타각적 굴절검사값을 기반으로 최적의 안경처방 머신러닝 알고리즘 개발
Purpose : As the Fourth Industrial Revolution progresses, to develop machine learning to draw subjective prescription values by using objective refraction, ocular aberrations, and pupil size. Methods : Myopic subjects (1000 eyes) with no ocular or systemic diseases that could affect vision and no history of ocular surgery were participated. I-Profilerplus (Zeiss, Berlin, Germany) was used to measure objective refraction, ocular wavefront-aberration, and pupil size. For subjective-refraction, spherical refraction (S, diopters), astigmatic refraction (C, diopters), and astigmatic axis (Ax, °) were measured using a Visuphor500 (Zeiss, Berlin, Germany). After the measurements, the machine learning model was developed using Python (version 3.10) and checked its prediction performance. Results : In the subjective refraction, the factors affecting the spherical refractive power were the highest in the order of objective spherical refractive errors, defocus aberration, spherical aberration, and trefoil aberration had the highest impact on spherical refractive power, while objective cylindrical refractive errors, defocus aberration, coma aberration, and trefoil aberration had the highest impact on cylindrical refractive power. However, the astigmatic axis was affected only by objective astigmatic axis. There was no difference between subjective refractive errors and machine learning predicted refractive errors for spherical refraction, cylindrical refraction, and astigmatic axis(p=0.976, 0.948, and 0.349, respectively). Conclusion : A machine learning model that predicts the subjective refractive errors was developed, and the prediction accuracy was confirmed through there was no significant difference between the predicted refractive errors and the subjective refractive errors. Therefore, it is thought that it can be used as basic data to derive accurate eyeglass prescription for personalized prescriptions in the future.
목적 : 4차 산업혁명이 진행됨에 따라 타각적 굴절검사값, 수차 및 동공크기 등을 이용하여 최적의 안경처방값 을 도출해주는 머신러닝(machine learning)을 개발하고자 하였다. 방법: 시력에 영향을 줄 수 있는 안질환 및 전신질환이 없고 안구 수술 이력이 없는 근시안(1,000안)을 대상으로 진행하였다. I-Profilerplus(Zeiss, Berlin, Germany)를 사용하여 타각적 굴절이상도(objective-refraction) 및 안구수차(ocular wavefront-aberration), 동공 크기를 측정하였고, 자각적 굴절이상도(subjective-refraction)는 Visuphor500(Zeiss, Berlin, Germany)를 사용하여 구면 굴절력(S, Diopter), 원주 굴절력(C, Diopter), 난시 축(Ax, °)을 측정하였다. 측정 후, 파이썬(Python, version 3.10)을 이용하여 머신러닝 모델 생성 및 예측 성능을 확인하였다. 결과: 자각적 굴절이상도에서 구면 굴절력에 영향을 미치는 요인은 타각적 구면 굴절력, defocus aberration, spherical aberration, trefoil aberration 순으로 높았고, 원주 굴절력에 영향을 미치는 요인은 타각적 원주 굴 절력, defocus aberration, coma aberration, trefoil aberration 순으로 높았으며, 난시 축은 타각적 난시축만 영향을 미치는 것으로 나타났다. 구면 굴절력, 원주 굴절력, 난시 축의 자각적 굴절이상도와 머신러닝 예상값은 차이가 없는 것으로 나타났다(p=0.976, 0.948, and 0.349, respectively). 결론 : 자각적 굴절이상도를 예측하는 머신러닝 모델을 생성하였고, 해당 모델의 예측된 값과 자각적 굴절이상 도와 유의한 차이가 없는 것을 통해 예측 정확도를 확인하였으며 앞으로 개인 맞춤형 처방을 위한 정확한 안경처 방값을 도출하는데 기초자료가 될 수 있을 것으로 생각된다.