Are Smartphone Pics Good Enough to Help Diagnose Pediatric Skin Conditions?

Patrick McMahon, MD


May 16, 2019

This transcript has been edited for clarity.

Hi. My name is Patrick McMahon. I am a pediatric dermatologist at the Children's Hospital of Philadelphia (CHOP). I'd like to talk to you about a study published in JAMA Dermatology[1] that sought to evaluate whether parents can take high-enough-quality photographs to allow pediatric dermatologists, like myself, to render a diagnosis. Can we practice teledermatology directly from a family to the doctor? There is a need for faster turnaround and better access for families, specifically for a field like pediatric dermatology, which experiences long wait times.

The study was conducted with 40 parents and their children who came in to our pediatric dermatology clinic here at CHOP. We asked them to upload a picture of their child's skin condition to their child's online medical record while they were there in the office. The staff enrolling them tried to simulate what it would be like if they were doing this from home. They asked them a few simple questions, such as how long the skin rash or condition had been there and some symptoms surrounding it.

We provided half of the families with an instruction sheet with some tips on how to focus the camera and take the images. The other families received no instructions at all.

Then the patients and parents went through their normal clinic visit, saw one of my colleagues, and had a diagnosis provided. That was all put in the chart. At a later time, I reviewed the information in a blinded way, based on the photographs and information provided, and "guessed" the diagnosis.

We also graded the image quality to see whether there was a correlation between poor image quality and inability to render an accurate diagnosis.

We found that smartphone cameras really are good enough. Whether or not we provided instructions, the image quality was actually really high. There was a trend toward better diagnoses in people who were provided with the instruction sheet, but it was not statistically significant.

There was an 89% concordance rate between the in-person diagnosis and my virtual diagnosis, which is reassuring.

Three cases were excluded on the basis of poor image quality. In my mind, those were cases which, in real life, would have required more photographs or more information in order to provide a diagnosis.

I think this really speaks to the utility of using smartphone technology and virtual telemedicine to reach patients when they need us. One of the pain points in my position as a specialist is that I sometimes don't see patients until weeks or months after they need me. They will show me excellent photographs on their phone, iPad, or some other device that I wish I would have seen months ago.

This [process of teledermatology] helps us meet them when they need us. We wanted to validate that this was a service that we could provide with high quality. Thankfully, this study helped us answer that question.

I look forward to further studies in the future to validate this through real-life telemedicine encounters. We're piloting some of that already so that we can start practicing this kind of medicine to help improve access for patients when they need us. We also hope to build out a process that is more integrated with our medical records.

Thank you very much.

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.