The impact of artificial intelligence on double reading of mammograms
- 23 April 2024
- journal article
- research article
- Published by Media Sphere Publishing Group in Russian Journal of Preventive Medicine
- Vol. 27 (5)
Abstract
Mammography is a gold standard for early diagnosis of breast tumors. Independent double-reading of mammograms by two different radiologists is recommended to use in order to provide a high quality of preventive examination and to detect additional cases of malignant neoplasms. Nevertheless, this approach increases the work load and time of mammography results receiving. OBJECTIVE To assess the impact of artificial intelligence (AI) algorithm on the duration of double reading of mammograms. MATERIAL AND METHODS The study was conducted at the Moscow Reference Centre for Radiology Diagnostics (Research and Practical Clinical Center for Diagnostics and Telemedicine Technologies, telemedai.ru). Mammograms were reviewed and radiology reports were prepared in the Unified Radiological Information Service of the Unified Medical Information and Analytical System in Moscow. Twelve radiologists who specialize in mammogram descriptions participated in the study. Russian-made softwarewas used for automatic analysis of mammograms. Two scenarios were implemented as a part of the study: the first — double-reading by radiologists, the second — first reading by AI algorithm, second reading by radiologist. RESULTS In the first scenario, a double reading was performed for 480 mammograms, in the second one — 510 mammograms. The average duration of the double-reading for the first scenario was 34:12:18 (hh:mm:ss), for the second one — 11:28:58. Application of AI algorithm allowed to reduce the duration of double-reading by 66.4% (p<0.0001). CONCLUSION The use of artificial intelligence algorithm in analysis of preventive mammography examinations allowed to reduce the duration of double-reading and work load on radiologists.Keywords
This publication has 17 references indexed in Scilit:
- An Artificial Intelligence–based Mammography Screening Protocol for Breast Cancer: Outcome and Radiologist WorkloadRadiology, 2022
- Effect of artificial intelligence technologies on the CT scan interpreting time in COVID-19 patients in inpatient settingRussian Journal of Preventive Medicine, 2022
- AI-based Strategies to Reduce Workload in Breast Cancer Screening with Mammography and Tomosynthesis: A Retrospective EvaluationRadiology, 2021
- Effect of artificial intelligence-based triaging of breast cancer screening mammograms on cancer detection and radiologist workload: a retrospective simulation studyThe Lancet Digital Health, 2020
- International evaluation of an AI system for breast cancer screeningNature, 2020
- A Deep Learning Model to Triage Screening Mammograms: A Simulation StudyRadiology, 2019
- Can we reduce the workload of mammographic screening by automatic identification of normal exams with artificial intelligence? A feasibility studyEuropean Radiology, 2019
- National Expenditure For False-Positive Mammograms And Breast Cancer Overdiagnoses Estimated At $4 Billion A YearHealth Affairs, 2015
- Mammographic Screening Programmes in Europe: Organization, Coverage and ParticipationJournal of Medical Screening, 2012
- Mammography screening: an incremental cost effectiveness analysis of double versus single reading of mammogramsBMJ, 1996