April 14, 2026
3 min read
Click here to read the Cover Story, “Eye rubbing: Screen for it, address causes, make patients aware of consequences.”
Eye-rubbing detection tool using AI
Eye rubbing is currently subjectively quantified by questioning the patient or using standardized questionnaires.
However, eye rubbing is often an unconscious reaction to a variety of stimuli, and people are mostly unaware of when and how often they rub their eyes.
We recently developed an innovative solution that uses an AI-powered smartwatch application to objectively detect, quantify and prevent eye rubbing, and it was tested in in a proof-of-concept study.
Smartwatches can function as activity trackers through specific sensors, and two sensors are of interest for eye rubbing, the accelerometer and the gyroscope, which detect and measure translational and rotational hand movements, respectively.
Inès Drira
Because each patient has a specific eye rubbing method, we identified and processed three main stereotypical patterns of rubbing: with the palms or heels of the hands, with the pulp of the fingers and with the knuckles. We trained our AI algorithm to differentiate between these movements and all other hand movements of everyday life.
Through this application, the smartwatch will be enabled to send a specific signal, such as a vibration or a small sound, every time patients rub their eyes. The application will also make it possible to store and analyze data, giving patients feedback on, for example, how often they rubbed their eyes in a week, how often they stopped after an alert was sent and how they improved from week to week. These data can then be correlated with ophthalmologic examinations. This will allow treating ophthalmologists to evaluate eye rubbing objectively, helping them educate patients and supporting them in their therapeutic decision-making. It will also provide new, objective data to confirm or disprove the causative relationship between eye rubbing and keratoconus.
In our study, we obtained good results in detecting eye rubbing, with an accuracy of nearly 94%. The next step will be to further perfect the algorithm in order to tailor it to each patient individually and apply it in clinical research.
For more information:
Inès Drira, MD, in private practice in Paris, can be reached at contact@docteurdrira.fr.
Habit reversal
Habit reversal was developed in the early 1970s by Nathan H. Azrin, and it has been successfully applied to a variety of habitual hand-to-head behaviors, such as nail biting, thumb sucking, hair pulling and eye rubbing.
The essential components are awareness training and competing response training. In awareness training, with the aid of a therapist, you learn to become aware every time your hand starts to approach the target area. Working on awareness is essential because these habits usually occur subconsciously, and you start to engage in the behavior before you know it. To increase awareness, you go through exercises in which you identify not only the actual behavior but the precursors. You cannot rub your eye unless your hands are up near your face, and your hand cannot be near your face unless it leaves wherever it was — your computer keyboard, a table, your lap. You practice identifying the early parts of that sequence of movements. With the therapist, you do it step by step. At first, you learn to recognize your hand touching your eye, then your hand near your face, your hand above your chin and your hand above the table where it was resting. Over many trials, you learn to become aware of these precursors, so that whenever you start to engage in the behavior, you recognize it immediately before it ends up in the actual behavior that is a problem.
Raymond G. Miltenberger
The second step is competing response training. As soon as you become aware that you are starting the behavior, you are taught to immediately do something incompatible. That can be something as simple as making a fist and holding it on your lap. It can be grasping an item that is incompatible with rubbing your eye. It can be grasping your hands in your lap or interlocking your fingers. It can be, if you are standing in line, thrusting your hands in your pockets. You practice this sequence with your therapist many times, and then you exercise it at home.
Another component that is often used within habit reversal is social support. A parent, a partner or a friend is trained to detect the behavior when it starts and immediately prompt you to initiate the competing response. When you do it, they praise you to reinforce the good behavior. Whenever possible, that social support should be implemented. When children are trained, their parents are involved in the session with the therapist from the beginning, and they are taught how to help prompt and praise the competing response. Parents are also taught to stop nagging and scolding their children when they see them engaging in repetitive behavior. Their role is to be supportive. In this way, the competing exercise becomes a pleasing experience because mom is happy with you, and when you start to engage in the old habit, the unpleasantness is no longer there — it is simply a reminder to do your exercise.
Our studies show that the rate of success of habit reversal is good. There are failures, of course, and in those cases, additional intervention components might be required.
For more information:
Raymond G. Miltenberger, PhD, BCBA-D, of the department of child and family studies, University of South Florida, can be reached at miltenbe@usf.edu.
Sources/Disclosures
Source:
Healio Interviews
References:
Disclosures:
Drira reports no relevant financial disclosures. Miltenberger reports no relevant financial disclosures.
Ask a clinical question and tap into Healio AI’s knowledge base.
- PubMed, enrolling/recruiting trials, guidelines
- Clinical Guidance, Healio CME, FDA news
- Healio’s exclusive daily news coverage of clinical data
<
Leave a Reply