Влияние технологий искусственного интеллекта на длительность описаний результатов компьютерной томографии пациентов с COVID-19 в стационарном звене здравоохранениястатья
Статья опубликована в журнале из списка RSCI Web of Science
Статья опубликована в журнале из перечня ВАК
Статья опубликована в журнале из списка Web of Science и/или Scopus
Дата последнего поиска статьи во внешних источниках: 8 апреля 2022 г.
Аннотация:ABSTRACT
The practical utility of thoracic computed tomography (CT) imaging in patients with suspected coronavirus infection (COVID-19)
has been repeatedly described. With a heavy workload during the pandemic, radiologists are pressed for time to interpret results.
Artificial intelligence (AI) technologies can reduce the time needed for CT study interpretation and protocol generation.
Objective. To evaluate the effect of the AI algorithm on the chest CT scan interpretation time for suspected COVID-19 in an inpatient
setting.
Material and methods. A retrospective study was conducted and the protocol is registered in ClinicalTrials.gov (NCT04489992).
The study was based on the data of patients who underwent chest CT scan between 08.04.20 and 01.12.20 in 105 municipal inpatient
healthcare organizations. Chest CT scans were performed according to standard scanning protocols. The radiologists analyzed
the scans with and without the Gamma Multivox Covirus AI service. The generation of medical report protocols was carried
out in the Unified Radiology Information Service as part of the Unified Medical Information and Analytical System of Moscow.
Results. 3133 CT studies with signs of COVID-19-associated pneumonia were analyzed without the AI (Group 1) and 63,379 with
the AI (Group 2). The median interpretation time in Groups 1 and 2 was 103.0 and 46.0 min, respectively. Analysis of the radiologist’s
chest CT findings interpretation time before and after AI implementation showed significant differences (p<0.0001). The
average chest CT interpretation time when using AI decreased by 29.4%.
Conclusion. The introduction of AI technology aimed at the pulmonary changes detection in COVID-19 according to chest CT
scanning into the practical work of inpatient radiology departments reduces the time of the interpretation protocol generation
by radiologists.