TTAC

Patient Exam Cameras – Whitepaper

 

Definitions & Overview

When considering patient exam cameras as a Telehealth technology, it is important for us to define some of the terminology that we are utilizing when referring to various models, as well as give some baseline definitions that allow for some frame of reference building.

Exam Camera Definitions

We broadly use the term “exam camera” when referring to a device that has live video output and supports one-handed operation. These cameras may have situations that require multiple hand use, but in general these cameras should be able to operate in a one-handed fashion. There are various subcategories of devices within exam cameras that we also explored for this toolkit. We use the term “patient exam camera” when referring to the AMD 2500 and GlobalMedia Total Exam Camera, as they have unique features such as the capacity to freeze images and built-in adequate light sources. These cameras are specifically created for and marketed to Telehealth programs. Another subcategory of exam cameras is “camcorders.” Camcorders are consumer grade video devices that have the added capability of recording media to some form of internal or external memory. Note that whenever patient data is being stored, there are added potential concerns and considerations to be taken into account. “Point-and-shoot digital cameras” or “digital cameras” are a potential subcategory of consumer grade devices that have recording capabilities and can possibly be considered as exam cameras. Lastly “home monitoring” cameras are a type of video device that have a very simple interface with no physical controls. We were specifically asked to include home monitoring cameras as a part of this evaluation. A full list of the exam cameras that we evaluated for the 2010 Patient Exam Camera Toolkit can be found in the “Product Information” section of the toolkit.

Related Technical Definitions

When considering the terms “standard definition,” “high definition,” “frame rate,” “Interpolation” and “progressive scanning” it helps to have some clarity surrounding their meanings and how those meanings apply to the exam camera market.

Standard Definition

Standard Definition is an image or video signal that has a resolution of 640 pixels on the horizontal axis by 480 pixels on the vertical axis when it is NTSC (National Television System Committee) or a standardized American video signal. Standard definition video signals are the kinds of signals that are being sent through S-Video and composite cables. This roughly translates to approximately 300,000 pixels or 1/3 of a megapixel. Therefore, the still images obtained from standard definition video signals are less than a megapixel in size. The aspect ratio of standard definition video signals is 4:3, which means the width to height ratio is 4 pixels for every 3 pixels. Aspect ratios can be important when it comes to doing conversions between standard definition and high definition.

High Definition

High definition video signals have a few different definitions that depend on the manufacturer. Generally various consumer devices may claim high definition output that are either 720 or 1080 pixels high; both are recognized as high resolution. You may also see various letters attached to the numbers, i.e. 720p or 1080i. The letters refer to interpolation and progressive scanning. The numbers are more what is associated with the actual resolution of the signal. The size of the images obtained from a 720 high definition video signal are about 1 megapixel and the images obtained from a 1080 high definition video signal are about 2 megapixels. High definition devices typically use HDMI and component output cables and connectors, and may require signal converters to connect to standard definition devices. The high definition aspect ratio is 16:9, which means the width to height ratio is 16 pixels for every 9 pixels and typically appears as a widescreen image.

Frame Rates

Frame rate is the number of frames displayed per second, the frame rate number varies by device. Some common numbers are 24, 24, 30, 50, 60, and 72. On their own, these numbers are not especially clinically relevant, but become more important when in the context of how different images in the video signal are scanned.

Interpolation

When you see a high definition video signal designated as either 720i or 1080i, the “i” refers to to interpolation. Interpolation is a way of generating a video stream, where essentially every other horizontal line is scanned every other frame. Consider a 720 pixel tall image where you number every pixel, and then think about the image as even and odd rows of pixels. On the first frame all odd pixels will be scanned, then on the next frame all of the even pixels will be scanned, therefore really only ever refreshing half of the image every frame. This is important when thinking about the conversion of signals that are shot at higher frames per second rates; you can see an artifact where the horizontal lines appear uneven and the image jagged. The image is refreshing at such a fast frame rate that the odd and even lines don’t get a chance to match up. If a high definition signal is 1080i, it can have decreased quality over a 1080p signal.

Progressive Scanning

When you see a high definition video signal defined as 720p or 1080p, the “p” stands for progressive scanning. In progressive scanning, the entire image is being scanned every single frame. Users do not experience the artifacts that are typically seen with interpolated video signals.

Product Overview

Back to Top

Patient Exam Cameras

For the purposes of this review the patient exam camera devices are the AMD 2500 and the GlobalMedia Total Exam Camera. These devices utilize S-Video signal output. These devices have external power sources that take a DC power input. They have special built-in lights. The built-in lights are significantly different than those offered on other camcorders, because they are actually built in around the lens. This makes the light source so it is shines on axis with the camera, and provides a very strong fill light that doesn’t have some of the problems that you might encounter with a light that is off axis or a flash. Therefore, during macro and intracavity imaging you can get a very well illuminated image without shadowing and other artifacts. These units also have fairly intuitive manual focusing options, which allow the user to fine tune focal lengths which is a very clinically relevant strength. These patient exam cameras also have a unique feature called polarization, which blocks light reflections and the loss of data that can be associated with them in video signals. Another strength these models have is their ability to pause or freeze frame images during a live video encounter. This allows the user to freeze and pause the image they are trying to convey during live video, which ultimately translates to higher quality images minus both clinician and patient movements. When a video signal is being transmitted live even small movements can potentially degrade the video signal, so this freeze or pause feature is very clinically significant. These cameras don’t have in-camera recording capabilities; however there may be other options for hooking these devices up to outside sources to record data. Depending on your program’s view on patient data storage practices and data storage capabilities, this may or may not be a clinically relevant issue for you. The final element that stands out when considering these devices is the fact that they are relatively expensive. Oftentimes they leave programs looking for alternative exam camera options that provide adequate technological solutions and that can still meet their programmatic needs.

Camcorders

Camcorders comprise a large market made up of consumer based video products. There is little standardization involved with features, recording, cables, and video outputs. This wide market base and lack of standardization can make selecting one of these units for your program a more complex and difficult decision. The video outputs on newer camcorder models support high definition and the HDMI output seems to be fairly standard on those high definition units. These newer camcorders tend to simultaneously have standard definition outputs as well as well as high definition outputs. The standard definition output range from an actual S-Video output or composite cable directly in the camera or they may utilize what they frequently refer to as an A/V output. The A/V output can range from a proprietary output to a 3.5mm stereo output, both of which can be converted to an S-Video or composite video signal. S-Video and Composite are typically the input type on various videoconferencing units. Some of these devices have built in lights, but the clinical utility of the video lights are lost on the camcorder units. The lights tend to be underpowered for illuminating farther away and for up close imaging their location on the camera is not on axis with the camera lens. For up close imaging, the way that the lights are situated makes it so they don’t handle macro imaging well. The TTAC team found it necessary to use external lights to compensate for poor lighting on the camcorder models reviewed. Another differentiating factor is the fact that these devices can record and store data on both internal and external memory sources. This feature would need some policy development surrounding it, but could also be clinically useful in certain situations explored later in this paper. There are a number of focusing options with the camcorders; many have some degree of decent to good auto focusing options. We feel that when considering focusing in clinical imaging it is important to have both a automatic and manual option available. Some of the higher end camcorder models offer some degree of manual focusing, however to get to the functionality the user must perform manual menu manipulations. There is a feature in the auto focus mode that can be described as touch screen focusing, where you look at the LCD display and touch the area you want to focus on. The camera will then select the appropriate focal length for that area and adjust the image. This feature can be finicky, but gets better and more responsive with the newer models available. None of the camcorders have an easy mechanism to pause live video. You do have the option of capturing an image or capturing a live video and using menu options to go back to review it. These options require 2 to 4 steps at best typically between a live video and having the image displayed on a VTC endpoint. This is definitely a less convenient option when compared to the patient exam camera models on the market. This functionality may require increased training and device familiarity, but does have some potential clinical benefit as well. Lastly, the camcorders are relatively inexpensive as compared to patient exam cameras, but must be evaluated with caution to understand their other related shortcomings and potential applicability to your program.

Digital Cameras

The digital camera market is not generally thought of in conjunction with the exam camera market, mostly because they are generally viewed as still imaging devices. However, we found that a majority of the devices have some variety of an AV output and in some case an HDMI output. There are potential issues surrounding the fact that some models have the output but can’t display live video output. The TTAC team was able to do some proof of concept work to prove that you can actually use digital cameras to stream live video over different videoconferencing units. As with camcorders, the digital camera market is a larger market with many models and features to choose from. The digital cameras do not have built-in video lighting, but do typically have better built-in flash then camcorders. External lights are necessary in many clinical imaging situations when a digital camera is used as a exam camera. Digital cameras have the ability to record videos and images, the video format tends to be high definition and the images tend to be captured at higher resolution or more megapixels then when they are streamed live. When you compare images sizes to those captured on standard definition signals, there are some definite resolution gains that be taken advantage of with image playback. It takes fewer steps to playback video and images on the digital cameras then on the camcorders. Focusing during live video can be challenging and is model dependent, but note that you do not have the fine level of control that you will have with patient exam cameras and camcorders. Some of the cameras do have the touch to focus feature on their LCD screens, but this functionality was not widely tested therefore we are still unsure of its utility. As with camcorders, there is no easy way to pause or freeze live video. Digital cameras are relatively inexpensive when compared to patient exam cameras and even to some camcorders. However, their expected product lifetime and durability is much shorter than both patient exam cameras and camcorders.

Home Monitoring Cameras

Consumer grade home monitoring cameras will generally have standard definition outputs, typically S-Video and composite. They are pretty well defined by their simplicity and lack of interface, there are no controls and there are no buttons. You can’t freeze or pause live video and you can’t focus the video. We found the intended use of these cameras is generally to do fixed focal point imaging, not up close imaging. The macro capability is completely lacking; you can only typically get 1.5-2 feet away from the subject before the image quality is lost. The lack of macro functionality will not be clinically sufficient as an exam camera for most clinical needs. The home monitoring devices are very low cost devices and can even be described as “cheap” with regard to their construction and durability. We do not recommend the application of home monitoring cameras in Telehealth as exam cameras.

Making Exam Cameras Work with VTC

Back to Top

Frequently exam cameras are used in conjunction with videoconferencing units. There are a few considerations when connecting exam cameras to VTC units. These issues will be both exam camera and VTC device specific.

Videoconferencing Inputs

With videoconferencing there are some inputs that you have to be aware of on the devices themselves. The inputs are frequently labeled as an axillary input; perhaps VCR or DVD player, it depends on the make and model of the VTC unit. When considering some recognized manufactures, older Tandberg units support composite inputs. There are some Tandberg C series codecs that may also support HDMI, Component, Composite, and DVI inputs. This trend of supporting HDMI would be applicable to easily connecting devices like camcorders and digital cameras that have HDMI output. Polycom only supports S-Video input, which may require some type of converter if you have the composite outputs that are on many of the video devices. At this time, Vidyo does not support axillary inputs. They do have some manufacturer provided information that states you can you a “video scaler” device and take multiple outputs and run them through the single HDMI input on the codec. Lifesize room based VTC systems support HDMI, Component, Composite, S-Video, and DVI inputs.

Connecting and Converting

Depending on the exam camera model that you purchase, you may have to look at connecting and converting these various inputs and outputs.

Standard Definitions

If you are connecting a standard definition input on a VTC system to an S-Video connector, patient exam cameras and home monitoring cameras can easily support this connection. When connecting camcorders and digital cameras you may have the need to purchase a converter that converts a composite signal to S-Video or vice versa. You may also need a “gender changer” converter that allows you to change the sex of the cable and may help connect some of the other cables you have along the way. Many camcorders and digital cameras support standard definition connections through their AV outputs. There are also high definition converters that might be required for down conversion of HDMI or component output to standard definition. There are many models of down converters on the market. It is wise to pair the model specifically necessary for exam camera device components you are dealing with when making a purchasing decision. During testing, the TTAC team actually encountered some issues with aspect rations related to a down converter box. If you think back to the definitions section, you’ll remember that the high definition aspect ratio was 16:9. If you can imagine shrinking a 16:9 signal down to a 4:3 signal, you can see how the images could become very tall and skinny. This occurs because the down converter is compressing everything to fit into the standard definition 4:3 aspect ratio. Some converters may offer the option of letterboxing, where it shrinks the 16:9 image to fit in a standard definition image box while maintaining the 16:9 image properties, then fills in the empty space with a black bar above and below the signal. The aspect ratio does not suffer, but the image size does.

High Definition

HDMI and Component inputs will be well supported by camcorders and digital camera models. However, patient exam cameras and security cameras do not support this functionality. You can purchase high definition converters as well. These are up-converters, they take S-Video or Composite video signals and translate them to something that can be playing over HDMI or Component inputs. You do not gain any resolution in up-converting these signals, and again aspect ratios may be an issue. You will be going from a 4:3 aspect ratio and stretching it to the wider 16:9 ratio. This can cause artifact that makes images will appear shorter and wider.

USB Based Converters

Desktop videoconferencing is a currently a growing market, many of the manufactures offer options for desktop video. All of these systems take USB web cameras as their video input. There are also USB converters out there that will take S-Video and composite signals and using a USB dongle or attachment, convert the signals. Depending on the desktop videoconferencing software and USB converter that you are using, you may be able to select this USB converted video as the video image or video source on the desktop video application. This opens up the possibilities for using exam cameras with desktop videoconferencing. You would have to still switch between video inputs in the software and you would lack the functionality of picture in picture, dual video and sharing content; but this possibility may give your program more videoconferencing options.

Making Exam Cameras Work with Store and Forward

Back to Top

Making exam cameras work with store and forward Telehealth systems is another possible use for the exam cameras we evaluated. Store and Forward systems use various interfaces to grab video signals and make them work with computers. Frame grabbers and various file systems are used to be able to access stored media on internal and external memory sources.

Frame Grabbers

With frame grabbers you can capture both standard and high definition video signals onto a PC or laptop using a special card. These cards are often integrated into a Store and Forward system in some way. The requirements you are looking at are very similar to VTC. It is a fairly straight forward connection to S-Video or Composite inputs on these devices, and the same connectors and converters are going to apply.

Full Resolution Captured Content

With full resolution captured content, we are referring to stored and recorded media on digital cameras and camcorders. This is one of the benefits that you gain from using digital cameras and camcorders, you have the option of sending higher resolution content through Store and Forward Telehealth systems. If you can access the device through a USB cable or removable media, you will have the option to send and transmit these larger megapixel images. This is perhaps an atypical use for the exam camera; however it can be a clinically useful use and something that camcorders and digital cameras do support.

Image Comparisons

When talking about image comparison, we viewed this topic on a higher level focusing on the strengths and weaknesses of exam cameras as opposed to individual model comparisons. If you are interested in how the individual cameras performed, we encourage you to take what you learn in this section and look at the “Sample Media” section in the 2010 TTAC Patient Exam Camera Toolkit. In the toolkit you will find side by side comparisons of all of the images we obtained with rating scales and commentary accompanying the images. When evaluating images we were mostly examining how the exam cameras handled image detail and color accuracy. We kept our commentary vendor ambiguous for the most part, because we are looking more at the various classes of device. While we were capturing images we took similar images under similar conditions with all of the exam cameras that we were assessing. We took images of the back, the oral cavity (mouth), the eye, a scar located on a volunteer’s leg, and Vitiligo located on a volunteer’s arm. All of the images were captured live through an S-Video capture card. Therefore, these images simulate what you would be seeing if you were looking at a VTC monitor with a standard definition video signal. The images are not cropped or edited; they are simply straight video feeds that we obtained.

Mouth Images

These images were obtained in a room with mixed lighting; the exam cameras were approximately 4 inches away from the subject. The exam cameras were mounted on a tripod for stability.

Patient Exam Camera

Images obtained using the Patient Exam Cameras in the mouth generally seemed to be clear, lack shadowing, and have fairly accurate image accuracy and color accuracy.

Older Camcorders

Images obtained using the older camcorders experienced white dots in areas of reflection; this is caused by reflection from external light causing all image data in an area to be lost. This happens because some of the older sensors don’t handle reflections especially well and they cause specular highlighting. You will also notice a bit of shadowing, due to the fact that we had to use external lighting to illuminate the subject. The lack of built in light around the lens caused these issues. We noted slightly reduced color saturations and some graininess in the images; however, the image clarity was adequate.

Newer Camcorders

In some of the images obtained you will notice aspect ratio problems, the images appear taller and thinner. These issues were experienced during live video stream by one of the models assessed. On paused playback or image review the images have appropriate aspect ratios, however if you were using this model in live mode you would experience these aspect ratio problems over VTC. We also noted the coloring to be a bit blue; this was cause by the external light we were using. The camcorders would white balance against the external light and producing slightly bluer skin tones. Again you will experience the shadowing cause by using an external light and a bit of lack of clarity in the images. When images were output in high definition you may notice a black box around the images, this is called letter boxing. Some of the newer camcorders were supposed to handle low light situations well, the images that we obtained on these devices without light did in fact produce fairly decent images.

Digital Camera

Images obtained using the point and shoot digital camera with flash caused more pronounced shadowing in areas of the mouth. These cameras also experience image clarity and color accuracy issues similar to the camcorders.

White Balancing

When we talk about white balancing in these exam cameras, we are really talking about the ability of the exam camera to register color. This is an automated continuous algorithmic process, and sometimes a manual process, where the exam camera tries to find true white in an image and then balance the rest of the color in the image to it. The GlobalMedia Total Exam Camera uses a unique “skin tone balance” feature, where it balances the colors in the images based on the skin tones observed. In general, the exact performance of how well these devices picked up skin tones and had accurate colors depends upon the exact lighting situation, the distance, the subject, and the background. There were a lot of things that impacted the color composition of images obtained. Even in the patient exam cameras and the higher end camcorders white balance is a potential struggle to be aware of.

Back Images

These images were obtained in room with mixed lighting; the exam cameras were approximately 5 feet away from the subject. The exam cameras were mounted on a tripod for stability.

Patient Exam Camera

Images obtained using the patient exam cameras were fairly clear, and allowed you to see the moles on the volunteers back well. There were still some color accuracy issues that we noticed in the images. The back tones were more yellow and the arms experienced some bluing. The coloration issues may be due to the mixed lighting situation that the images were obtained under.

Digital Camera

We noted a lack of detail in some of the moles. Similar color accuracy issues were noted, with the back more yellow and the arms experiencing some bluing.

Newer Camcorders

Similar lack of detail was noted in the moles. However, the coloring tended to be a bit more balanced then the digital camera images.

Vitiligo Images

These images were obtained in room with mixed lighting; the exam cameras were approximately 4 inches away from the subject. The exam cameras were held in the imagers hand, therefore lacked some stability and experienced some imaging issues related to movement.

Patient Exam Camera

Images obtained using the patient exam cameras appreciated the finer details in the skin and the hair follicles. The exam cameras tended to experience white balancing issues and has some color accuracy issues in the images. There was also some specular highlighting noted in some of the images.

Older Camcorders

Images obtained using the older camcorders experienced color tones that were overall a bit more brown and pigmented as compared to the subjects skin. In the older camcorder images you can appreciate the finer details in the skin.

Newer Camcorders

The images were lacking a bit of image detail and experiencing some degree of color inaccuracy where the skin tones were hyperpigmented as compared to the subject.

Polarization

Only the patient exam cameras offer this unique feature. The AMD 2500 allows this feature to be turned on and off by the user. The image polarization feature reduces specular highlighting in reflective situations. In images where the polarization feature is on, you may lack some image depth and detail but the color accuracy is somewhat better.

Image Recording

In general, the patient exam cameras outperformed the camcorders and digital cameras with their macro capabilities when live imaging in up close situations. In this section we are thinking of ways to take advantage of having images recorded on camcorders and digital cameras to compensate for their live macro performance. When the internal and external memory of the camcorders and digital camera were used to capture and store images, they are captured in a much higher resolution or as larger megapixel images, then the standard 1-2 megapixels images that are transmitted in a live video signal. When the playback menus are accessed on these devices, the user has the ability to use the zoom functionality and zoom in with much greater resolution then could ever be obtained in a live video stream. Once the images were viewed on playback mode, they were on par with and in most cases better than the patient exam camera images. There are multiple menu selections necessary to get to these images in the camcorder models. Accessing these images in digital cameras is a slightly simpler process, but still requires at least one or two button manipulations.

Usability Review

Back to Top

For usability of the exam cameras we looked at two overarching categories, mechanics and ease-of-use. In mechanics we are looking at how the device feels in the hand and how durable and reliable the devices are. In ease-of-use we are looking at characteristics that point to how easily the device can be used as intended. We know that oftentimes there are issues when a device is too complex or too cumbersome to easily use, this can lead to under utilization or lack of utilization of a technology. Each area had individual metrics that were assessed on a 1-5 likert scale with 5 being the best. For definitions of what we were looking at for each of the metrics we used, please refer to our metric definition page.

Mechanics

We reviewed all of the cameras on a predefined set of metrics related to mechanics. The metrics were: overall feel, appearance, material, durability, mounting brackets/hardware, accessibility of ports, durability of ports, power and button layout and configuration. The patient exam cameras did rather well in the mechanics ratings, both averaging over a 4 on a 5 point scale where 5 is the best. They both have different approaches to button configuration, with the GlobalMedia being very simplistic with 2 buttons and the AMD being a bit more complex with buttons on the back end of the device. Most of the individual details come down to preference and depend on your programmatic needs when it comes to making a decision on the devices. The mechanics of the camcorders and digital cameras were also rated fairly well; we found them all to be fairly mechanically sound devices. We did find them all to be acceptable devices based on our ratings.

Ease-Of-Use

We reviewed all of the cameras on a predefined set of metrics related to ease-of-use. The metrics were: powering on/off, settings, connectivity and cables, recording and playback, ease of image freeze/review, lighting, macro, focus, color balance, and zoom. We did find the patient exam cameras easier to use than the camcorders and digital camera. The camcorders and digital cameras had mixed results; we felt for the most part that they were generally acceptable. Both the INSIGNIA and the GE Security cameras had lower overall ease-of-use scores.

Back to Top