Machine learning studies in medicine often fail to report how they handle missing data

Machine learning studies in medicine often fail to report how they handle missing data
Image: Intellect (generated)

In short: A systematic review of 152 clinical prediction model studies found that researchers frequently fail to report how they manage missing data. While many studies use machine learning, a significant portion of the literature lacks sufficient detail regarding the presence and handling of missing information, which is a common issue in medical research.

Researchers systematically reviewed published papers from 2018 to 2019 that used supervised machine learning to develop or validate clinical prediction models. They found that many studies simply omit patient data through complete-case analysis, a method generally advised against because it can lead to bias and reduced analytical power.

The review was limited to studies published within a specific two-year window focusing on supervised machine learning in medical fields.

Why it matters: Incomplete reporting of missing data in machine learning models can obscure potential bias and loss of analytical power.

Opens the Intellect+ feed with this card first. Swipe up for the next one.

Sources

To cite this card, link to https://intellectplus.co.uk/app/science-research-machine-learning-studies-in-medicine-often-fail-to-report-ho.

How we source, check and correct cards: Editorial standards.

General information, not professional advice. Spotted an error? Flag the card or see corrections and complaints.

Intellect Plus is a visual educational discovery platform for exploring fascinating facts and ideas. Brain Food, Not Brain Rot. What is Intellect Plus?

Related

Finding something worth knowing…