TRIPOD Statement
banner
tripodstatement.bsky.social
TRIPOD Statement
@tripodstatement.bsky.social
Reporting guidelines for clinical prediction (including for AI/ML)

TRIPOD-2015 (https://tinyurl.com/ywc8m624)

TRIPOD+AI (https://tinyurl.com/4s8f3ze6)

TRIPOD-LLM (https://tinyurl.com/y4ekxxre)

Visit -> www.tripod-statement.org
Complete and transparent reporting aids critical and assessing risk of bias in #predictiveAI. TRIPOD+AI and PROBAST+AI are key tools to improve #AI research in healthcare

TRIPOD+AI -> tinyurl.com/39pz3rfd
PROBAST+AI -> tinyurl.com/yt8vrvrf

#Digitalhealth #healthcareAI #machinelearning
April 25, 2025 at 11:05 AM
The new PROBAST+AI tool to assess quality & risk of bias of #predictive#AI models in healthcare is predicated on good reporting, i.e., by following the TRIPOD+AI guidance

PROBAST+AI
www.bmj.com/content/388/...

TRIPOD+AI
www.bmj.com/content/385/...

#MLSky #StatsSky #digitalhealth #machinelearning
March 25, 2025 at 8:42 AM
TRIPOD+AI (Expanded E&E): "If uncertainty intervals for individual prediction model outputs have been presented then provide details on how this was done" (www.bmj.com/content/385/...)

👇This new paper provides insight into uncertainty of risk estimate on decision making

www.bmj.com/content/388/...
February 13, 2025 at 12:14 PM
Underpinning the FUTURE-AI recommendations 👇 is transparency

Reporting guidelines like TRIPOD+AI are essential to ensure all key details are completely & transparently reported

Get them here -> www.bmj.com/content/385/...

#predictiveAI #machinelearning #trustworthyAI #healthcareAI #digitalhealth
February 7, 2025 at 11:15 AM
A periodic reminder that if you are writing up your study developing/validating a #machinelearning clinical prediction model then make sure you are reporting all the necessary information by following the TRIPOD+AI standards 😁

www.bmj.com/content/385/...

#MLsky #StatsSky #MedSky #transparency #AI
January 25, 2025 at 12:16 PM
TRIPOD+AI (www.bmj.com/content/385/...) ask authors to 'Report model performance estimates with confidence intervals'

This preprint by STRATOS provides a critical overview of the properties of 32 commonly reported performance measures

-> arxiv.org/abs/2412.10288

#MLSky #StatsSky #machinelearning
January 13, 2025 at 12:39 PM
Great to see "TRIPOD+AI and DECIDE-AI are the highest quality guidelines with respect to rigor of development and the involvement of the stakeholders" 😄 in

academic.oup.com/jamiaopen/ar...

Get TRIPOD+AI from www.bmj.com/content/385/...

#MLSky #StatsSky #MedSky
January 8, 2025 at 2:37 PM
We have a NEW PAPER in @naturemedicine.bsky.social on reporting recommendations for addressing the unique challenges of #largelanguagemodels (LLMs) in biomedical applications

www.nature.com/articles/s41...

#MLSky #StatsSky #medSky #AISky #artificialintelligence #generativeAI #transparency
January 8, 2025 at 10:24 AM
2024 saw us publish TRIPOD+AI, a major update to the TRIPOD reporting guideline for studies developing/evaluating prediction models using #machinelearning and statistical methods. Ensure your study is completely reported by following TRIPOD+AI

bmj.com/content/385/...

#AI #StatsSky #MLSky
December 24, 2024 at 12:33 PM
When we developed TRIPOD+AI for reporting clinical prediction models, we introduced an #openscience section to facilitate transparency and reproducibility asking authors to describe protocol, code, data availability

www.bmj.com/content/385/...

#MLSky #machinelearning #StatsSky #AI #reproducibility
December 13, 2024 at 9:42 AM
We have a website

www.tripod-statement.org

where more information can be found including translations - we will be uploading French and Chinese translations of TRIPOD+AI soon.

/end (for now 😉)
November 23, 2024 at 11:50 AM
Protocol development is a crucial in mapping out any research study. It provides key details on the rationale, objectives, design, data collection, analysis methods & dissemination plans before a piece of research is conducted, TRIPOD-P (in preparation) will provide guidance

tinyurl.com/53pe46wh
November 23, 2024 at 11:50 AM
Large language models and #generativeAI are making a huge impact in healthcare and beyond, so we have developed TRIPOD-LLM to provide recommendations (will be published soon, but currently available as a preprint)

tinyurl.com/bdhjv4hj

+website tripod-llm.vercel.app
November 23, 2024 at 11:50 AM
Clear and informative reporting of journal & conference abstracts is essential to help readers and reviewers identify potentially relevant studies and decide whether to read the full text, so we developed TRIPOD-Abstracts to provide reporting recommendations

tinyurl.com/52j8sad8
November 23, 2024 at 11:50 AM
An important aspect of systematic reviews is risk of bias assessment, so members of the TRIPOD Initiative also developed the PROBAST tool

tinyurl.com/4x9a8b4z (tool)

tinyurl.com/3e6kkekt (guidance)

An update for #machinelearing models (PROBAST+AI) will appear soon!
November 23, 2024 at 11:50 AM
With the large and growing number of prediction models, systematic reviews to evaluate and summarise the overall evidence available are important, so we developed TRIPOD-SRMA to provide guidance on reporting these studies

tinyurl.com/33ffsj9k

#evidencesynthesis #systematicreviews
November 23, 2024 at 11:50 AM
Accompanying the TRIPOD Cluster is a 38 page explanation and elaboration paper (containing 213 references) which provides examples of good reporting and methodological insights into developing and validating clinical prediction models with clustered data.

tinyurl.com/2kvvr4f9
November 23, 2024 at 11:50 AM
If you are developing models using data from multiple centres, locations, or different studies. Accounting for clustering can be important to capture and explore heterogeneity in prediction model performance, so we developed TRIPOD-Cluster to guide the reporting.

tinyurl.com/4bks4xmd
November 23, 2024 at 11:50 AM
An overview on how and why TRIPOD+AI was developed can by found in this short piece by lead author @gscollins.bsky.social

tinyurl.com/57dknawu
November 23, 2024 at 11:50 AM
In April 2024, we updated TRIPOD to reflect methodological advances (including sample size, #fairness, #openscience) and to make it agnostic to modelling approach, and cater for #machinelearning methods (we called this TRIPOD+AI).

tinyurl.com/2kc39b4s
November 23, 2024 at 11:50 AM
Accompanying the 2015 TRIPOD guideline is a detailed 73 page explanation and elaboration paper (containing 531 references) providing examples of good reporting and methodological insights into developing and validating clinical prediction models.

tinyurl.com/29k7a5wm
November 23, 2024 at 11:50 AM
...and to aid dissemination across different clinical areas we published the guidance simultaneously in 11 journals
November 23, 2024 at 11:50 AM
Our first guideline led by @gscollins.bsky.social, Carl Moons, Hans Reitsma & Doug Altman was published in 2015 and provides recommendations for clinical prediction models (primarily regression based models).

This made an impact and is widely used with ~9000 citations 🎉

tinyurl.com/3mbk7sv4
November 23, 2024 at 11:50 AM
We've made the move across...so here’s a little thread introducing the various TRIPOD reporting guidelines to provide standards for #transparency and completeness of clinical prediction models to facilitate #reproducibility and guide #openscience

#MLSky #StatsSky #statistics #academia #EpiSky
November 23, 2024 at 11:50 AM
Our latest recommendations for reporting #AI/#machinelearning prediction model studies in healthcare are available in the @bmj.com

bmj.com/content/385/...

Don't forget the further guidance for each reporting recommendation in the supplementary material

#MLSky #statsSky #reportingstandards
November 19, 2024 at 10:36 PM