Can we use a smartphone for clinical imaging?
Clinical photographs play a crucial role in orthodontic treatment. We commonly use these images for diagnosis, communicating with patients, and monitoring treatment progress. Traditionally, we feel that digital single-lens reflex (DSLR) cameras are the best option for capturing these photographs. However, advancements in smartphone camera technology have led to a growing trend of using smartphones for clinical photography. Despite this shift, we still do not know if there are any significant differences between images taken with smartphones and those captured with DSLR cameras.
A team from Cambridge, London, and Bristol, in the beautiful South of England, examined this interesting subject. The EJO published their paper.
Smartphones or digital SLRs for clinical dental photography: is there a difference?
Kirandeep Nandhra et al,
European Journal of Orthodontics, 2025. https://doi.org/10.1093/ejo/cjaf033
What did they ask?
They did this study to
“See if there was a difference in the image quality, dimensional accuracy, colour shade accuracy and ease of use of DSLR and smartphone cameras for orthodontic images”.
What did they do?
They did a cross-sectional study to evaluate images from Smartphones and DSLR cameras using the following stages:
- They identified the four most popular smartphones and three semi-professional DSLR cameras.
- An experienced operator then took standard orthodontic images using the cameras.
- A panel of five experienced clinicians assessed the image quality and colour shade accuracy of the images using visual analogue scores.
- Three experienced operators evaluated the ease of use of the cameras.
- They then used the manufacturer’s information to evaluate the sensor size.
- Finally, they looked at dimensional accuracy by taking images at a standard distance from the patient. They then evaluated image distortion between the cameras.
The statistician then constructed linear mixed effects models to evaluate the outcomes. At this point, the statistical analysis was beyond my understanding. But it seemed logical.
What did they find?
These were the main results.
- The DSLR had a greater sensor size than the smartphone cameras
- They pooled the data for the types of cameras. When they examined the image quality, the DSLR cameras were superior to the smartphones for every intra- and extra-oral view.
- There was no difference in the ease of use of the DSLR and the smartphones.
- The DSLR had better dimensional accuracy than the smartphone.
- No device was better than the other for colour shade accuracy.
Their overall conclusion was
“A DSLR camera, with a ring flash, should be considered to be the “gold standard” for clinical photography, due to the reduced quality rating and image distortion produced by smartphone cameras”.
What did I think?
Sometimes, the most interesting and useful studies are those that are straightforward. This study fits into that category. The methodology was clear and relevant to the study’s aims, and the team produced a well-written paper.
Importantly, they acknowledged the limitations of their study. Most were minor, but they noted that while the use of Visual Analogue Scales (VAS) was established and relevant, they did not ask the assessors to explain their rationale for the scores.
Nevertheless, this approach allowed them to analyse the data with minimal “white noise.” This is a valid point. However, it also meant that they did not assess whether the images were clinically acceptable.
As a result, we can conclude that while DSLR cameras may be “better” than smartphones for imaging, we still do not know if smartphones can be used effectively. This ultimately depends on our personal standards. If we aim for the highest quality, we should use DSLR cameras. However, if we are willing to accept lower-quality images, smartphones may suffice.
In short, this situation is akin to clinical treatment; our decisions are influenced by our standards and abilities. This paper points us in the right direction.
Key points
My blog software can generate key points for each post. I am trying this for the next few posts. Can you let me know what you think in the comments?
- Methodology of the Study: The researchers employed a cross-sectional design, capturing images using popular smartphones and semi-professional DSLR cameras. These images were then assessed through expert evaluation, sensor analysis, and distortion measurement.
- Comparison of Smartphone and DSLR Cameras in Clinical Photography: A study evaluated whether there are significant differences in image quality, dimensional accuracy, color shade accuracy, and ease of use between smartphone cameras and DSLR cameras for orthodontic images.
- Key Findings on Image Quality and Accuracy: DSLR cameras produced superior image quality and dimensional accuracy compared to smartphones, though no differences emerged in color shade accuracy, and ease of use was comparable for both device types.
- Implications of the Results: While DSLR cameras are considered the ‘gold standard’ for clinical photography due to better image quality, smartphones may still be adequate depending on individual standards for image acceptability.
- Study Limitations and Conclusions: The study acknowledged limitations, including not assessing clinical acceptability of images, but overall suggests that DSLR cameras remain the preferred choice for high-quality clinical images, whereas smartphones could be acceptable depending upon standards.
Emeritus Professor of Orthodontics, University of Manchester, UK.
the issue of using smart phones and not mentioned, and then indemnity companies have highlighted, is data security. Smart phone are not the most data secure. IIRC in UK and Europe GDPR applies and meaning that any data has to be on UK/EU sited data servers
Maybe I’m old-school as I learned on a DSLR, but I much prefer images taken with a camera than a phone. I agree with Ross, data security and GDPR would prevent me using a personal phone, I think you’d need one specifically for that use and to stay within the clinic, but the risk of abuse is high. I don’t know how the apps for aligner companies work, if they just keep the photos within the app?
433 / 5 000
Most photographers who shoot with a DSLR camera use a long (fixed) focal length of 90-200mm. This is not the case for photos taken with a smartphone, which allows macro shooting across a focal length range from wide angle to telephoto. This is why we often see photos taken with a smartphone at a wide angle with very significant distortion.
I don’t think it’s a good idea to have smart phones in the operatory or in the operating room.
Assume any picture taken on a smart phone will go on the internet!
The message sent to the staff and patients that it is ok to use a smart phone in the clinical space is awful.
Moreover, the patients should not use the smart phone in the clinical space as well!
The overall conclusion that the digital single lens reflex camera produces the higher quality clinical images, will come as no surprise to the busy practitioner. However, the Smartphone is an item that almost every patient-family unit possesses, at home. How many of us already employ the home Smartphone as the best way for the patient to convey a sudden and unexpected problem in the many weeks or even months “between appointments?” It offers the significant advantage to have an immediate decision whether the patient needs to be seen and treated in the orthodontic office, or to simply have the parent clip off the sharp wire or remove the loose bracket? This is emergency tele-orthodontics, where snap decisions may be made by the orthodontist, on line, while neither requiring to physically occupy the orthodontist’s chair, nor to waste clinic time, nor to disturb the family’s equilibrium with unwanted and often superfluous travel. Worth mentioning.
I cannot help but think that the willingness to use smartphones falls into the category of trying to dumb things down and make things easier when in reality this outcome is totally expected. Did anyone realyy think that the quality of pictures taken with a phone was anywhere near the standard that has been around for decades with SLR cameras? Also in this category is the belief that scans with intra-oral scanners can replace photographs. But then maybe I am just “old fashioned”.
It begs the question as why there anre not attempts to raise the bar in the specialty rather than looking for “easy street” all the time
Thanks for starting this discussion, Kevin,
…and as a photographer – I agree the quality from smartphones cameras will never match that of DSLR – now mainly mirrorless cameras.
Here is another challenge:
I moved away from film photography to digital in 1995.
I was able to enjoy the efficiency of the digital clinical imaging workflow. It also allowed me to stop taking study models.
Today when some of those patients bring in their children I can impressively show them their ancient pre-treatment photos quickly.
When I started 3-D scanning with the Itero in 2011, I could see the potential to ditch the digital camera – which I did last year when the Lumina scanner appeared with its photographic quality imaging – if you are concerned that these images look stylised – the original “video” image is still accessible.
So now that workflow has become even more impressive – no more messing around with intraoral mirrors, flash and sterilising retractors and the best part – all that data is now in the cloud.
I’m glad I stayed with the one system over the last 15 years.
… and I am sure many of us have also tried 3D face scanning with the iPhone.
I could elaborate but this is already a long reply.
We’ve now been told (by a corporate) that all patient records older than 11y (ie not been in contact for 11y) are going to be deleted. We can choose to keep them but have to contact patients individually for express permission. It’s so useful to be able to look back at notes and photos from previous treatments, in fact today someone is booked in who last came in 2008, so I am quite concerned at this loss of records.
What was the make and model of each DSLR used and which macro lens was used on each DSLR? In adddition, what ring flash was used on each? To sum up, could we have a look at the meta data generated by each DSLR image?
“The three semi- professional DSLR cameras used were: Nikon 3500D, Canon 2000D and Canon 100D, all with the ring-flash and macro-lens recommended by their respective manufacturer. The iPhone 15 Pro, Samsung Galaxy S23 Ultra, Google Pixel 8 Pro and Xioami 13T Pro smartphones set to their default settings and using their in-built flash were used for comparison. To replicate conventional intra- and extraoral orthodontic
clinical photographs, operators were given the manufacturer recommended settings of the DSLR cameras for extraoral and intraoral photographs. The smartphone cameras were not adjusted as there were no recommended settings for clinical photography. We do however acknowledge that there are smartphone editing applications and
auxiliaries which can be utilized to improve the quality of photographs. These include using third-party apps such as Adobe lightroom mobile with the images in JPEG format and autofocus, adjusting the exposure, shutter speed and aperture setting according to the device and photography conditions, and using a continuous light source, e.g. dedicated LED light. “
Thanks for starting this discussion, Kevin. …And to answer your question, Mark Wertheimer:
Here is a challenging story:
I moved away from film photography to digital in 1995.
I was then able to increase efficiency by delegating clinical photography and improve the whole imaging workflow.
It also allowed me to stop taking study models.
Today when old patients brings in the children I can impressively show them their ancient pre-treatment photos.
When I started 3-D scanning with the Itero in 2011, I could see the potential to ditch the digital camera which I did last year when the Lumina scanner appeared with its photographic quality imaging.
Now that workflow has become even more impressive – no more messing around with intraoral mirrors, flash, sterilising retractors and the best part – 14 years of data – now in the cloud – I’m glad I stayed with the one system.
I could elaborate but this is already a long reply.
Hello John!
Yes, great discussion and I know that you are passionate about the accuracy of clinical imaging. I also switched to iTero intraoral imaging in lieu of DSLR intraoral photographs, much to the pleasure of my patients and staff. I found the scans via MDC (multi direct capture) technology are more accurate (less linear distortion than DSLR- 50 microns) and capture more data including the entire dentition and palate – all in 3D vs 2D.
I made the switch reluctantly to ditch my DLSR for extraoral pics during covid when my battery died and camera shops closed. At a pinch I used a practice ap that was available. Yes, I did get a few “gold fish” faces initially, but the updates and templates for clinical photography have improved to the extent that distortion is clinically undetectable (bench test – photo of parallel grid lines cf DSLR- using the clinical practice ap, not the regular smartphone camera function). Re the safety and wisdom of a smartphone (or iPad?) in the clinic, depends what platform you are using. I show the patients that I am using an integrated practice ap designed for clinical photography that will assist my diagnosis and treatment planning – within 10 minutes I have more comfortably captured in 3D the entire original dentition and palate down to 50 microns, may view “aligned” and / or “restorative” setups, all within the patient’s face, as well as shoot optional speech videos that Zachrisson mandated years ago for benchmark orthodontic diagnostics.
As Kevin acknowledged in the limitations, this study did not evaluate clinical acceptability of images, nor did it take into account the additional diagnostic capabilities of integrated practice aps supported by smartphones. (Hopefully a follow-up). If my goal was only to obtain high quality images, I may venture to resurrect my DSLR. I am not after a photography award (although I challenge most to detect the capture technology) – I need and patients deserve standard of care, reliable, valid, accurate, safe and rapid data capture to enhance diagnostic and treatment capability. I am sure that it won’t be long until we may additionally capture the entire face as well as the intraoral environment with a single scanner, to highest standards of accuracy and patient safety. I only wish Sheldon Baumrind was alive to see his vision become clinical reality.
*VV provides educational sessions on behalf of Align Technology
IT. WILL. NEVER. REPLACE. THE. HORSE.
So, 20 years ago the International Ortho Congress was in Paris and the UK was one of the Orthodontic Villages that did a day of lecturers and they had Sverker Toreskog (not an orthodontist but a prosthodontist) as a speaker and he kicked off his talk by saying that if he would have done anything different with his career, he’d have taken more pictures and he envied orthodontists because they are in the habit of taking pictures and models. For everyone else, it was a chore and impediment.
Even at that, taking the pictures in the earlier part of my career involved sending a patient to “medical photography” in some hospitals or the consultant asking for the camera to be summoned and it would be taken from the back of a cupboard and the mirrors and retractors would be located and settings would be dialled in – it was a disruption.
Then digital came along….
I once had a boss who took these things seriously and said the colour and picture quality of slide film was so far ahead of digital photography there was no way the £10k or whatever the early price was could be justified on a digital camera, even though it was saving hundreds of quid a month on film. Plus there was that decade where everyone in the UK was convinced that people would be “digitally manipulating” their results on photographs.
The effort involved in taking pictures with a DSLR compared to a film camera is about the same chair side, but the DSLR was miles ahead in terms of ease of storage and retrieval afterwards, and eventually the price was right and they became the standard and we’re done with film cameras. There’s a DSLR for each chair and saying “I’m going to need to take some photos” does still cause some eyerolling from the team but only as much as saying “bracket off, LL5” . The paraphernalia is ready within a few seconds and off we go.
BUT…
…a few months ago, I had a patient in with a suspected tumour. I took a few camera photos and a few phone photos. The phone photos were with my oral surgeon probably before the patient left the practice. He knew what he was looking at and advised the guy see him without delay. In terms of trauma, spreading infection, and remote management the trade off between image quality and the stakes will leave the phone superior.
It has gone beyond the convenience of storage and retrieval and on to the convenience of processing and interpreting the image and comparing to other images – particularly if it’s sequential in orthodontic movement.
A DSLR might be the big boy on the block but to a generation of people currently working they are going to look as technologically advanced as the Hogwarts Express. The smartphone will be easier to use and more familiar to them and the image quality – if it matters – could be modified by apps. The data protection needs to be ironed out. But these things are systems more than devices and when a critical mass of users use it, we’re probably going to move to that as the new standard.
I think we just have to find a way to make it work better for the low stakes stuff, because it’s not going away.
Stephen Murray
Swords Ortho