Thyroid disease

Integration of Artificial Intelligence Decision Aids to Reduce Workload and Enhance Efficiency in Thyroid Nodule Management

JAMA Netw OpenPublished Diagnostic study

Authors' conclusion

This diagnostic study suggests that an optimized AI strategy in thyroid nodule management may reduce diagnostic time-based costs without sacrificing diagnostic accuracy for senior radiologists, while the traditional all-AI strategy may still be more beneficial for junior radiologists.

Study context

Population
Patients with thyroid nodules on US; retrospective set 1048 patients, 1754 nodules (July 2018–July 2019); prospective set 268 patients, 300 nodules (May–December 2021); 16 junior and senior radiologists16 radiologists
Comparison
Optimized artificial intelligence (AI) strategy vs traditional all-AI strategy
Outcome
Diagnostic performance, time-based cost and assisted diagnosis, optimized vs traditional all-AI strategy (prospective set)
Methodological detailsretrospective development set, prospective comparison set, Data analyses completed in September 2022
  • Retrospective development set
  • Prospective comparison set
  • Data analyses completed in September 2022

Performance

As reported in the abstract

Mean task completion time

Junior reader 11: all-AI vs optimized (increased with optimized)

15.2 seconds (95% CI, 13.2-17.2) to 19.4 seconds (95% CI, 15.6-23.3)

Junior reader 12: all-AI vs optimized (increased with optimized)

12.7 seconds (95% CI, 11.4-13.9) to 15.6 seconds (95% CI, 13.6-17.7)

Senior reader 14: all-AI vs optimized (shorter with optimized)

19.4 seconds (95% CI, 18.1-20.7) to 16.8 seconds (95% CI, 15.3-18.3)

Senior reader 16: all-AI vs optimized (shorter with optimized)

12.5 seconds (95% CI, 12.1-12.9) to 10.0 seconds (95% CI, 9.5-10.5)

Further findings

Sensitivity and specificity, readers 11 to 16

No significant difference in sensitivity between the 2 strategies
range, 91%-100%
No significant difference in specificity between the 2 strategies
range, 94%-98%

US features not improved by AI assistance

Junior radiologists: cystic or almost completely cystic, anechoic, spongiform nodules, nodules <5 mm
Senior radiologists: cystic or almost completely cystic, anechoic, spongiform, very hypoechoic, taller-than-wide, lobulated or irregular nodules, extrathyroidal extension

Original publication

Integration of Artificial Intelligence Decision Aids to Reduce Workload and Enhance Efficiency in Thyroid Nodule Management. JAMA Netw Open. Published 1 May 2023. doi:10.1001/jamanetworkopen.2023.13674

Composed by AI from the published abstract and checked against it. How Briefs are made →

Get them every week

New studies on this topic in My Week and in your Monday e-mail.

We e-mail you a sign-in link and a code. No password, no spam.