👇👇
RT @GSCollins: Whilst these reviews focussed on #machinelearning, the findings have been observed in much earlier studies using regression…
Whilst these reviews focussed on #machinelearning, the findings have been observed in much earlier studies using regression methods - https://t.co/A9dECixlxg - https://t.co/UCDHWh5YJ6 - https://t.co/rLYcmldqlw +many more. We seemingly haven’t learnt from
@tunguz However, if you want go beyond covid I agree there is a lot of low quality prediction model studies (e.g., https://t.co/FTyHaQfDNm, https://t.co/CVGZ4a6kHB, https://t.co/H31DUoCmTA, https://t.co/a90fmleV8Y, https://t.co/WC9HQUGJae and many many mor
RT @gscollins: @modeRNepi Disappointingly, the number of events often not reported in non-ML reviews too (https://t.co/XxNBA6nBmR, https://…
@modeRNepi Disappointingly, the number of events often not reported in non-ML reviews too (https://t.co/XxNBA6nBmR, https://t.co/y722hXIAjE, https://t.co/aNrXReSQmI + many more) despite being a key item in the TRIPOD reporting statement (https://t.co/5Zbnf
@amancayork @Richard_D_Riley @laure_wynants @MaartenvSmeden Existed well before covid (e.g., https://t.co/hZTKuT51la, https://t.co/mt1nznp9Wz, https://t.co/Gcptj2Gl38, https://t.co/XxNBA6nBmR, https://t.co/aNrXReSQmI) - but amplified during covid arguably
@randyboyes @PWGTennant I can point you to review after review showing how poor GLMs are used in prediction regardless of whether ML or stats (https://t.co/aNrXReSQmI, https://t.co/26dCCXV5oT, https://t.co/hZTKuT51la, https://t.co/g7P3pUU0di, + many more),
RT @GSCollins: @IAmSamFin @georgemsavva @MaxALittle @benthebray @Farzad_MD @AledadeACO Absolutely, there are plenty of systematic reviews e…
@IAmSamFin @georgemsavva @MaxALittle @benthebray @Farzad_MD @AledadeACO Absolutely, there are plenty of systematic reviews evaluating prediction models using regression highlighting major flaws in methodology and reporting (e.g., https://t.co/hZTKuT51la, h
@WalterReade @kdpsinghlab @dinga92 @MaartenvSmeden @AndrewLBeam @ivivek87 @ADAlthousePhD @IAmSamFin @signormirko Many regression based prediction models are doing on tiny sample sizes (few outcome events), categorise continuous predictors, ignore missing d
RT @GSCollins: @TenanATC Journals riding the ML wave/driving the hype. There's also poor study design, poor/weak validation, unfair compari…
RT @GSCollins: @TenanATC Journals riding the ML wave/driving the hype. There's also poor study design, poor/weak validation, unfair compari…
RT @GSCollins: @TenanATC Journals riding the ML wave/driving the hype. There's also poor study design, poor/weak validation, unfair compari…
RT @GSCollins: @TenanATC Journals riding the ML wave/driving the hype. There's also poor study design, poor/weak validation, unfair compari…
@TenanATC Journals riding the ML wave/driving the hype. There's also poor study design, poor/weak validation, unfair comparisons against non-ML (https://t.co/g7P3pUU0di). But we observe poor methods also in established non-ML prediction literature too (e.g
Developing risk prediction models for type 2 diabetes: a systematic review of methodology and reporting http://t.co/bB0fGkd9Lc #bmcmedicine
Abstract | Developing risk prediction models for type 2 diabetes: a ...: Risk prediction models for type 2 diabe... http://t.co/dWpfciep
Abstract | Developing risk prediction models for type 2 diabetes: a ...: Risk prediction models for type 2 diabe... http://t.co/0JHD9yH
Breaking News: Abstract | Developing risk prediction models for type 2 diabetes: a ... http://t.co/mNP5Bun
Abstract | Developing risk prediction models for type 2 diabetes: a ...: Risk prediction models for type 2 diabe... http://t.co/wStLQ01
T2 Diabetes risk prediction models: many studies => poor methods & generally poor level of reporting http://t.co/wyGa4Z0
Abstract | Developing risk prediction models for type 2 diabetes: a ...: Risk prediction models for type 2 diabe... http://t.co/TNC5gwX
Abstract | Developing risk prediction models for type 2 diabetes a ...: Risk prediction models for type 2 diabet... http://t.co/Hwhd62z
BioMedCentral: New research from #BMCMedicine discusses risk prediction models for diabetes http://ow.ly/6oZJo http://t.co/YByOqnz
#bmcmedicine Developing risk prediction models for type 2 diabetes: a systematic review of methodology and reporting http://t.co/JhjVfXG
New research from #BMCMedicine discusses risk prediction models for diabetes http://ow.ly/6oZJo
RT @BMC_series: Numerous risk prediction models for diabetes are affected by weak methods and poor reporting #BMCMedicine #OA http://t.co/12I8ITv
Numerous risk prediction models for diabetes are affected by weak methods and poor reporting #BMCMedicine #OA http://t.co/12I8ITv