Nowadays, most pathologists use digital systems within their research projects and practices to analyze images, communicate, and store data. The use of digital systems, whether be it surgical, clinical, forensic, or other branches of pathology, has considerably improved communication and consultation among specialists and also allows for faster diagnosis for patients.
When pathologists use a digital system, they must make sure to select the right imaging equipment. Here, a digital camera can be used to produce a high-quality pathology image, as it can reliably and accurately reproduce what a pathologist observes in the eyepiece of the microscope.
The main elements that comprise a high quality pathology image are high dynamic range, excellent color, and high resolution. It is also imperative that the camera does not add unwanted artifacts to the image (has low noise), can be easily linked to a computer (through a USB interface, for instance) and pathologists are able to detect features even in dark areas (which means the camera must have high sensitivity).
This article analyzes the fundamentals of imaging within the field of pathology and the factors that need to be taken into consideration when choosing an appropriate digital microscopy camera.
Accurate color reproduction is the most important element when choosing a camera for pathology. The captured digital image contains data that can lead to a patient’s diagnosis, and therefore, the camera should be able to display color that matches exactly what the pathologist sees when peering into the eyepiece of the microscope.
In order to make the image more visually appealing, most consumer-grade camera models automatically post-process their images to enhance the saturation of certain colors and, as a result, do not accurately capture the original colors.
A camera that is specifically designed for scientific use should be selected, as it can produce images that are properly calibrated via gamma and white balancing. A computer monitor that is properly calibrated is also equally important when assessing the color of digital pathology images.
Sample image above: prostate carcinoma taken with Lumenera’s INFINITY2-2 Microscopy Camera.
Camera sensitivity and noise
When choosing a digital camera, image quality is the result of the camera’s noise and sensitivity levels. The sensitivity of the camera involves a combination of its pixel size, quantum efficiency (QE), and charge conversion efficiency.
Though pixel sizes and higher efficiencies are important for increasing sensitivity, they must be considered along with the noise specifications of the camera. Read noise and dark current noise are two main noise sources that affect pathology applications. It is important that these values are as low as possible without compromising on high sensitivity.
Quantum efficiency (QE)
QE is defined as the ability of the camera to transform incident photons into usable signal electrons. The camera sensor’s ability to convert the light energy into an electronic signal varies with the wavelength of light.
QE curves are generally included in camera datasheets, showing the camera’s QE response across the spectrum, typically from the shortest visible wavelengths (Blue) and up into the longer wavelengths of the near infrared (NIR).
The following figure illustrates some example QE curves obtained from Lumenera’s INFINITY3-3UR microscopy camera datasheet. Another element that plays an important role in the camera’s sensitivity is pixel size. If the pixels are larger, the camera’s sensor will be more sensitive to the incoming light since there is a larger surface area to collect the incident light.
The image comparison shown below can help demonstrate the impact of pixel size and QE on an image. This comparison was made using two INFINITY microscopy cameras that possess different pixel sizes and QEs.
The same light source and lighting conditions were used to capture these images on a microscope in the same environment.
Initially, the images did not seem very different in terms of brightness, but in order to accomplish this, the camera with smaller pixels and a lower QE (image shown on the right) needed 4 times longer exposure time. The difference in exposure time demonstrates the level of sensitivity of one camera from the other.
Comparison images between the INFINITY3-3URC (left) and INFINITY1-3C (right), captured with identical gain and gamma settings.
When side by side comparison was made on the QE curves for each camera, as illustrated in the following graph, the difference in required exposure time is further highlighted.
As shown in the graph below, the maximum QE for the INFINITY3-3URC camera is almost 50% greater than the highest value for the INFINITY1-3C, leading to a more sensitive camera.
Graph comparing quantum efficiency curves between: INFINITY3-3URC (solid lines) INFINITY1-3C (dashed lines)
In addition, when the size of each pixel is increased, the sensor’s overall sensitivity also increases. This is because the incident light is grouped into the sensor’s larger areas, which means less amount of light is needed to hit these larger targets.
In the example described above, the pixels on the INFINITY3-3URC camera are of 4.54 x 4.54 μm, while the pixel size of the INFINITY1-3C camera is of 3.2x3.2 μm each. QE and pixel size are usually found on the camera’s datasheet.
The difference in exposure time demonstrates the sensitivity level of one camera from the other. One advantage to having a high sensitivity camera is that less light is needed to illuminate a slide, as in the case where the maximum illumination is restricted due to the lamp wattage of the light source, or where particular optical filters may be used.
In addition, a more light-sensitive camera enables shorter exposure times, which directly influence the frame rate and thus, the refresh rate of the image on screen during framing and focusing operations. To put it simply, the camera will work at maximum frame rate in lower lighting conditions.
An important element for producing high-quality images is to reduce the camera’s noise levels through its design. The minimum noise value for the camera is established by read noise, rendering anything below minimum threshold unusable.
This type of noise is most relevant for short exposure times and is measured in electrons (e-). Dark current noise is a temperature and time dependant noise value that is caused by heat within the camera.
It is expressed in electrons per second (e-/s), and increases over time and at elevated temperatures. Dark current noise will make the read noise value less prevalent for long exposures, such as with fluorescence confocal microscopy.
The appearance of noise in an image can be best understood by using an example. Shown below is an enlarged part of the image from the first example taken with Lumenera’s INFINITY1-3C camera.
In the left image, it can be seen that there is a low noise level and the detail in the bottom right corner is sharp and clearly visible. The image on the right side has simulated noise, rendering the patterns and shapes in the tissue indistinguishable.
Featured Above: Comparison of an image with a normal amount of noise (left) vs. an image with a lot of simulated noise (right). Both taken with the INFINITY1-3C.
One can obtain best results when the signal-to-noise ratio is as high as possible, with its upper bound defined by the dynamic range of the camera. The dynamic range refers to the ratio between the maximum amount of electrons that can be produced by a pixel and its noise floor.
It shows the ability of the camera to create usable data in both bright and dark areas of the image at the same time. A high level of sensitivity in a camera can be achieved only when the noise floor is as low as possible, thus enabling the sensor to detect even a small amount of incident light.
Furthermore, a high QE will convert more photons to electrons, thereby providing a stronger signal. The combination of high QE and low noise will result in a camera with a high signal-to-noise ratio, offering excellent image quality.
A camera with sufficient resolution should be selected to image a large part of a slide in a single frame, while maintaining detail and clarity of the specimen. However, it is important to achieve a balance between sensitivity and resolution as these are codependent.
For equal dimension sensors, one with a higher pixel count will essentially have smaller pixels, making them less sensitive to light. Therefore, it is important to choose a camera that has adequate resolution, but not too much, to maintain a high sensitivity level.
The selected resolution should be sufficient enough to differentiate the smallest amount of detail that is anticipated to be encountered in the application. This approach will help ensure that nothing is missed owing to lack of clarity in the image.
In the example given below, the pixel size between the INFINITY3-6URC and the INFINITY3-3URC cameras is identical; however, the sensor size of the INIFINTY3-6URC is larger (1” vs. 2/3”), and thus, has a higher total number of pixels.
Field of view comparison between Lumenera’s INFINITY3-3URC and INFINITY3-6URC microscopy cameras
The image resolution will also be affected by the selected objective for the microscope. As the numerical aperture (also called ‘NA’) is increased, the ability of the objective to resolve two small areas of the specimen close to one another is also increased.
This is because smaller NA objectives become diffraction limited when trying to resolve small details in the image. Though lower NA objectives enable less light to reach the camera sensor, it will increase the depth of field, enabling the observer to focus on a thicker part of the specimen at one time.
A higher magnification objective can also help resolve the smaller detail, but it will considerably reduce the field of view. This is the reason why a large NA is required when viewing the fine details of the specimen, while maintaining a large field of view.
An often overlooked, yet equally important feature of a camera is its interface – how it connects to the computing environment.
In a laboratory environment, when working within a budget and attempting to reduce the costs as well as simplifying the equipment set up, it is vital to choose a camera that eliminates the need for a custom peripheral or interface card to communicate with the PC.
Since computers are already equipped with this technology, using a standard interface, such as USB, will help control equipment cost and complexity. In addition, this also provides ease of integration by offering a more hassle-free, user-friendly, plug-and-play set up.
After considering bit-depth and resolution, users may want to consider which USB version is suitable for their particular applications. Large data payloads can now be moved at high transfer rates, thanks to recent advancements in USB 3.0 that is about 10 times faster than its predecessor, USB 2.0. Most new computers are factory equipped with USB3.0 ports, or for older computers a PCIe expansion cards offering USB 3.0 interfaces can usually be added.
Image analysis software
Users will want to make sure that their cameras are integrated with image analysis software, or work with third party software products. The INFINITY cameras from Lumenera come complete with easy-to-use software at no extra charge, and are also integrated with leading software technology partners such as Media Cybernetics (Image Pro Premier), MicroManager (ImageJ), Molecular Devices (Metamorph), and National Instruments (LabVIEW). The INFINITY camera models also support image acquisition into patient records management software packages that include the ability to use a TWAIN capture interface.
Total cost of ownership
Price always plays a role in product selection when shopping for any new product. When selecting an imaging solution, aspects such as image quality, image analysis software, ease of integration, warranty, and product support should be considered into the total cost of the imaging solution.
These imaging and budgetary factors can easily be addressed by pathologists with Lumenera’s range of USB 2.0 and USB 3.0 INFINITY cameras, all of which come with a 4-year warranty.
Sample pathology images
BPAE Fixed Cell taken with the INFINITY3-3UR
Why choose Infinity?
Lumenera’s Infinity cameras offer the following benefits:
- Compatible with any microscope with a c-mount
- Very easy to install
- User friendly, intuitive software included, for both Mac and PC
- Simple USB interface with no need for extra cards
- Free software updates from Lumenera’s website
- Industry leading 4-year warranty
Lumenera Corporation, a division of Roper Industries, headquartered in Ottawa, Canada, is a leading developer and manufacturer of high performance digital cameras and custom imaging solutions. Lumenera cameras are used worldwide in a diverse range of industrial, scientific and security applications.
Lumenera solutions provide unique combinations of speed, resolution and sensitivity in order to satisfy the most demanding digital imaging requirements. Lumenera customers achieve the benefit of superior price to performance ratios and faster time to market with the company's commitment to high quality, cost effective product solutions.
Sponsored Content Policy: News-Medical.net publishes articles and related content that may be derived from sources where we have existing commercial relationships, provided such content adds value to the core editorial ethos of News-Medical.Net which is to educate and inform site visitors interested in medical research, science, medical devices and treatments.