Acuity VEP: improved with machine learning
Acuity VEP: improved with machine learning
复制标题
DOI：
10.1007/s10633-019-09701-x
复制
发表时间：
2019-10-01
影响因子：
1.4
通讯作者：
Heinrich, Sven P.
中科院分区：
文献类型：
--
作者：
Bach, Michael;Heinrich, Sven P.
目的视觉敏锐度-视觉诱发电位方法基本上都是使用在许多检查尺寸（或空间频率）上获得的信息来获得敏锐度的测量。通常使用幅度，有时与相位或噪声测量结合使用。在我们的方法中，我们采用稳态的低对比度棋盘格刺激，并获得六个不同的检查大小的幅度和意义，产生12个数字。基于规则的“启发式算法”（Bach等人，Br J Ophthalmol 92：396-403，2008. 10.1136/bjo.2007.130245）的成功率超过95%，109例病例的行为和客观敏锐度之间的一致性限度（LoA）为+/- 0.3LogMAR。我们在这里的目的是测试使用这个相对较小的数据集的机器学习技术是否可以实现类似的LoA。鉴于机器学习（ML）的最新进展，我们将广泛的ML算法应用于该数据集。这是在R的“插入符号”框架内使用总共89种方法完成的，其中基于规则的方法和多元回归方法表现最好。对于交叉验证，我们使用刀切（leave-one-out）方法，基于在所有剩余108个案例上训练的ML模型预测每个案例。结果ML方法在许多不同类型的ML算法中都能很好地预测视力。仅使用振幅值（丢弃p值）改善了结果。近一半的测试ML算法实现了比启发式算法更好的LoA;几个“随机森林”或“多元回归”类型的算法实现了低于+/- 0.3的LoA。在启发式方法失败的情况下，敏锐度被成功预测。然后，我们将使用Bach等人[1]数据集训练的ML模型应用于2018年的新数据集（78例），并发现启发式算法和ML方法的LoA均为+/- 0.259，几乎是一线改进。结论ML的方法似乎是一个有用的替代基于规则的分析敏锐度VEP数据。实现的准确度相当或更好（在任何情况下，基于ML的敏锐度与行为敏锐度的差异都不超过+/- 0.29 LogMAR），可测试性更高，接近100%。检查可能的陷阱。
Purpose Acuity-VEP approaches basically all use the information obtained across a number of check sizes (or spatial frequencies) to derive a measure of acuity. Amplitude is always used, sometimes combined with phase or a noise measure. In our approach, we employ steady-state brief-onset low-contrast checkerboard stimulation and obtain amplitude and significance for six different check sizes, yielding 12 numbers. The rule-based "heuristic algorithm" (Bach et al. in Br J Ophthalmol 92:396-403, 2008. 10.1136/bjo.2007.130245) is successful in over 95% with a limit of agreement (LoA) of +/- 0.3LogMAR between behavioral and objective acuity for 109 cases. We here aimed to test whether machine learning techniques with this relatively small dataset could achieve a similar LoA. Methods Given recent advances in machine learning (ML), we applied a wide class of ML algorithms to this dataset. This was done within the "caret" framework of R using altogether 89 methods, of which rule-based and multiple regression approaches performed best. For cross-validation, using a jackknife (leave-one-out) approach, we predicted each case based on an ML model having been trained on all remaining 108 cases. Results The ML approach predicted visual acuity well across many different types of ML algorithms. Using amplitude values only (discarding the p values) improved the outcome. Nearly half of the tested ML algorithms achieved an LoA better than the heuristic algorithm; several "Random Forest"- or "multiple regression"-type algorithms achieved an LoA of below +/- 0.3. In the cases where the heuristic approach failed, acuity was predicted successfully. We then applied the ML model trained with the Bach et al. [1] dataset to a new dataset from 2018 (78 cases) and found both for the heuristic algorithm and for the ML approach an LoA of +/- 0.259, a nearly one-line improvement. Conclusions The ML approach appears to be a useful alternative to rule-based analysis of acuity-VEP data. The achieved accuracy is comparable or better (in no case the ML-based acuity differed more than +/- 0.29 LogMAR from behavioral acuity), and testability is higher, nearly 100%. Possible pitfalls are examined.