Purpose

To evaluate if an artificial intelligence (AI) model can better select candidates that would demonstrate visual field (VF) progression in order to shorten the duration or the number of patients needed for a clinical trial

Design

Retrospective cohort study

Methods

7,428 eyes of 3,871 patients from the University of Washington Department of Ophthalmology VF Dataset were included. Progression was defined as at least 5 locations with greater than 7 dB of change compared to baseline on two consecutive tests. Progression for all patients, a subgroup of the fastest progressing based on survival curves, and patients selected based on an elastic net Cox regression model were compared. The model was trained on pointwise threshold deviation values of the first VF, age, gender, laterality and the mean total deviation (MD) at baseline.

Results

13% of all patients met the criteria for progression at five years. Differences in survival were observed when stratified by MD and age (p < 0.0001). Those at risk of progression included patients 60 to 80 years old with an initial MD < -5.0. This subgroup decreased the sample size required to detect progression compared to the entire cohort. The AI model-selected patients required the lowest number of patients for all effect sizes and trial lengths. For a trial length of 3 years and effect size of 30%, the number of patients required was 1656 [95% confidence interval (CI), 1638–1674], 903 [95% CI, 884–922], and 636 [95% CI, 625–646] for the entire cohort, the subgroup, and the model-selected patients, respectively.

Conclusion

An AI model can identify high risk patients to substantially reduce the number of patients needed or study duration required to meet clinical trial endpoints.

中文翻译：

---

神经保护试验的视野终点：AI 驱动的患者丰富案例

目的

评估人工智能 (AI) 模型是否可以更好地选择能够证明视野 (VF) 进展的候选者，以缩短临床试验所需的持续时间或患者数量

设计

回顾性队列研究

方法

来自华盛顿大学眼科 VF 数据集的 3,871 名患者的 7,428 只眼睛被纳入。进展定义为至少 5 个位置与两次连续测试的基线相比变化大于 7 dB。对所有患者的进展、基于生存曲线进展最快的亚组和基于弹性净 Cox 回归模型选择的患者进行了比较。该模型接受了第一个 VF 的逐点阈值偏差值、年龄、性别、偏侧性和基线时的平均总偏差 (MD) 的训练。

结果

13% 的患者在 5 年时符合进展标准。当按 MD 和年龄分层时，观察到存活率的差异 (p < 0.0001)。有进展风险的患者包括 60 至 80 岁且初始 MD < -5.0 的患者。与整个队列相比，该亚组减少了检测进展所需的样本量。AI 模型选择的患者在所有效应大小和试验长度下需要的患者数量最少。对于 3 年的试验长度和 30% 的效应大小，所需的患者数量为 1656 [95% 置信区间 (CI)，1638–1674]、903 [95% CI，884–922] 和 636 [95 % CI, 625–646] 分别针对整个队列、亚组和模型选择的患者。

结论

人工智能模型可以识别高风险患者，从而大幅减少满足临床试验终点所需的患者数量或研究持续时间。