Digital Clinical Photography: Part 3 - Getting Close Up
The first article in this series presented an argument to encourage doctors to buy digital cameras and to incorporate clinical photography into their clinical skills and routine practice. The second article provided some clues as to how to avoid buying a camera which wouldnt take good closeup photographs. This third article was written to provide some basic but useful advice about how to use ones newly purchased digital camera to take high quality clinical photographs of skin lesions.
The laws of physics relating to optics conspire to make closeup photography a much more difficult exercise than most other photographic pursuits. The biggest problems associated with closeup photography are firstly getting adequate amounts of light from the object under scrutiny, and secondly, getting that small object in sharp focus. To understand these two critical issues and how they relate to taking good closeup photos one needs to be aware of some basic but important facts about how cameras, both digital and analogue/film, take pictures.
As one takes a photograph, light reflected from an object first passes through a series of adjustable glass lenses (plastic in cheaper cameras!) and then through a mechanical iris, the aperture of which can be quickly adjusted wider or narrower to allow more or less light into the camera. The light then enters the camera through a mechanical shutter which is opened for a determined amount of time. Depending upon the type of camera being used (digital or film-based/analogue) the light finally strikes either an electronic light sensor or a traditional film with a coating of light sensitive silver halide crystals in one or more layers with associated pigments. In a digital camera the electronic light sensors individual light-sensitive electrical elements, or pixels measure exactly how much light and of which colour or frequency has landed on each of them and immediately transfers the huge amount of information to the cameras electronic wizardry. This then converts the information into a picture which can be seen in the liquid crystal display (LCD) screen on the back of the camera very soon afterwards. In the light sensitive emulsion of a film, individual tiny crystals of silver halide in different layers of the film change irreversibly when they are struck by photons of light but the film must then be kept away from all extraneous light until it can be chemically treated (developed and fixed) so that it cannot degrade with loss of information.
Aperture & Shutter Speed
Exactly how much light enters the camera is controlled by the automatic functions of the cameras internal computer which almost instantaneously adjusts the width of the mechanical iris in the cameras lens (i.e. the aperture) and determines how long the shutter is kept open (i.e. the exposure time). The wider the iris is opened and the longer the shutter is kept open the more light can enter the camera to land on the light sensor or the film. If the light sensor in the camera detects a very low level of ambient light the cameras computer may widen the iris aperture and/or keep the shutter open for longer in order to capture enough light to take an adequately illuminated image. Manually controllable digital cameras can be adjusted by photographers to override the automated features of the camera.
Yet another factor relating to cameras, light and photographic images is the ISO rating which is a measure of the intrinsic sensitivity of the electronic light sensor or of the film to light. A low ISO rating indicates a low sensitivity to light whereas a high ISO indicates a high light sensitivity. In a traditional film or analogue camera a fast or very light-sensitive film has a high ISO rating and is made with a thin coating of emulsion containing comparatively large crystals of silver halide which can individually capture many light photons and therefore will react to relatively low levels of light. However the tradeoff is that these large crystals create visibly grainy photos. By contrast, slower films with lower ISO ratings are made with emulsions containing very tiny silver halide crystals each of which can only react to much smaller numbers of light photons and therefore to respond at all there must be a high level of ambient light. However photographs made with this type of film are highly detailed and have no visible grain.
Unlike film which is made with a specific and unchangable sensitivity to light, a digital camera electronic light sensor can be adjusted to be more or less sensitive to light. While this functionality can be used to great advantage it is not without some tradeoff! Due to the limitations of currently available technology, increasing a light sensors sensitivity to light when theres not much ambient light available results in the increased prominence of the internal electronic noise occurring within the sensor. This can result in the appearance of grain or mottling, especially within darker areas of pictures and it is similar to the grain found in pictures taken with film cameras using low ISO film. This problem of grain is worse if the light sensors individual pixels are especially small as is the case with many small and highly portable digital cameras which usually have quite small sensors crammed with many megapixels because manufacturers think that consumers will be attracted to cameras with even more megapixels, a new phenomenon known as the megapixel race! While these cameras may well create highly detailed images in bright light, images taken in low-light settings will be very grainy and of low quality. By contrast those expensive, heavy and bulky digital single-lens reflex cameras have quite large light sensors with large individual light-sensitive pixels which which create pictures with virtually no graininess in all but the darkest circumstances (as described in the previous article). So the take-home message which must be repeated yet again is more megapixels isnt necessarily better - caveat emptor!
When one takes closeup pictures of very small things such as patients freckles or skin lesions, it is because they are so tiny that they cannot reflect very much light back to the camera. Therefore, even in a reasonably brightly lit room its highly likely that extra illumination will be needed to create adequately exposed closeup photos. If an additional source of light isnt used, its likely that the cameras shutter will be held open for far longer than 1/30 of a second, this being the critical level beyond which those inevitable minor movements of the camera while being hand-held or of ones subject (remember, were discussing taking pictures of patients who move at least as much as we do!) will cause ones photos to be blurred and useless. With more light on the subject the shutter speed will be faster and movement of the patient or of the camera will be less of an issue and so it is extremely important to have a very bright source of light shining on the patient to ensure that ones photos are properly illuminated and not blurry because of hand movements.
Nearly every digital camera has a built-in flash light. Unfortunately these flash units are usually totally useless for closeup photography because most cameras cannot adequately restrict the amount of light released by the flash units during closeup photography. As a result, if one tries to use a cameras inbuilt flash to adequately illuminate a skin lesion while the camera is held only a few centimetres from the skins surface, the resulting image will be totally white!
Many digital cameras have a hot-shoe on their top surface and one might assume that buying an external flash unit to sit on the hot-shoe would be the answer to this problem. Unfortunately this is commonly a useless strategy because one must buy a flash unit which can reduce its output sufficiently in response to feedback from the cameras internal light sensor and most of the cheaper cameras do not offer this as an option. Those cameras which do have through-the-lens (TTL) monitoring for flash control are usually the expensive, heavy and bulky digital SLR cameras which have very protruberant lenses which can get in the way and cast a pronounced shadow on the closeup subject when the flash is fired! If one is to use an external flash which has the capacity to control its light output for closeup photography one must use an external bracket and hotshoe, which results in a cumbersome combination of camera, bracket and flash unit!
One option is to use a dedicated ring flash unit with TTL metering. Ring flash units are usually quite expensive, often costing more than $500, but they are capable of providing excellent illumination. One minor problem with their use is that the light source which surrounds the front of the cameras lens so evenly illuminates the skin that all shadow and therefore any clues to the 3 dimensional nature of a skin lesion is lost. Nikon sell an interesting ring light which clips onto the front of some of its cameras to illuminate closeup objects for macrophotography. It differs from a ring flash because it produces a constant output of very white light from a number of powerful white LEDs when it is switched on. Unfortunately it can only be used with a few specific Nikon Coolpix cameras but from the authors personal experience using it with a Nikon Coolpix 4500 it works very well!
How White Is White?
Holding a very bright light with an incandescent globe close to patients will be uncomfortable for them, partly due to the high intensity of the visible light and partly due to the intense heat. By contrast, fluorescent lights are much cooler but they emit a different shade or hue of white, with a more blue-green cast to it than the relatively warmer and redder light emitted by ordinary incandescent light globes. This difference in the colour of white becomes very obvious when one compares photos of skin taken under fluorescent lights with photos taken under incandescent lights. In the former, the skin tones may look slightly yellow or green rather than bright pink or even red skin tones seen in the latter. Nearly all digital cameras can either automatically adapt to the type of light source and many will also allow users to manually set the white balance setting within the cameras menu options according to the light source being used. This enables the cameras electronics to adjust how the image is created so that skin colours are as natural as possible.
However, even if one forgets to correctly set the cameras internal white balance setting for the light source used to illuminate patients skin lesions, it is possible to make adjustments to the images after the pictures have been taken by using appropriate computer software. Of course, this takes extra time (which one just doesnt have in the context of a busy day) and requires more than a bit of familarity with the software, not to mention having to buy the software in the first place! Its actually much easier and quicker to take the photos again with the correct white balance setting than to alter the photos with software once theyve been taken.
To summarise, its important to use a good bright light, keep it close enough to the patient to provide enough light for a well exposed picture (but not so close that its intolerable for the patient!) and ensure that the cameras white balance setting is set correctly before the pictures are taken.
Bright But Blurred!
Next comes the issue of ensuring that the object being photographed is in sharp focus. Along with this issue comes another concept called depth of field. This term describes the range of distance from the camera in which everything is in sharp focus including the object of interest in the photograph, perhaps some things closer to the camera than the object of interest and perhaps also things which are behind the object. Think of it this way: try to imagine looking at a line of dominoes standing on end, separated by 1cm and extending off into the distance directly away from you. Now imagine focussing on one particular domino one metre away from you. With a big depth of field perhaps all the dominoes both 10cm in front of the individual domino on which youve focused your cameras optics and all the dominoes within 50cm behind that individual domino may well be in focus. By contrast with a small depth of field perhaps only one domino in front and 2 behind may be in focus and every other domino in front and behind of those few dominoes in focus will be blurry. The laws of optics also determine that the closer one gets to ones object (and therefore larger the object appears in the image) the narrower the depth of field becomes. This means that ones margin for error in focussing on a skin lesion can be very tiny and it accounts for why so many closeup photographs taken by inexperienced photographers are simply out of focus. It also explains why its possible for the very tip of a nodular lesion to be in sharp focus while the base of the lesion and surrounding skin may be completely out of focus! This can be disconcerting especially if the surrounding skin has features of interest. Similarly if one relies on the cameras automatic focussing mechanism to take accurately focussed images one may be in for a few surprises. The author has commonly had to discard photos of an out of focus flat macule close beside a sharply focussed ruler all of 1-2 mm thick! So be careful exactly what your camera is looking at when it automatically focusses. In some cameras its possible to move the focussing area away from the exact centre of the picture, either up or down or to one side in order to ensure that one can focus accurately. Alternatively, one may have to focus manually on ones target assuming that the camera has a manual focus feature.
One way to increase the depth of field is to narrow the aperture of the iris in the cameras lens. Cheaper cameras often provide no manual control over the aperture, also known as the f-stop, which has a numerical value inversely proportional to the actual size of the aperture, i.e. the larger the f-stop number, the smaller is the aperture. With an infinitely small aperture the depth of field is theoretically infinite but clearly the smaller one makes the lens aperture (or the larger the f-stop) the less light actually gets into the camera. So to compensate for a tiny lens aperture required to increase the depth of field as much as possible one can extend the duration of the shutter opening. However the longer one keeps open the cameras shutter the greater the risk that either the patient or the camera youre holding will move thereby causing blurring of the picture! Faster shutter speeds minimize the risk of blurring caused by unwanted movement but reduces the amount of light. Increasing the cameras ISO or light sensitivity setting can generate unwanted graininess in the picture. Increasing the brightness of the light shining on the patients skin can help but this is limited by the patients tolerance of the light and heat given off by the light source. So all these factors are inter-related and all conspire to make it hard to get a good closeup photograph!
It is possible to use a tripod stand to support the camera so that it remains still throughout an exposure longer than 1/30 of a second but it can take quite a bit of time to set up a tripod (extending the legs, attaching the camera and then positioning the tripod and camera sufficiently close to the patient) and still that wont stop the patient from moving during a long exposure when extra light isnt available. Overall tripods are cumbersome and just not worth the effort in almost all common clinical settings unless one is regularly taking pictures of exactly the same part of a patients anatomy (e.g. for iridology). Sudden tiny movements of ones camera closer to or away from the patient or movements of the patient towards or away from the camera due to breathing, minor tremor or even ones heartbeat can be enough to cause blurred pictures. Some cameras are rather slow at getting the focus and taking the picture (called shutter lag) and if the delay between ones pressing the cameras shutter button and the camera first establishing the correct focus and then finally taking the picture there may be quite a bit of movement of the camera relative to the patients skin lesion! One strategy the author uses to minimize errors of focus due to these extremely minor movements of the camera relative to the patient is to brace the camera against the patient. In this setting the end of the cameras lens may only be a few centimetres from the patients skin and it is relatively easy to extend one or more fingers from ones left hand while it also supports the cameras lens or body. It takes a bit of practice to do this quickly, gently and without casting a shadow onto the patients skin lesion but it definitely helps to minimise movement between camera and patient. Patients will tolerate this level of physical intimacy for reasonable periods of time assuming one maintains a clinical demeanour!
Because its all too easy to take hopelessly blurred closeup photos of patients skin lesions its extremely important that you always look at the photos before you ablate or excise the lesion youve just photographed or before the patient leaves your consulting room so that you can try to take the pictures again, this time in focus. Its better to look at the photos on ones computer monitor because the relatively small LCD screens on the back of most cameras are just too inaccurate to assess adequacy of focus even if one can zoom in a few times. If you dont make it your policy to do this every time you may well end up extremely disappointed to discover that your photos were blurred and useless for the purpose of comparison at the same time as the patient returns some weeks or months later for review of a suspicious mole . The extra time you take to do this will be appreciated by your patient who will probably also enjoy looking at the photos youve taken!
Exactly Where Was That Spot I Photographed Last Year?
Important as it is to document the appearance of individual skin lesions by taking closeup photographs, it is just as important to document exactly where on the body the lesions are (or were). Context photos taken which show exactly where on the face, body or limb a skin lesion is situated are a bit easier to take than closeup photos and are taken just far enough away from the patient so that one can easily see where the lesion is e.g. on the left cheek or on the lower back, etc. It is even more important to take context photos when the patient has multiple skin lesions such as dysplastic naevi which need to be followed up in the future. Some patients have a large number of these lesions and one may first need to take an overall picture of a patients entire back for example. It may also be helpful then to zoom in a little bit closer to take some photos of perhaps four quadrants of the patients back. It can also be useful to draw numbered circles around specific and suspicious lesions using a marker pen (wipes off with alcohol swabs). Finally one might then take highly detailed closeup photos of each individual circled and numbered skin lesion. This attention to detail will make it easy to recognize individual skin lesions and ensure that if one reviews the patient in the future there will be little difficulty recognizing which skin lesion is which and which of the many lesions has changed. This process obviously takes time especially when one has to also check that each photo is adequately exposed and is in sharp focus.
How Big Was That Spot Last Month?
It is also extremely useful to take extra photos with a ruler placed close beside a skin lesion so that future review of the lesion will clearly demonstrate if the lesion has enlarged. For complete accuracy it is worthwhile taking photos with a ruler in both vertical and horizontal planes relative to the skin lesion so that growth predominantly in one dimension can be measured. Finally, with the Medicare item numbers and fees for excision of skin cancers now dependent upon both the site and size of the lesion, and with pathologists often understating the size of a skin lesion because of elastic recoil causing lesions to miraculously shrink in size once removed (while at the same time the hole left in the skin following excision equally enlarges!) taking both context photos and photos with rulers beside the lesion will be adequate proof of ones claims should they ever be questioned by Medicare.
Another principle is to use light and shade to enhance the three-dimensional appearance of lesions. Varying the position and the angle of the illuminating light on raised skin lesions will cast shadows which can enhance ones perception of depth and improve the quality of the appearance of a skin lesion. However its also important to be aware that skin is sometimes a bit greasy and can reflect light so brightly that it can reduce the visible detail of a lesion. This may not be obvious when one is looking through the cameras viewfinder but will be very obvious when one examines the image immediately afterwards on the computer monitor and may force one to take the pictures again! You may also want to document the appearance of blanching which is so typical for a basal cell carcinoma when adjacent skin is stretched away from opposing sides of the lesion. It is possible, although somewhat clumsy, to use the fingers of ones left hand to stretch the skin away from opposing sides of a skin lesion at the same time as the left hand braces the camera securely against the patients body. However, if this technique proves to be too awkward one may need to ask the patient to do the skin stretching!
Finally, if you decide to purchase a dermatoscope specifically for the assessment of pigmented skin lesions you might also want to purchase a camera capable of connecting to a dermatoscope by means of an adaptor. Unfortunately there are not many cameras which can be used for this purpose. As mentioned in the last issue, the author uses a Heine dermatoscope with a well and truly obsolete Nikon Coolpix 4500 camera which is used solely with the dermatoscope. This combination takes excellent dermatoscopic photographs and can zoom in quite a bit to further enhance the magnification provided by the dermatoscope. It is obviously beyond the scope of this article to discuss dermatoscopy in any depth. However if you already use a camera to document skin lesions and if you use or plan to use a dermatoscope then it is logical to consider going the final step by documenting what you see in the dermatoscope using digital photography! And its easier to take good quality dermatoscopic photos than other forms of closeup photos because the dermatoscopes provide excellent light with their built-in illumination and there is little risk of blurring due to unwanted movement because the dermatoscope is obviously resting on the patients skin!
Dermatoscopes are ludicrously expensive and if one has to purchase another digital camera just to use with the dermatoscope thats yet an extra expense. However these work tools are tax-deductible and in the long-term the extra expense is definitely worthwhile! It takes quite a bit of time to become competent at interpreting what one sees with a dermatoscope but being able to preserve or document what one sees with photography can assist the learning process as its then possible to compare ones dermatoscopy photos with those found in textbooks. Relying on ones memory is just not good enough!
To sum up, if you have an appropriate camera, good lighting, and the patience to practice, it is possible to take wonderful and very useful photographs during consultations. While it may take a little while to become efficient at using this photographic technique it can be extremely beneficial to patient management and it is highly likely to improve patient satisfaction.
The next article in this series will focus (sorry about the pun!) on how to get your photos onto your computer, what use you can make of your digital photos, how they can be improved if they have minor deficiencies and how to look after them.
Posted in Australian eHealth