Univ. of Toronto Researcher: “I did not realize quite how bad [the lack of reproducibility and poor quality in research papers] was.”

---

Many areas of science have been facing a reproducibility crisis over the past two years, and machine learning and AI are no exception. That has been highlighted by recent efforts to identify papers with results that are reproducible and those that are not.

Two new analyses put the spotlight on machine learning in health research, where lack of reproducibility and poor quality is especially alarming. “If a doctor is using machine learning or an artificial intelligence tool to aid in patient care, and that tool does not perform up to the standards reported during the research process, then that could risk harm to the patient, and it could generally lower the quality of care,” says Marzyeh Ghassemi of the University of Toronto.

In a paper describing her team’s analysis of 511 other papers, Ghassemi’s team reported that machine learning papers in healthcare were reproducible far less often than in other machine learning subfields. The group’s findings were published this week in the journal Science Translational Medicine. And in a systematic review published in Nature Machine Intelligence, 85 percent of studies using machine learning to detect COVID-19 in chest scans failed a reproducibility and quality check, and none of the models was near ready for use in clinics, the authors say.