One of the top three causes of permanent blindness in many parts of the world, extreme myopia, is predicted to induce visual impairment in patients based on machine learning models created by Japanese researchers.
High myopia, a condition in which a person has excessive shortsightedness, allows them to focus on close things but not far ones.
Although high myopia can be corrected with contacts, glasses, or surgery, it is not only uncomfortable; it also frequently results in pathologic myopia, which is the disease that causes the majority of blindness due to complications.
“We are aware that machine-learning algorithms perform effectively in tasks like detecting alterations and issues in myopia,” Tokyo Medical and Dental University (TMDU) in Japan’s main author Yining Wang stated.
However, the goal of this study was to look into a different topic: how effective these algorithms are at making long-term forecasts.
In order to do this, the group examined the visual acuity of 967 Japanese patients after three and five years had gone by using a cohort study.
They created a dataset using 34 characteristics, including age, corneal diameter, and current visual acuity, which are frequently obtained during ocular examinations.
Next, they put a number of well-known machine learning models to the test, including support vector machines and random forests. When it came to forecasting visual impairment at five years old, the logistic regression-based model outperformed the others.
Predicting results, though, is just one aspect of the tale.
Senior author Kyoko Ohno-Matsui of the university stated, “It’s also important to present the model’s output in a way that is convenient for making clinical decisions and easy for patients to understand.”
The researchers visualized the classification model using a nomogram in order to accomplish this. A line with a length indicating the significance of each variable for visual acuity prediction is allocated to it.
These lengths can be translated into points, which can then be tallied to produce a final score that illustrates the likelihood of future visual impairment.
Individuals who lose their vision permanently frequently experience physical and financial hardships as a result of their diminished freedom.
This study has demonstrated that machine-learning models have high potential to assist address this increasingly significant public health challenge, which will benefit individuals as well as society at large, even though the model still has to be assessed on a larger population.
