Photo tips for clinical images?
October 25, 2004 12:35 PM   Subscribe

Where can I find some easily understandable tips on taking good clinical images with a digital camera?

One aspect of my job is working with the production and print end of medical illustrations, and I need a way to explain to a largely inexperienced group how to produce digitial images that are usable in a medical journal. Basically, how to take good clinical photos: what settings, lighting, and framing techniques work best to achieve clear images at high-enough resolutions. But I Am Not a Photographer (and yes, I told them so, but they still want my help, so I'd like to be able to provide some guidance). I can tell you all about the minimal resolution and file formats we need, but nothing about actually taking a picture. Are there simple guides, preferably online, that can elucidate the basic principles of good digital photography -- and better yet, how they apply good clinical photography?
posted by melissa may to Technology (7 answers total) 1 user marked this as a favorite
 
Ring lights are good for even lighting with minimal shadows. They're frequently used in macro photography (and to some extent fashion, but those are big, expensive rings). If you're taking clinical images, you might need to focus at short distances, so a macro lens might be useful.
posted by Civil_Disobedient at 6:09 PM on October 25, 2004


I am not a medical photographer, but this is what I would advise in your situation.

(1) Get a camera with the largest LCD possible
(1b) Pay close attention to the image in the LCD as you compose the image. If the image in the LCD looks bad, the picture will probably be bad.
(1c) Since an SLR can't use its LCD for preview, only use an SLR if it gives you access to features that you need and cannot find on a P&S camera.
(2) Set the camera to use minimal compression (high quality JPEG should be fine, TIFFs or RAW images are unnecessarily complicated unless the journal specifically requests it)
(3) Set the camera to use the largest image size possible (6MP if it is a 6MP camera)
(4) turn the on-camera flash OFF. If space is limited, use a ring light mounted around the camera's lens. If more space is available, use separate lights. If reflections or hot-spots obscure important details, move the lights before moving the camera.
(5) Avoid mixing different kinds and temperatures of lights
(6) Take lots of pictures. A good photographer exects about 10% of their pictures to come out as expected. If you need 10 good pictures, take 100 shots.
(6b) Get a memory card that will hold at least 200 pictures at the camera's maximum megapixel setting.
(6c) Never delete a picture while it is still in the camera. Always review images on a computer with a good monitor before deleting anything.
(7) If the object you are shooting is within a meter of the lens, activate the camera's "macro" feature.
(8) Read the camera's instruction manual to find out what its focus and light limits are.
(8b) Pay close attention to the AutoFocus system. Does the camera always focus on the object in the center of the scene? Is there a multipoint AF system?
(9) When you aren't at work, take pictures of plants, insects, wall textures, etc. Look for objects that are similar in size, shape, color, and texture to your medical subject matter.
posted by b1tr0t at 9:16 PM on October 25, 2004


How good do the images have to be? Is this something we're talking about taking w/ a point and shoot camera in the middle of a (time constrained) procedure, or is this something they can/need to spend time setting up?

C_D's suggestion on lighting is good if they can spend money on equipment. Lighting and color balance are both going to be issues, and ring-lighting helps with both.

* Bring your own lighting, unless you're confident shooting w/ the ambient light will give you a good enough exposure and color (aka -- spend some time shooting in the ambient light beforehand)

* Macro lenses help you shoot close-ups of something small, so this is probably what you want. But, short macro lenses require you to get REALLY CLOSE, which may not be what you want. 100mm macro lens should let you sit back a ways and still get a good shot.

* Get your color balance/white balance right. Color will probably count in these photos, so at the very least ensure your camera is setup for the right white balance. If you're super serious, get a grey card and set your white balance based on that.

* Probably the bigger challenge will be loosing detail in the shadows or in the light. You'll want to spot-meter (usually w/ the on camera 'spot meter') the areas you really want to see in the final image. This will probably result in some areas being too dark or too light, but I'm assuming there's one focal point you're most interested in, rather than trying to balance the picture as a whole.

* Bring something to steady your camera - tripod, monopod, whatever. The need for this decreases as your lighting situation improves, but it's better to have it and not need it...

* As for framing, I'd concentrate folks on making the image understandable. No need for Hitchcockian perspective, orient the camera so that the 'up' side of the picture is the 'up' side of the patient. Try to include enough of the surrounding anatomy that a viewer can easily orient themselves to what's going on (in at least 1 picture, probably near the beginning).


Sorry, I don't know any sites that concentrate on clinical shooting. In fact, I've never done anything like that myself, and I'm by no means an expert.
posted by daver at 9:32 PM on October 25, 2004


Is this something we're talking about taking w/ a point and shoot camera in the middle of a (time constrained) procedure...

Bingo. Thanks for asking for the clarification. These are oral surgeons taking shots of bloody gums and teeth during long and complex surgeries, so yes, they're very much in a point and shoot situation.

The most common problems I see from the production end: images that are low-res (at 72 or 96 dpi, generally); too small; irregularly sized so that we can't mount a nice consistent series; or that are badly lit and washed out. All of your answers so far have been extremely helpful at addressing these points and I appreciate it more than I can say. This is a last-minute request and I am feeling very much out of my depth.

Just a point of clarification if you have the time: we accept jpegs, but prefer tiff or eps if possible because they are higher-compression formats. Why are they more complicated to take than jpegs? If this requires a major tutorial, please don't worry about it.

(FYI, we also print histology, but I'm not going to even get into that with them -- I almost never see really good, clear tissue shots, with the occasional exception that's so shockingly crystalline that I figure that the authors who provide them must have really very well-funded labs.)
posted by melissa may at 10:07 PM on October 25, 2004


You can convert from JPG to TIFF with just about any image program or viewer known to man.

Don't worry too much about color balance if you're shooting with a flash or if you're shooting RAW files. Under no circumstance should you ever use the camera's built-in flash for close shots. The light will be off-center, and your image will be underexposed.

Also, don't bother with a tripod if you're trying to shoot around surgeons watching the clock. You just won't have the time to set up. If you have a decent flash or good enough light, you should be able to shoot fast enough that it won't matter.

If you chimp your shots (look at the LCD after you shoot) you'll get a good idea whether you're getting the shot or not.

Keep an eye out for things that will "fool" the meter. Extreme dark or light will make the shot overly bright or dark (respectively). Shooting a teeth close-up, for example, will require you to compensate for the brightness of the teeth. The camera will assume everything is neutral grey, so your shot will come out underexposed. This is the same reason why out-of-camera shots of snow scenes look grey instead of white. Most modern digicams have what's called Exposure Compensation to, well, compensate.
posted by Civil_Disobedient at 10:24 PM on October 25, 2004


Photography guide for plastic surgeons [courtesy of the Wayback Machine, since the source site has fubared their link]
posted by nakedcodemonkey at 10:43 PM on October 25, 2004


Just a point of clarification if you have the time: we accept jpegs, but prefer tiff or eps if possible because they are higher-compression formats. Why are they more complicated to take than jpegs? If this requires a major tutorial, please don't worry about it.

TIFF or RAW images are often substantially larger than a low-compression jpeg, so you can't fit as many on a memory card. More importantly, RAW images must be converted to TIFF or JPEG with a separate program on the computer. This extra step can only add confusion and error to the process. When you set a camera to "high(est) quality JPEG" you get an image that has no visible artifacts, and can easily be manipulated and stored by common tools.

A bit of "photoshopping" can substantially improve digital pictures, particularly when there is poor light. The problem you face is where to add this step in the work-flow. You may be tempted to instruct the medical technicans to do this, but then they've got to fork out $600 for photoshop, a newer computer (don't get me started on PCs likely to be found at medical offices!), and learn good digital-darkroom skills. It is probably more efficient to have one of the magazine/journal staff process all the images prior to publication. This is a slightly more complex decision than this paragraph would suggest.

In summary - formats other than JPEG require more post-processing by the technician. Regardless of image format, some degree of post-processing will substantially improve image quality.

These are oral surgeons taking shots of bloody gums and teeth during long and complex surgeries

If these procedures are anything like my wisdom teeth extraction, then instruct the photographers to select cameras that can work inside a waterproof or water-resistent housing. A camera soaked in blood and tooth powder won't last long.

images that are low-res (at 72 or 96 dpi, generally); too small; irregularly sized so that we can't mount a nice consistent series

Ask for 6 megapixel, uncropped images. This means the journal has to do the cropping, but it also means that the journal gets to choose how the image is cropped. I don't know if it supports a waterproof housing, but you may want to look at cameras like the canon S60 or S70. Of particular note is the lens, which will go to 28mm wide. With a pricy SLR and a $1500 lens, you can only get a little wider - roughly 24mm. A wide lens is extremely useful in close-up shots.

the occasional exception that's so shockingly crystalline that I figure that the authors who provide them must have really very well-funded labs

I've taken some of my favorite pictures with a sub-$100 soviet-era film camera. Knowing how focus and exposure work makes all the difference.
posted by b1tr0t at 12:51 AM on October 26, 2004


« Older Hair-raising   |   Salty Beer Newer »
This thread is closed to new comments.