We are honoured to be recognised as a 'Notable Advance of 2020' @NatureMedicine, ensuring transparency in medical #AI.
Thank you to all who supported, and continue to support, SPIRIT-AI & CONSORT-AI
@CONSORTing@EQUATORNetwork
Calling all patients, carers, healthcare professionals, and researchers working on or using #clinicaltrials: Help shape the next update of the CONSORT and SPIRIT reporting guidelines!
Sign up here to have your say on what should be included: https://t.co/uMDatbde6e
4/6 We have a special interest in trials of new technologies, #AI in healthcare and the #microbiome. If you are submitting a trial in one of these areas, make sure that you submit the relevant checklist with your paper. @ConsortSpiritAI https://t.co/k4xSw8JsmD
"#AI applied to breast cancer treatment outcomes documented limited external validation & model calibration, with substantial risk of bias, significant gaps in reporting, and poor code & data availability"
- https://t.co/RdhYKhsZzO
Use TRIPOD https://t.co/dNuw3dWwmi
#mltwitter
Just published today @NatureMedicine! Reporting of early-stage evaluation of #AI: specifically focussing on feasibility, human factors and safety. Great to be part of this initiative spearheaded by ⭐️Baptiste Vasey⭐️Peter McCulloch & the DECIDE-AI team!
https://t.co/MpsEmtoPTd
SUPER excited this paper is now out. The facade of a division between ethics and science has contributed to the problems we now face when trying to make AI equitable. To better support #ethicalAI and #healthequity using AI in healthcare we must... 1/5
Just launched the 2nd (and final) round of the TRIPOD-AI Delphi survey seeking consensus on how to report studies developing/validating #MachineLearning based clinical prediction models.
DM me if you want to participate.
#statstwitter#mltwitter#ml4h#medtwitter#epitwitter
Fantastic to see reporting guidelines included in this @WHO evidence generation framework for AI-based medical devices! Includes @CONSORTing, SPIRIT, @ConsortSpiritAI, DECIDE-AI, STARD-AI, and @TRIPODStatement-ML.
.@WHO’s Evidence Generation Framework for #AI- based Medical Devices, Including a use-case for Cervical Cancer screening is a result of extraordinary collaboration and I’m delighted that Women’s Health is showcased as an area of immense potential for implementing innovation
New quality assessment tool QUADAS-AI for #AI diagnostic accuracy studies, compliments upcoming reporting standards @STARD_AI.
Led by the brilliant @viknesh_s plus an incredible team:
Announcing a new initiative in @NatureMedicine to develop a quality assessment tool for #ArtificialIntelligence-centered diagnostic test accuracy studies: QUADAS-AI
=> https://t.co/4Ksuydk6X2
Newly published protocols in @BMJ_Open describing our development process behind upcoming reporting guidelines & risk of bias tools for #MachineLearning based clinical prediction model studies (https://t.co/GbUyyS6ROh) & diagnostic test accuracy studies (https://t.co/mRG6OndA3x)
Our #openaccess protocol is now available in @BMJ_Open describing the methods to develop TRIPOD-AI reporting guideline and PROBAST-AI risk of bias tool for #MachineLearning prediction models. (https://t.co/wsmHgILkGw)
Are you developing or validating a prediction model using artificial intelligence? The new @TRIPODStatement extension is on its way!
TRIPOD-AI is registered as a reporting guideline under development with us: https://t.co/jOT2PFBbC0
Our new study looking at completeness of reporting of #MachineLearning based clinical prediction models against the TRIPOD Statement (https://t.co/zG0ZFM37mv).
Findings will feed into our upcoming guidance for reporting #MachineLearning prediction models)
@GSCollins@CarlMoons