Recently, the Tokyo Medical and Dental University researchers have created a machine-learning model using
that effectively predicts and visualizes the long-term risk of visual impairment (
).
Myopia or shortsightedness is a condition that affects distance vision where there is no trouble seeing objects that are near, but have difficulty in viewing objects that are farther away.
People with extreme shortsightedness, also called high myopia are unable to focus on distant objects. Contact lenses , glasses, or surgery can correct their eyesight, but having high myopia is not just inconvenient.
Advertisement
About half of the cases result in pathologic myopia, and blindness is most commonly caused by complications from pathologic myopia. Individuals who are permanently blinded frequently experience physical and financial hardships due to their lack of independence.
How Machine Learning Help Predicts Future Myopic Risks?
Although machine-learning algorithms are already effective at recognizing changes and difficulties in myopia, this recent study examined the algorithms’ long-term prediction capabilities.
A nomogram comprising roughly 34 variables related to myopia was evaluated using a machine-learning model based on logistic regression. These factors were then combined to create a score that represents the long-term risk of visual impairment.
The best prediction for the risk of visual impairment was at 5 years. The
These scores predict visual acuity and help decision-making in future clinical assessments.
In summary, the study demonstrated that machine-learning models have a high potential to assist in cutting down the future risk of blindness- an increasingly significant public health challenge, which will benefit individuals as well as society at large.
Reference :
- Machine Learning Models for Predicting Long-Term Visual Acuity in Highly Myopic Eyes – (https:jamanetwork.com/journals/jamaophthalmology/article-abstract/2810735)
Source: Medindia
