Previously submitted to: Journal of Medical Internet Research (no longer under consideration since Nov 08, 2021)
Date Submitted: Jun 9, 2020
Can artificial Intelligence Support Clinical Decision Making in the Management of Hepatocellular Carcinoma Patients?
Background:
IBM Watson for Oncology (WFO), a cognitive artificial intelligence clinical decision support system for therapy selection, has been used in some cancers with potential benefits.
Objective:
We aim to validate the use of WFO in hepatocellular carcinoma (HCC).
Methods:
Treatment recommendations were provided for 530 HCC patients between Jan. 2013 and Jun. 2013 at three tertiary centers. WFO and a blinded second review by a multidisciplinary team (MDT) provided treatment recommendations for all cases in 2018. The degree of overall treatment concordance and specific treatment concordance between WFO and MDT was evaluated in different Barcelona Clinic Liver Cancer (BCLC) stages. Univariable and multivariable regression analyses were performed to identify factors associated with discordance.
Results:
The overall concordance rate was 52.1% between the second MDT review and WFO, with 60.0%, 54.9%, 40.3%, and 50.8% for BCLC stages 0, A, B, and C, respectively. For BCLC stage 0, hepatectomy was the recommended treatment by both MDT and WFO without a significant difference (P = 0.10). For BCLC stage A, hepatectomy was recommended most by both MDT and WFO, despite a significant difference in recommendation rate (P < 0.001). For BCLC stage B/C, Transcatheter arterial chemoembolization (TACE) was recommended more by WFO and hepatectomy more by MDT (all P < 0.001). The discordance between WFO and MDT was more often observed in female patients (P = 0.018), and patients with PLT ≤ 100 × 109/L (P < 0.001), portal hypertension (P < 0.001), or biliary obstruction (P = 0.014).
Conclusions:
WFO could provide moderate concordant treatment recommendations like those recommended by MDT for patients at BCLC stage 0, but not for those at BCLC stages A to C, to whom MDT recommended more aggressive treatments than WFO.
Clinicaltrial:
Citation
Request queued. Please wait while the file is being generated. It may take some time.
Copyright
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.