You are here

  • Home
  • Online First
  • Differences in performance of acute ischemic stroke artificial intelligence platforms

Article Text

Article menu
Download PDFPDF
Neuroimaging
Differences in performance of acute ischemic stroke artificial intelligence platforms
  1. Mohamad Ezzeldin1,2,
  2. Adam Delora3,
  3. Ameer E Hassan4,
  4. Rime Ezzeldin5,
  5. Christopher Hadjialiakbari3,
  6. Eryn Percenti6,
  7. Jordan Torres6,
  8. Yazan J Alderazi7
    1. 1Department of Clinical Sciences, College of Medicine, University of Houston, Houston, Texas, USA
    2. 2Neuroendovascular Surgery, HCA Houston, Houston, Texas, USA
    3. 3Emergency Medicine, HCA Houston, Kingwood, Texas, USA
    4. 4Department of Neurology, University of Texas Rio Grande Valley, Harlingen, Texas, USA
    5. 5Medicine, Jordan University of Science and Technology Faculty of Medicine, Irbid, Jordan
    6. 6Internal Medicine, HCA Houston, Kingwood, Texas, USA
    7. 7Cerebrovascular Disease and Neurointerventional Surgery, HCA Gulf Coast Division, Webster, Texas, USA
    1. Correspondence to Dr Mohamad Ezzeldin; mohamadezzeldin{at}hotmail.com

    https://doi.org/10.1136/jnis-2024-022373

    Statistics from Altmetric.com

      Request Permissions

      If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

      Thank you for your interest in our manuscript.1 ,2 Real world studies are critical in understanding the limitations and performance of these software packages and how they perform in different environments. This facilitates generalizability to the users' unique settings and use case. On review of the studies you have cited, we found that many of them have findings similar to our study.1 In our retrospective performance analysis, Rapid large vessel occlusion (LVO) had a specificity of 0.85 and a sensitivity of 0.87, with a positive predictive value of 0.46 and a negative predictive value of 0.97. Viz LVO had a specificity of 0.96 and a sensitivity of 0.87, with a positive predictive value of 0.75 and a negative predictive value of 0.98. This is consistent with previous studies, including some of the studies cited in your letter. According to Soun et al, Rapid LVO had a sensitivity of 0.96, specificity of 0.85, negative predictive value of 0.99, and positive predictive value of 0.53.3 Schlossman et al showed that Rapid LVO demonstrated a sensitivity of 0.90, specificity of 0.86, positive predictive value of 0.45, and negative predictive value of 0.98.4 The studies cited in your letter show values very similar to those we have reported and are reassuring with respect to the validity of our findings.

      In your letter, you present an average of the values for sensitivity, specificity, positive predictive value, and negative predictive value. Taking three …

      View Full Text

      Footnotes

      • X @jordantorres_md

      • Contributors ME is the guarantor, ensuring the integrity of the work, overseeing the revision process, and addressing concerns raised during the review process. All authors contributed to drafting the manuscript and approved the final version.

      • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial, or not-for-profit sectors.

      • Competing interests HCA Healthcare has a minority equity stake in VIZ.ai and also purchases products from the company for use in its facilities. AEH is a consultant and a speaker for Viz.ai, who developed the Viz LVO algorithm that we tested, and he is also involved in a study sponsored by Viz.ai (LVO Synchronise Study) looking at the impact of Viz LVO implementation on patient timing and outcomes. ME is a speaker/consultant for Viz AI and Imperative Care and has investments in Galaxy Therapeutics. In the interest of full disclosure, below is a list of conflicts of related companies that are in a related field that we do not believe directly impacted this study. YJA has stock in Sanavention. AEH is a consultant/speaker for Medtronic, Microvention, Stryker, Penumbra, Cerenovus, Genentech, GE Healthcare, Scientia, Balt, Viz.ai, Insera therapeutics, Proximie, NeuroVasc, NovaSignal, Vesalio, Rapid Medical, Imperative Care, and Galaxy Therapeutics. AEH is a principal investigator of COMPLETE study (Penumbra), LVO SYNCHRONISE (Viz.ai), Millipede Stroke Trial (Perfuze), and RESCUE ICAD (Medtronic). AEH is on the steering committee/publication committee of SELECT, DAWN, SELECT 2, EXPEDITE II, EMBOLISE, CLEAR, ENVI, DELPHI, and DISTALS. AEH is DSMB of the COMAND trial.

      • Provenance and peer review Not commissioned; internally peer reviewed.

      Linked Articles