Juan Romero1, Sharon Chiang2, Daniel M Goldenholz3. 1. Harvard Beth Israel Deaconess Medical Center, Boston MA, United States. Electronic address: jromero5@bidmc.harvard.edu. 2. University of California San Francisco, San Francisco, CA, United States. Electronic address: Sharon.Chiang@ucsf.edu. 3. Harvard Beth Israel Deaconess Medical Center, Boston MA, United States. Electronic address: daniel.goldenholz@bidmc.harvard.edu.
Abstract
PURPOSE: Recently a realistic simulator of patient seizure diaries was developed that can reproduce effects seen in randomized clinical trials (RCTs). RCTs suffer from high costs and statistical inefficiencies. Using realistic simulation and machine learning this study aimed to identify a more statistically efficient outcome metric. METHODS: Five candidate deep learning architectures with 54 permutations of hyperparameters were compared to the traditional standard, median percent change (MPC). Each were also tested for type 1 error. All models had similar outcomes, with appropriate low levels of type 1 error. RESULTS: The simplest model was equivalent to a logistic regression of a histogram of individual percentage changes in seizure rate, requiring 21-22% less patients to discriminate drug from placebo at 90% power. This model was referred to as LPC. CONCLUSION: Future studies to validate LPC may enable faster, cheaper and more efficient clinical trials.
PURPOSE: Recently a realistic simulator of patient seizure diaries was developed that can reproduce effects seen in randomized clinical trials (RCTs). RCTs suffer from high costs and statistical inefficiencies. Using realistic simulation and machine learning this study aimed to identify a more statistically efficient outcome metric. METHODS: Five candidate deep learning architectures with 54 permutations of hyperparameters were compared to the traditional standard, median percent change (MPC). Each were also tested for type 1 error. All models had similar outcomes, with appropriate low levels of type 1 error. RESULTS: The simplest model was equivalent to a logistic regression of a histogram of individual percentage changes in seizure rate, requiring 21-22% less patients to discriminate drug from placebo at 90% power. This model was referred to as LPC. CONCLUSION: Future studies to validate LPC may enable faster, cheaper and more efficient clinical trials.
Authors: Philippa J Karoly; Daniel M Goldenholz; Dean R Freestone; Robert E Moss; David B Grayden; William H Theodore; Mark J Cook Journal: Lancet Neurol Date: 2018-09-12 Impact factor: 44.182
Authors: Daniel M Goldenholz; Shira R Goldenholz; Robert Moss; Jacqueline French; Daniel Lowenstein; Ruben Kuzniecky; Sheryl Haut; Sabrina Cristofaro; Kamil Detyniecki; John Hixson; Philippa Karoly; Mark Cook; Alex Strashny; William H Theodore; Carl Pieper Journal: Epilepsy Res Date: 2017-07-25 Impact factor: 3.045