Kazutaka Kamiya1, Ik Hee Ryu2, Tae Keun Yoo3, Jung Sub Kim2, In Sik Lee2, Jin Kook Kim2, Wakako Ando4, Nobuyuki Shoji4, Tomofusa Yamauchi5, Hitoshi Tabuchi5. 1. From Visual Physiology, School of Allied Health Sciences. Electronic address: kamiyak-tky@umin.ac.jp. 2. Kitasato University, Kanagawa, Japan; B&VIIT Eye Center, Seoul, South Korea. 3. Kitasato University, Kanagawa, Japan; B&VIIT Eye Center, Seoul, South Korea; Department of Ophthalmology, Aerospace Medical Center, Republic of Korea Air Force, Cheongju, South Korea. 4. the Department of Ophthalmology, School of Medicine. 5. Department of Ophthalmology, Tsukazaki Hospital, Hyogo, Japan.
Abstract
PURPOSE: To compare the achieved vault using the conventional manufacturer's nomogram and the predicted vault using machine learning, in a large cohort of eyes undergoing posterior chamber phakic intraocular lens (EVO implantable collamer lens [ICL]; STAAR Surgical) implantation. DESIGN: Reliability and validity assessment and comparison. METHODS: Our study comprised a total of 1,745 eyes of 1,745 consecutive patients (mean age ± standard deviation, 26.2 ± 6.8 years) undergoing ICL implantation. At 1 month postoperatively, we quantitatively measured the actual vault, and compared it with the predicted vault using machine leaning of anterior segment optical coherence tomography metrics. RESULTS: All machine learning methods, such as support vector regressor (SVR), gradient boost regressor (GBR), random forest regressor (RFR), and linear regressor (LR), showed significantly less mean absolute error and higher percentages of eyes within 50-200 µm of the targeted ICL vault than the conventional nomogram (P < .001). The RFR provided the lowest mean absolute errors and the highest percentages of eyes within 50 to 200 µm, followed by the GBR, the LR, and the SVR (P < .01). CONCLUSIONS: Machine learning of the preoperative anterior segment optical coherence tomography metrics, especially the RFR, provided significantly higher predictability of the ICL vault than the conventional manufacturer's nomogram, suggesting that it will become an aid for predicting the ICL vault and subsequently selecting the proper ICL size in daily practice.
PURPOSE: To compare the achieved vault using the conventional manufacturer's nomogram and the predicted vault using machine learning, in a large cohort of eyes undergoing posterior chamber phakic intraocular lens (EVO implantable collamer lens [ICL]; STAAR Surgical) implantation. DESIGN: Reliability and validity assessment and comparison. METHODS: Our study comprised a total of 1,745 eyes of 1,745 consecutive patients (mean age ± standard deviation, 26.2 ± 6.8 years) undergoing ICL implantation. At 1 month postoperatively, we quantitatively measured the actual vault, and compared it with the predicted vault using machine leaning of anterior segment optical coherence tomography metrics. RESULTS: All machine learning methods, such as support vector regressor (SVR), gradient boost regressor (GBR), random forest regressor (RFR), and linear regressor (LR), showed significantly less mean absolute error and higher percentages of eyes within 50-200 µm of the targeted ICL vault than the conventional nomogram (P < .001). The RFR provided the lowest mean absolute errors and the highest percentages of eyes within 50 to 200 µm, followed by the GBR, the LR, and the SVR (P < .01). CONCLUSIONS: Machine learning of the preoperative anterior segment optical coherence tomography metrics, especially the RFR, provided significantly higher predictability of the ICL vault than the conventional manufacturer's nomogram, suggesting that it will become an aid for predicting the ICL vault and subsequently selecting the proper ICL size in daily practice.
Authors: Yong Li; Feihui Zheng; Li Lian Foo; Qiu Ying Wong; Daniel Ting; Quan V Hoang; Rachel Chong; Marcus Ang; Chee Wai Wong Journal: Diagnostics (Basel) Date: 2022-06-08
Authors: Eun Min Kang; Ik Hee Ryu; Geunyoung Lee; Jin Kuk Kim; In Sik Lee; Ga Hee Jeon; Hojin Song; Kazutaka Kamiya; Tae Keun Yoo Journal: Transl Vis Sci Technol Date: 2021-05-03 Impact factor: 3.283