Thanks for visiting Imaging and Machine Vision Europe.

You're trying to access an editorial feature that is only available to logged in, registered users of Imaging and Machine Vision Europe. Registering is completely free, so why not sign up with us?

By registering, as well as being able to browse all content on the site without further interruption, you'll also have the option to receive our magazine (multiple times a year) and our email newsletters.

Vision is a cut above

Share this on social media:

Topic tags: 

Greg Blackman investigates some of the imaging applications used by the medical profession

The medical profession is ultimately concerned with patient care. The phrase primum non nocere, which translates as: ‘first, do no harm’, is one of the primary precepts that medical students are taught at medical school. Non-maleficence, a term derived from the very same maxim, warns physicians of the possible harm medical intervention might cause and that sometimes doing nothing is preferable to treating the condition. Beneficence is the other side of the ethical coin, and teaches that, while physicians are expected to refrain from causing harm, they also have a duty of care to their patients, and possess the expertise and skill to carry out that duty. Most medical decisions involve a balancing act between beneficence and non-maleficence, between the benefits and risks of treatment.

Without delving too much further into the very murky depths of medical ethics, it is safe to say that, as technology advances and new medical techniques come to light, patient care is improved and the risks associated with treatment reduced. Medical imaging techniques are no different; for example, the speed of line scan cameras used to capture images from Optical Coherence Tomography (OCT) helps minimise the time the procedure takes. Speaking about OCT, Dr Joachim Linkemann, product manager at Basler Vision Technologies, comments: ‘The faster they [the cameras] are, the less time the procedure takes, which improves patient comfort.’

OCT is a non-invasive imaging technique based on interferometry. It is used in ophthalmology, cancer diagnosis and skin examinations to provide a 3D image of the tissue. Near-infrared light (approximately 850nm) is used at a bandwidth of 10-20nm and the coherence length of the scattered light allows slices through the tissue of a few 100μm to be visualised.

In the context of ophthalmology, Linkemann explains: ‘The advantage of OCT is that it is a non-contact technique and allows the ophthalmologist to gain a microscopic 3D image of the back of the retina.’ With headquarters in Ahrensburg, Germany, Basler’s line scan and area scan cameras are used for medical imaging techniques, such as OCT.

One alternative is for the doctor to simply look at the back of the eye using white light. As NIR light is used with OCT, there is no reflex from the iris and therefore the patient’s eye requires no preparation for the procedure. NIR can also penetrate deeper into the tissue than white light.

Linkemann warns: ‘The camera technology must have low noise with no background stripe or wavy patterning, as the signal received is in the form of a stripe pattern.’ In addition, cameras used in OCT must be high quality and sensitive.

X-ray vision

Advances in the sensitivity of X-ray imaging devices – with the advent of image intensifiers and later with flat-panel digital detectors – have resulted in a reduction in the dose of radiation to which the patient is exposed. This is especially important with fluoroscopy, an X-ray imaging technique that provides physicians with real-time images of the patient’s internal structures, as the technique involves radiation exposure over time and therefore a greater total absorption.

Fluoroscopes typically consist of a fluorescent screen sensitive to X-rays connected to an X-ray image intensifier and a CCD camera. Alternatively, flat-panel digital detectors are used, which reduces the need for image intensifiers.

Fluoroscopes generate large numbers of images, which the vision system must be able to acquire and process in real time. Digital Subtraction Angiography (DSA), for example, which is a type of fluoroscopy technique used to visualise blood vessels, works by subtracting a reference image of the area under investigation from the acquired image as a contrast medium is injected into the veins. This allows radiologists to visualise blood vessels as the contrast medium moves through them.

With the increasing use of digital imaging technique, image data becomes much more readily accessible. Image courtesy of Pleora Technologies.

Italray, an Italian manufacturer of radiological equipment, is using image acquisition boards and software from Dalsa in its X-Frame line of digital imaging systems, which includes the X-Frame CCD series for radiology and fluoroscopy applications. X-ray images are acquired by a 1,024 x 1,024 CCD camera connected to an image acquisition board by Dalsa, which provides high-speed image acquisition of up to 60fps. The X-ray images are then sent to Dalsa’s vision processor for real-time processing, primarily recursive filtering and edge enhancement. ‘The Dalsa vision engine, with its independent processor, is very useful because it can be dedicated to real-time processing without overloading the host system, which is used for display and storage,’ says Ciro Rebuffat, software development manager at Italray.

Italray is also developing a line of radiology and fluoroscopy systems based on a flatpanel digital detector. One of the advantages of flat-panel detectors over more traditional fluoroscope setups is their high sensitivity to X-rays and therefore their potential to reduce radiation doses. In addition, detectors can be manufactured to larger dimensions than image intensifiers and, therefore, a flat-panel system can accommodate a much larger range of exams – the device can be used for chest X-rays, for example.

With the increasing use of digital imaging techniques in hospitals and medical centres, image data becomes much easier to access. Pleora Technologies’ iPort video connectivity solutions are used in the medical sector as an interface for diagnostic imaging. It allows the user to transport images and video over a standard IP network in real time, with low system latency and high reliability. For example, a digital X-ray detector can be connected using iPort technology to an IP network allowing the images to be displayed in real time.

Robert Lee, VP of sales and marketing at Pleora, comments: ‘Reliability in collecting and transporting the images, with no loss of resolution is absolutely key.’ Radiologists are concerned with minimising the patient’s dose of X-rays. Using reliable image transport technology is therefore crucial, as retaking the image would result in the patient being exposed to an additional dose of radiation.

Lee notes that the resolution of digital X-ray images must be maintained; in other words, all of the X-ray dosage that the patient experiences must translate into useable image resolution. Therefore, video and images are transferred in an uncompressed format, as image compression can result in data loss or artefacts. Pleora’s iPort technology uses Gigabit Ethernet to transport large image files quickly and reliably.

X-ray image of barium enema, used most commonly to check bowel health. Image courtesy of Dalsa.

One of the advantages with networking the equipment is the improved flexibility it provides – the instrument can be taken to the patient rather than the other way round. X-ray machines can be wheeled into operating rooms, which is especially important with certain injuries, such as spinal injuries, where patients shouldn’t be moved.

With specific medical applications, such as in live surgery where the video feed is guiding the surgeon’s tools, the latency of image transfer becomes especially important. Rudi Rincker, VP of business development at Pleora, remarks that in these circumstances, latency must be reduced to a point where what the surgeon is seeing on the screen is the same as what is happening on the operating table. Pleora’s iPort technology provides low-latency data transfer suitable for this type of medical application.

Vision-guided surgery

Surgical navigation systems are still a relatively new technology. However, according to Claudio Gatti, co-CEO of Claron Technology, which develops vision software for medical applications, the technique has become the standard of care when dealing with neurological or spinal injuries.

Claron Technology’s MicronTracker is a navigational tool for surgeons and uses Point Grey Research’s Bumblebee2 stereovision cameras to track where a surgeon’s instrument is inside the patient. The system operates by overlapping 3D Computed Tomography (CT) or Magnetic Resonance Imaging (MRI) scans, generated prior to surgery, with real-time image data produced by the Bumblebee2 cameras during the procedure. For instance, neurosurgeons will use CT scans to pre-plan a surgical procedure for resecting a brain tumour. During the procedure, the cameras from Point Grey track small, chequered marker patterns, called Xpoints, on the external portion of the tool; these report the position and orientation of the instrument inside the body. The system then overlays the real-time data with the preoperative 3D model from the CT scans, to give the surgeon a virtual view of what is ahead or around the tool.

There are other tracking systems available using infrared light, but the MicronTracker uses vision and passive lighting to identify the marker. According to Gatti, the navigation systems currently in use are large, complex and costly systems. With the computer vision-based approach, the navigation technology is being scaled down towards simpler and more compact systems.

‘There were two main challenges in designing the system,’ explains Gatti. ‘The robustness of the recognition and being able to do it in real time, and, secondly, the precision of the measurement – an accuracy of 0.3mm rms across a field of view up to two metres in diameter.’

The robustness of the recognition is dependent on the detection algorithm and the design of the pattern placed on the surgical tool. In addition, the software needs to be able to adapt to the lighting conditions. ‘Surgical lighting can be very powerful, measuring hundreds of thousands of lux,’ says Gatti. ‘However, this light is often focused on a specific area, so some areas can be very bright while others will be in almost complete darkness.’

Point Grey Bumblebee2 cameras have the functionality to cycle through different exposure modes depending on the lighting conditions. If there is a large range of illumination, the MicronTracker’s high dynamic range mode is designed to handle it.

The precision of the measurement was addressed by a very accurate detection algorithm and high precision custom calibration of the camera. ‘The resulting detection algorithm allows the identification of the Xpoint on the x-y plane with an accuracy of 1-2 per cent of a pixel,’ comments Gatti.

MicronTracker can also use multiple stereoscopic cameras to extend the field of measurement, and each camera can support either binocular or trinocular vision depending on the surgical procedure and the degree of accuracy required.

Gatti notes that this type of computer vision-based product is driven largely by advances in software. In addition, with a steady stream of faster CCD sensors becoming available and higher communication bus bandwidth, this technology could become much more common within hospitals.

Nurse! Scalpel!

In preparation for surgery, surgeons will typically specify the instruments required for the next day’s procedure from a pick list, from which technicians will make up and sterilise a kit containing the necessary surgical instruments. Manually assembling kits of instruments for the variety of surgical procedures performed in a typical hospital can be time-consuming and potentially inaccurate.

The US company Censis, has developed a software product called Censitrac, which ensures the correct instruments are loaded into surgical kits. The system operates through a 2D data matrix barcode, electrochemically applied to each instrument, which is scanned during the assembly process using a DataMan 7500 ID reader from Cognex.

‘In the past we assembled instruments based on a document that we called a count sheet,’ says Pat Stefanik, registered nurse and central sterile manager at Saint Thomas Hospital, Nashville, Tennessee. ‘This task was complicated by the wide range of individual preferences among the different surgeons and by the large number of non-traditional instruments that are used by our surgeons. For example, some like all their curettes together and all their scissors together, while others want all the ringed instruments together.’

The instruments are typically a silvery metal, normally either stainless steel or aluminium. Some have a matt surface while others have a mirror surface. Some instruments are flat and others are curved. The reflectivity of the surface makes it difficult to form a good image of the mark. Cognex’s DataMan 7500 series of ID readers contain an integrated diffuser that provides soft illumination required for highly reflective parts, such as electrochemically etched marks on shiny round surfaces, and incorporates software that handles a wide range of degradations to the appearance of the code.

It’s not simply a matter of ensuring the correct instrument is in the correct kit, but also tracking its movements. In the event of a biological test problem with a steriliser load, the hospital needs to know exactly which instruments were in the load and their current location. Using an automated tracking system also allows a record of which instruments were used in a procedure and to determine exactly when maintenance is due.