Diagnostic Accuracy of AI for Opportunistic Screening of Abdominal Aortic Aneurysm in CT: A Systematic Review and Narrative Synthesis

Maria R Kodenko; Yuriy A Vasilev; Anton V Vladzymyrskyy; Olga V Omelyanskaya; Denis V Leonov; Ivan A Blokhin; Vladimir P Novik; Nicholas S Kulberg; Andrey V Samorodov; Olesya A Mokienko; Roman V Reshetnikov

doi:10.3390/diagnostics12123197

Diagnostic Accuracy of AI for Opportunistic Screening of Abdominal Aortic Aneurysm in CT: A Systematic Review and Narrative Synthesis

Diagnostics (Basel). 2022 Dec 16;12(12):3197. doi: 10.3390/diagnostics12123197.

Authors

Affiliations

¹ Research and Practical Clinical Center for Diagnostics and Telemedicine Technologies of the Moscow Health Care Department, Petrovka Street, 24, Building 1, 127051 Moscow, Russia.
² Department of Biomedical Technologies, Bauman Moscow State Technical University, 2nd Baumanskaya Street, 5, Building 1, 105005 Moscow, Russia.
³ Department of Information and Internet Technologies, I.M. Sechenov First Moscow State Medical University (Sechenov University), Trubetskaya Street, 8, Building 2, 119991 Moscow, Russia.
⁴ Department of Fundamentals of Radio Engineering, Moscow Power Engineering Institute, Krasnokazarmennaya Street, 14, Building 1, 111250 Moscow, Russia.
⁵ Federal Research Center "Computer Science and Control" of Russian Academy of Sciences, Vavilova Street, 44, Building 2, 119333 Moscow, Russia.

Abstract

In this review, we focused on the applicability of artificial intelligence (AI) for opportunistic abdominal aortic aneurysm (AAA) detection in computed tomography (CT). We used the academic search system PubMed as the primary source for the literature search and Google Scholar as a supplementary source of evidence. We searched through 2 February 2022. All studies on automated AAA detection or segmentation in noncontrast abdominal CT were included. For bias assessment, we developed and used an adapted version of the QUADAS-2 checklist. We included eight studies with 355 cases, of which 273 (77%) contained AAA. The highest risk of bias and level of applicability concerns were observed for the "patient selection" domain, due to the 100% pathology rate in the majority (75%) of the studies. The mean sensitivity value was 95% (95% CI 100-87%), the mean specificity value was 96.6% (95% CI 100-75.7%), and the mean accuracy value was 95.2% (95% CI 100-54.5%). Half of the included studies performed diagnostic accuracy estimation, with only one study having data on all diagnostic accuracy metrics. Therefore, we conducted a narrative synthesis. Our findings indicate high study heterogeneity, requiring further research with balanced noncontrast CT datasets and adherence to reporting standards in order to validate the high sensitivity value obtained.

Keywords: QUADAS; abdominal aortic aneurysm; artificial intelligence; computed tomography; opportunistic screening.

Publication types

Review

Grants and funding

This paper was prepared by a group of authors as a part of the medical research project (No. USIS (in the Unified State Information System for Accounting of Research, Development, and Technological Works): 122112400040-1) “Reference datasets for sustainable development of artificial intelligence-based diagnostics to minimize the long-term impact of the COVID-19 pandemic on the health of the Moscow population”.