TRIPOD Statement

NEW PAPER in @bmj.com "Dealing with continuous variables and modelling non-linear associations in healthcare data: practical guide"

--> www.bmj.com/content/390/...

#methodologymatters #StatsSky #EpiSky

NEW PREPRINT "Critical Appraisal of Fairness Metrics in Clinical Predictive AI"

-> arxiv.org/abs/2506.17035

We identified 62 fairness metrics (and growing) - unsurprisingly it's all a bit of a mess...with most metrics not fit for purpose

#predictiveAI #fairness #machinelearning #StatsSky #MLSky

Maternal early warning scores shown to be methodologically weak and at high risk of bias - Journal of Clinical Epidemiology www.jclinepi.com/article/S089...

Item 10 of the TRIPOD+AI asks
(www.bmj.com/content/385/...)

"Explain how the study size was arrived at, and justify that the study size was sufficient to answer the research question. Include details of any sample size calculation"

Here's why it's important 👇

Importance of sample size on the quality and utility of AI-based prediction models for healthcare Rigorous study design and analytical standards are required to generate reliable findings in healthcare from artificial intelligence (AI) research. On…

**New Lancet DH paper**

"Importance of sample size on the quality & utility of AI-based prediction models for healthcare"

- for broad audience
- explains why inadequate SS harms #AI model training, evaluation & performance
- pushback to claims SS irrelevant to AI research

👇
tinyurl.com/yrje52fn

Let’s raise the standards!

Adopting TRIPOD+AI means advancing equitable, accountable AI ready for clinics.

Check guidelines: tripod-statement.org 📖

Together, we can ensure AI serves patients first. #OpenScience #AIforGood

Who uses TRIPOD-AI?
👩🔬 Researchers designing models
👨⚕️ Clinicians evaluating tools
👩💻 Developers building algorithms
📋 Journals & peer reviewers
A shared framework for responsible innovation. 🤝 #DigitalHealth #HealthcareAI

Why TRIPOD-AI?
🔹 Ensures models can be replicated/validated
🔹 Builds trust with clinicians & patients
🔹 Reduces bias risks in outcomes
🔹 Bridges code to real-world care
Better reporting = better science for healthcare challenges. 🌍💡
#AIethics #MedEd

TRIPOD-AI’s pillars:
✅ Transparent data sources & preprocessing
✅ Full model architecture/training details
✅ Rigorous validation (internal/external)
✅ Ethics checks & bias mitigation
✅ Clear clinical impact
No more “black box” AI!
🔍 #EthicalAI #MachineLearning

TRIPOD+AI statement: updated guidance for reporting clinical prediction models that use regression or machine learning methods The TRIPOD (Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis) statement was published in 2015 to provide the minimum reporting recommendations for studie...

A guideline to boost transparency in AI-driven medical prediction models! Evolved from TRIPOD, it ensures studies are reproducible, ethical, and clinically meaningful. Crucial for trustworthy #AI in healthcare.

Read the paper: bmj.com/content/385/...

#HealthTech

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

UK EQUATOR Centre Publication School Fit for purpose: The secrets of success in writing, publishing, and disseminating research articles

Early-career researchers and students in health research! Join the UK EQUATOR Centre's Publication School to learn how to plan, write, and publish excellent journal articles.

Date: Mondays, 23 June – 14 July 2025
Time: 9:30 – 13:30 GMT+1
Where: Online via Zoom

bit.ly/equatorpubsc...

UK EQUATOR Centre Publication School Fit for purpose: The secrets of success in writing, publishing, and disseminating research articles

Health researchers, have you ever wondered how to:

• Negotiate authorship
• Choose a journal
• Develop good writing habits
• Write each section of an article
• Revise your own writing
• Respond to peer review

The UK EQUATOR Centre's Publication School is here to help! bit.ly/equatorpubsc...

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

PROBAST+AI: an updated quality, risk of bias, and applicability assessment tool for prediction models using regression or artificial intelligence methods The Prediction model Risk Of Bias ASsessment Tool (PROBAST) is used to assess the quality, risk of bias, and applicability of prediction models or algorithms and of prediction model/algorithm studies....

*NEW PAPER*

PROBAST+AI: an updated quality, risk of bias & applicability assessment tool for prediction models using regression or AI methods

PROBAST+AI consists of two distinct parts:
- model development (quality assessment tool)
- model evaluation (risk of bias tool)

www.bmj.com/content/388/...

NEW PAPER in the @bmj.com "PROBAST+AI: an updated quality, risk of bias, and applicability assessment tool for prediction models using regression or #artificialintelligence methods"

www.bmj.com/content/388/...

#StatsSky #MLSky #AI #MethodologyMatters

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

[ #Webinar ] D-8 until our next webinar! ⤵️

💬The importance of transparency in predictive AI: the role of reporting guidelines
🗣️ @gscollins.bsky.social (@ox.ac.uk)

💻https://sesstim.univ-amu.fr/fr/content/webinar-quantim-gary-collins

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/...

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/...

FUTURE-AI: international consensus guideline for trustworthy and deployable artificial intelligence in healthcare Despite major advances in artificial intelligence (AI) research for healthcare, the deployment and adoption of AI technologies remain limited in clinical practice. This paper describes the FUTURE-AI f...

FUTURE-AI: international consensus guideline for trustworthy and deployable artificial intelligence in healthcare www.bmj.com/content/388/... #artificialintelligence #healthcare #bioinformatics #ethics #trust

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

NEW PAPER in @bmj.com "FUTURE-AI: international consensus guideline for trustworthy and deployable #artificialintelligence in healthcare" from design, development, validation to regulation, deployment & monitoring

www.bmj.com/content/388/...

#StatsSky #MLSky #MedSky

👇 Item 10 of TRIPOD+AI

"Explain how the study size was arrived at (separately for development and evaluation), and justify that the study size was sufficient to answer the research question. Include details of any sample size calculation"

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

#MLSky #StatsSky

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

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

The TRIPOD-LLM reporting guideline for studies using large language models https://www.nature.com/articles/s41591-024-03425-5 (read free: https://rdcu.be/d5Nh3) 🧬🖥️🧪 web app: https://tripod-llm.vercel.app/

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

ICYMI: Last week, led by @daniellebitterman.bsky.social, we published the TRIPOD-LLM recommendation for reporting biomedical studies using large language models.

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

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