Smartphones are Revolutionizing Pill Identification
As a pharmacist, I receive many questions regarding medications from patients and other clinicians. One set of questions that often comes up is, “What is the medication I take that is a green capsule,” or “The patient said they take a white tablet, thats about ¼ inch long,” or “They brought in all these medications to the ER, but don’t know what they are.”
For years, if a pharmacist or other professional wanted to identify a pill or solid dosage pharmaceutical (i.e. tablets, capsules), they had to turn to a drug compendium that included the capability of identifying said pill based on colors, markings, shape, and so on. Its not hard to imagine how tedious the task could be. And often times, these could return anywhere from a handful of potential pills to hundreds. The drawback was that often times, an exact image to compare against was not provided.
However, due in part to the great speed technology has come, many are trying to find a ‘smarter’ way of conducting pill identification. The National Library of Medicine (NLM) has created a project titled “Pillbox,” where they have sought to create an online tool to help users identify medications accurately while supplying high-depth resolution images.
Interestingly, researchers at the NIH, including Jesus Caban and colleagues are investigating novel ways of identifying medications using currently available computational photography. Using the nations prescription formulary, the NIH is looking to create a system in which it would be possible to identify medications through the imprint of the pill, color, and shape in multiple types of viewing conditions. Their results can be seen at their 19 IEEE International Conference on Image Processing (ICIP).
While the NIH is pushing research in this field, there are multiple startups also looking to tackle this field. Recognizing the potential payoffs in investing in a problem that has costly consequences, these startups are coming out with multiple patents on identifying medications using smartphone technology.
The premise of these companies is that as most users carry a smartphone with them, and that the built in sensors and camera have come a long way, there must be some way to use it to identify medications. Several examples of such companies include MedSnap, ID My Pill, and MyPillSense.
Many of these apps require the user to use some standard backdrop to help identify medications. The most advanced I have seen has been the unit created by Medsnap, which based on my interview with Dr. Hymel of Medsnap, allows them to get a better identification. The obvious negative to this however, is that the user is then responsible for an additional item to have with to use the app.
For myself as a pharmacist, I find these new innovations fascinating. It is my expectation that with time, it will become common practice for providers to use their phones to identify their patients medications with nothing more than a picture.
Imagine when that patient comes into the ER with their medications in generic bottles or mixed together – soon, you should be able to take out your phone and accurately identify the medications in minutes. Future directions include the ability to automatically upload this data to your EHR and reconcile medications. There are also potential future uses in the outpatient setting. Visiting nurses would have a new tool to use when they visit their patients at home to verify medications. Patients could use these apps to verify that the pill they are about to take is one they are supposed to. Other uses could also include clinical trials for pharmaceutical companies and as a mechanism for enhanced drug information or interaction checkers.