Authors' Reply: Interpreting Agreement in AI-Assisted American Society of Anesthesiologists (ASA) Classification - Takeaways - MDSpire
Coming Soon: Introducing MDSpire News. Learn more
Conexiant’s news site is now MDSpire News. Learn more

Authors’ Response: Understanding Agreement in AI-Assisted American Society of Anesthesiologists (ASA) Classification

  • By

  • Bo-Han Lin

  • Cheng-Wei Lu

  • I-Ching Hou

  • Tzu-Yu Lin

  • Chin-E Liu

  • Chong-Sun Khoi

  • October 21, 2026

Share

  • 1

    The original analysis used unweighted Cohen κ to assess exact agreement between AI-generated and clinician-assigned ASA classifications.

  • 2

    Linearly weighted Cohen κ was 0.893 with Taiwan NHI data and 0.560 without NHI data.

  • 3

    Among 63 patients, all 9 disagreements with NHI data involved adjacent ASA categories; without NHI data, 16 of 17 disagreements were adjacent.

  • 4

    In 10,000 paired patient-level bootstrap replicates, the unweighted κ difference with versus without NHI data was 0.365 (95% CI, 0.292-0.440).

  • 5

    Interpretation by ASA category is limited by the predominance of ASA II patients and small ASA III-to-IV subgroups, and κ does not establish comparable performance across categories.

Original Source(s)

Authors' Reply: Interpreting Agreement in AI-Assisted American Society of Anesthesiologists (ASA) Classification

  • by Bo-Han Lin, Cheng-Wei Lu, I-Ching Hou, Tzu-Yu Lin, Chin-E Liu, Chong-Sun Khoi

  • October 21, 2026

Related Content

Nipple or skin-sparing mastectomy flap: can preoperative breast-MRI predict post-operative complications?

Correction to: Technical refinements of pediatric laparoscopic percutaneous extraperitoneal closure for adult indirect inguinal hernia

Correction: VFA/TAMAI predicts aggravated organ damage after central line removal due to catheter‑related bloodstream infection in patients with enterocutaneous fistula