Facial recognition has been capturing more than its fair share of headlines recently and most of the attention focuses on the negative aspects of this rapidly evolving technology. While we acknowledge that there are potential issues relating to privacy, civil rights, the maturity of the technology, and possible misuse of facial recognition technology, there are many interesting facial recognition and monitoring applications that are and will soon be making positive impacts – especially in the healthcare space.
What is facial recognition?
Facial recognition is a category of biometric software that maps an individual’s unique facial features mathematically and stores the data as a faceprint, like how fingerprints are captured and stored. Using special algorithms and powerful software, a live video capture or digital image is compared to the stored faceprint to attempt to discover or validate a person’s identity.
Facial recognition has become more commercially viable only recently due to:
- High quality cameras come down in cost or are embedded in more devices
- Recognition algorithms are getting better with improved artificial intelligence
- Facial recognition requires large amounts of computing power to process and compare “real-time” images to a database consisting of millions of faces. As computing costs are dropping, computing resources can easily be accessed through APIs in the cloud which makes it easier for facial recognition to be embedded in more technology and applications
What are the key use cases for facial recognition in healthcare?
During our technology visioning sessions with our healthcare clients, we regularly discuss how facial recognition and monitoring will improve patient care, free up staff time for more value-added services, and create a better experience for patients, their families and healthcare staff.
Some of the use cases where we see facial recognition playing a key role now and over the next several years include:
- Patient check-in process
- Real-time emotion detection
- Patient tracking
- Managing chronic pain and medication usage
- Diagnosis of certain diseases or conditions
- Staff identification
- Securing facilities
Patient check-in process
Using facial recognition, the hospital check-in process could be greatly simplified by letting a person admit themselves to a facility without filling out any paperwork or showing identification – they would simply let the system scan their face. In addition to speeding up the process, facial recognition could be used to verify their identify and allow the provider to personalize their experience.
Real-time emotion detection
Another case is to use facial recognition to evaluate facial cues to interpret the emotional state of patients including emotions such as anger, fear, disgust or sadness to aid in providing appropriate assistance. This technology can also be used to improve patient safety by identifying patients who are at risk of unsafe behaviors like accidentally removing a breathing tube and alerting caregivers when a specific emotion is detected.
Facial recognition can track patients and others within a facility without the need for a physical tracking device. The system could identify whether a specific person should have access to certain floors or other restricted locations, or if they have wandered outside of their pre-defined ‘territory.’ This use case is especially relevant in nursing homes or assisted living facilities as a way of keeping patients safe.
Managing chronic pain and medication usage
Facial recognition could be used to scan a patient’s face to determine their pain level in order to manage chronic pain and medication usage. A smartphone app called PainChek assesses the pain levels in people with dementia who have challenges with verbal communication and develops a pain score in real time which can speed up the diagnosis and treatment of pain.
Diagnosis of certain diseases or conditions
Facial monitoring and eye-tracking combined with artificial intelligence is revolutionizing medicine by recognizing and diagnosing diseases. This technology takes advantage of modern computers to analyze, sort, and find patterns across huge amounts of data and serve as an extension to a doctor’s experience and knowledge allowing faster and more accurate diagnoses.
Face2Gene is one company that takes advantage of the fact that so many genetic conditions have a tell-tale “face”— a unique constellation of features that can provide clues to a potential diagnosis. Their software can diagnose up to 50% of the 8,000 known genetic syndromes using facial patterns.
Another company called RightEye is developing an automated autism test. The model’s eye-tracking test recently predicted autism spectrum disorder 86% of the time in more than 400 toddlers. This tool is not yet approved by the FDA, but the early results are very promising.
Pairing facial recognition algorithms with cameras in patient rooms can also verify who is in the room before displaying patient health information on video screens. This use case could also activate specific activities such as turning on the charting dictation whenever a doctor enters the room.
Facial recognition technology can be used to scan all visitors entering the facility to detect and identify individuals of interest to foster better security. It could be used to identify individuals and match them to a list of individuals who are known drug seekers or others who are no longer allowed access to the facility. Facial recognition can also ensure that unauthorized people do not gain access to restricted areas.
Where does facial recognition go from here?
Like so many other emerging frontiers of personal data, there are many benefits and conveniences associated with facial recognition technology that need to be balanced with making sure that implementations address the important issues of HIPAA compliance, privacy, civil rights, bias and legality.
It’s inevitable that facial recognition technology will become more pervasive in everyone’s lives in the coming years, and here are some guidelines that we will follow as a firm:
- As consultants, we must be open to and continue to seek ways that facial recognition can make peoples’ lives better. At the same time, we must also be aware of the risks of technology use to help design mitigations , as well as ensure compliance with applicable privacy and security regulations.
- As technologists, we need to be cognizant of and highlight abuses of facial recognition technology when we encounter them and continue to raise awareness for strong protections around privacy and civil rights. We need to make sure that proper safeguards and assurances are included in all implementations of this technology.
- As consumers, we need to pay attention to what parts of our identities are being tracked, where our identities are stored and managed, how and where the information will be used, and who will have access to that information.
- As citizens, we must keep track of legislation and advocate for privacy rights and laws that include proper protections while harnessing the power of facial recognition and artificial intelligence for good.
Our healthcare thought leaders, Principal Phil Crompton and Associate Alexis Diamond, contributed to this post.