According to a story from cacm.acm.org, researchers have been slowly developing an AI software that can diagnose diseases simply by watching the way in which a person types on a computer keyboard. Such a program could help lead to the diagnosis of many diseases, particularly movement disorders, earlier than currently available techniques.
Diagnostic Potential
The software could have implications beyond just disease that specifically affects movement, however. It may be possible to diagnose mental health issues such as depression and schizophrenia as well. This may seem hard to wrap your mind around, but detections technologies are so precise that they can easily detect subtleties in typing activity down to a single millisecond. Combined with the learning abilities of AI, even slight abnormalities in typing can become a useful pattern that the program can pick up on.
Research So Far
The idea’s potential effectiveness has already been displayed in research, such as this study conducted at the University of Texas. An AI program was able to detect early stage Parkinson’s disease by recording keyboard data from patients over time. Typing data has the potential to be a valuable aspect of medical records for many rare patients in the future if this technology is ultimately adapted for broader use. Tracking feedback from touch screens could also be utilized in a similar fashion.
Another research project has used touch screen data, such as swiping, tapping, and scrolling (along with keyboard strokes) to detect schizophrenia. Typing analysis has also been used to detect Huntington’s disease, heart problems, and sleep disorders.
Faster Diagnosis, Better Treatment
When it comes to Parkinson’s disease, progression of this illness can lead to serious declines in steadiness and coordination, including with hand and finger movements. Inconsistencies in the pace and rhythm of typing, as well as typing speed, can be vital indicators of Parkinson’s, even in its early stages. Over time, patients must type in a manner that is less efficient thanks to slowed movement and stiffness. Diagnosing with keyboard and touch screen data also leads to substantially earlier detection of Parkinson’s disease. In progressive diseases, early detection means early treatment which can greatly slow the course of the disease and preserve a patient’s functionality for longer.
