The optimization of the vault-predicting formula based on the anterior segment measurements from artemis insight 100
Hao Wang, Bo Zhang, Wenwen Du, Zaohe Sun, Qi Fan, Chenjiu Pang
- 发表年份
- 2024
- 引用次数
- 2
- 访问权限
- 开放获取
摘要
Abstract To optimize and evaluate the accuracy of the vault-predicting formula generated from a very high-frequency digital ultrasound robotic scanner (Artemis Insight 100). The relationship between the achieved lens vault (LVa) at one month after intraocular collamer lens (ICL) implantation surgery and the predicted vault (LVp) was analyzed by a retrospective study, and an optimized formula was built up. Then, the accuracy of the optimized vault-predicting formula was evaluated in a prospective study by comparing the LVa and the predicted vault from the optimized formula (LVop). The retrospective study included 77 patients (133 eyes) while the prospective study enrolled 90 patients (170 eyes). The difference between LVp and LVa at one month after surgery was statistically significant ( P < 0.05), and the linear regression analysis of LVa against LVp yielded a good fit ( R 2 = 0.68). The optimized vault-predicting formula was LVop (μm) = 1.21 × LVp (μm) + 124.73. In the validation study, the difference between LVop and LVa was not statistically significant ( P = 0.10), and a good agreement between LVop and LVa was shown by Bland–Altman analysis. The optimized vault-predicting formula could predict the actual LV after ICL implantation surgery, help to select an appropriate ICL size and reduce the need for re-operation.
关键词
相关论文
Statistical Learning Theory
Yuhai Wu, Vladimir Vapnik
1999
Artificial intelligence: a modern approach
1995
Applied Nonlinear Control
Jean-Jacques Slotine, Weiping Li
1991
A new optimizer using particle swarm theory
R.C. Eberhart, James Kennedy
2002