanning two-photon fluorescence microscope used for three-dimensional imaging of living cells. The control and acquisition of the data has to be achieved quickly to reduce the risk of cell damage and be perfectly synchronised to avoid image distortion.

The Solution:
National Instruments powerful LabVIEW graphical programming environment provided us with an ideal platform to rapidly develop a powerful yet user-friendly control and acquisition system for the laser scanning two-photon fluorescence microscope.

Article

The need for biologists to image living cells that have increased complexity is on going. A two-photon fluorescence microscope is a specific type of microscope that can help meet these demands. This article will outline how, with the help of National Instruments LabVIEW programming environment, it was possible to construct one of these microscopes.
A two-photon fluorescence microscope works by tightly focusing very short pulses of laser light (100 femtoseconds) onto the sample. By raster scanning the focal spot across the sample (x and y axis) and collecting the fluorescent light that originates from the focus it is possible to form an image of a living cell. Furthermore, by moving the sample up or down (z-axis) it is possible to acquire a number of stacked images from which a three-dimensional image can be formed of the sample.
Commercial two-photon fluorescence microscopes, as with any other state-of-the-art equipment, are extremely expensive pieces of equipment. It was the aim of this work to construct the device from a traditional optical microscope. The success of the project largely depended upon the creation of a flexible, user-friendly program that could control all parts of the microscope while simultaneously acquiring data in a fast, concise and systematic way.
Our microscope, shown in Figure 1, had five separate pieces of external equipment that needed to be controlled.

1. A pair of scanning galvanometric mirrors were used to raster scan the laser beam in the x and y axis. By applying a voltage (between 10V) from the Digital-Analogue (D-A) output channel of our DAQ (NI PCI-6052E) to each of the mirrors we were able to have direct control over the position of the scanned beam inside the microscope.
2. The z-control (up and down) movement of the focal position was controlled by a DC-servo motor (from Physik Instrumente) placed against the focal control knob of the microscope. LabVIEW drivers came automatically with the device making the integration of this non-NI piece of equipment within LabVIEW very straightforward.
3. A photomultiplier tube (PMT), from Hamamatsu, was used to detect small amounts of fluorescent light originating from the focal spot of the laser. This device works by converting small levels of light into an output voltage, the higher the voltage, the more light is detected. A single analogue-digital input was therefore used to measure this voltage.
4. An electronic shutter was used to block the laser beam while no image was being taken. A digital output from the DAQ was used to change the position of the shutter to either ‘open’ or ‘closed’.
5. A digital CCD camera (Q-image) with a Firewire interface was used to acquire the traditional optical image of the cell. The use of the NIVision made the integration of the camera into LabVIEW a very simple procedure.
The block diagram, in Figure 2, outlines how this equipment is integrated within the whole system.

The control panel was designed in a way so that a non-technical person could use the microscope. The initial settings for microscope need to be first selected within the control panel (Figure 3). The laser scanning region is entered by selecting its top left and bottom right corners. Other variables such as the size and speed of scan, the number and separation of stacks, the time interval between each time-lapse measurement are set as inputs. These parameters, as well as the name of the image and root directory are all saved together with the acquired image data.

Vast quantities of data are collected during a stacked, time lapse measurement. It was therefore important for the program to automatically name and place data within an intelligent directory structure so that the user could quickly access a specific data set. This was achieved very easily with LabVIEW. Furthermore, the data automatically arranges itself into a 2-D matrix, displayed on the computer screen and saved to disk as raw data. Finally, if selected, the image can also be saved as a jpg image.
Once the start button is pressed, the coordinates for the raster scan are automatically created and ‘loaded’ into the two D-A output channels. The use of DAQmx allowed the quick configuration of the DAQ channels so that the two D-A output channels (one for each mirror) were triggered and synchronised with the A-D input (PMT voltage). This ensured that the PMT signal could be directly associated with raster scan position. Once this raster scan is completed, the z-axis is then automatically moved and another scan is started. This process continues until the whole stack is acquired. Finally, the shutter is automatically closed until the next time-lapse measurement is to be taken.
In the future, the microscope will be upgraded in two ways. Firstly, the data acquisition card will be replaced with one that has a larger sampling rate allowing the acquisition rate to become faster. Secondly, to obtain extra complementary information from the image, an additional PMT will be added to the microscope. The signal from this PMT will need to be measured synchronously with the other PMT. The power and flexibility of LabVIEW allow these two hardware upgrades to be easily implemented within the current program. This ability of LabVIEW to be able to grow along side our rapidly changing experimental arrangement is vital for us helping us save both valuable time and money.

Author Information:
For more information on this Case Study, contact:
Iain G. Cormack
Institut de Ciencies Fotoniques, Mediterranean Technology Park
Castelldefels (Barcelona)

Browse All Case Studies »