This demonstrates the feasibility of a programme I’ve set forth, for the ‘exact’ simulation of ANY dose-escalation design.
— David C. Norris, MD moved to Mastodon 🦣 (@davidcnorrismd) May 5, 2021
Influenced by @MarkusTriska‘s exacting perspective on ’exactness’, however, I now use the more precise term ‘MCSE-free’. 2/https://t.co/gsxmgQIbc8
Braun’s work draws its motivation from the chiropractic of #OneSizeFitsAllogist CRM skeleton manipulation — oh, excuse
— David C. Norris, MD moved to Mastodon 🦣 (@davidcnorrismd) May 5, 2021
me, I mean #calibration — to optimize various dose-centered performance characteristics. 4/https://t.co/pER5rjPdma
My own approach to these calibration tasks is by contrast elementary: just generate all possible paths in the trial.
— David C. Norris, MD moved to Mastodon 🦣 (@davidcnorrismd) May 5, 2021
I call this ‘complete path enumeration’ (CPE), to distinguish its intent from the aims of the related DTP idea of @ChristinaBYap et al. 6/https://t.co/dpJkahup3R
This work has also provided the occasion for me to write my first R6 class (h/t @hadleywickham https://t.co/qwwUBYce9u),
— David C. Norris, MD moved to Mastodon 🦣 (@davidcnorrismd) May 5, 2021
enabling efficient memoization of intermediate results which would be wasteful to recompute during CPE.
But there’s something I ought to tell you … 8/
All told, the net speedup is remarkable: I can now CPE (𝑣.𝑡.) a typical-sized CRM trial in mere seconds. The VIOLA trial CPE which
— David C. Norris, MD moved to Mastodon 🦣 (@davidcnorrismd) May 5, 2021
took 17 minutes in an earlier vignette now takes just 1.4 sec on 6 cores. 10/https://t.co/LjJ6MWL3Ax
The vignette concludes with a Nelder-Mead calibration to optimize the so-called ‘probability of correct selection’ (PCS) of ‘the’ MTD. This takes about 25 minutes, admittedly an order of magnitude greater than the 2 mins Braun attains for his benchmark relating to this task. 12/
— David C. Norris, MD moved to Mastodon 🦣 (@davidcnorrismd) May 5, 2021
For a fixed design, you pay the cost of CPE—mere seconds!—just ONCE; then the matrix gears of WWTT https://t.co/SJwXc7PV0I engage, yielding a comprehensive #patientcentered #trialsafety picture via R’s blazing-fast matrix math. 14/https://t.co/Yf4U7yMY4v
— David C. Norris, MD moved to Mastodon 🦣 (@davidcnorrismd) May 5, 2021
… #oncology #phase1 statisticians could engage with substantive and patient-centered questions of #pharmacology: 16/https://t.co/WDnA4sxAvf
— David C. Norris, MD moved to Mastodon 🦣 (@davidcnorrismd) May 5, 2021
I also need to get this v0.2-3 accepted onto CRAN. Including @rustlang code has created difficulties for Win-builder, in particular.
— David C. Norris, MD moved to Mastodon 🦣 (@davidcnorrismd) May 5, 2021
My hope that CRAN’s ongoing analysis will smooth out this process for the community gives me some consolation, but waiting is hard! 18/18 pic.twitter.com/BPCqLaf7Uo