AI cuts lesion measurement time by 34% - Takeaways - MDSpire
Coming Soon: Introducing MDSpire News. Learn more
Conexiant’s news site is now MDSpire News. Learn more

AI cuts lesion measurement time by 34%

Automated lesion measurements increased agreement on treatment response, although expert-generated proposals were accepted more often and required fewer adjustments.

  • By

  • Andrea Surnit

  • September 3, 2026

  • 4 min

Share

  • 1

    AI-assisted lesion measurements reduced reading time by 34%, from 105 seconds to 71 seconds per patient.

  • 2

    AI assistance increased interobserver agreement on RECIST response classification by approximately 8 percentage points.

  • 3

    AI-assisted measurements showed greater deviation from expert-derived references, with a mean absolute error of 4.35 mm.

  • 4

    Radiologists accepted 61% of AI-generated proposals, compared to 77% of expert proposals, indicating less trust in AI.

  • 5

    Expert assistance provided the lowest measurement error, with a mean absolute error of 1.51 mm, compared to AI and unassisted measurements.

Original Source(s)

AI cuts lesion measurement time by 34%

  • by Andrea Surnit

  • September 3, 2026

Related Content

AI diagnostics and what clinicians can verify in oncology

"The more AI changes what we do … the less able we are to check it.”

Signal, not noise: learning to measure emphysema beyond the −950 HU threshold

Hot tub lung may mimic refractory pneumonitis

Rare MAC organisms on lung biopsy and an overlooked exposure history shifted the diagnosis after prolonged immunosuppression.