Effect of artificial intelligence-based triaging of breast cancer screening mammograms on cancer detection and radiologist workload: a retrospective simulation study

Abstract

Background We examined the potential change in cancer detection when using an artificial intelligence (AI) cancer-detection software to triage certain screening examinations into a no radiologist work stream, and then after regular radiologist assessment of the remainder, triage certain screening examinations into an enhanced assessment work stream. The purpose of enhanced assessment was to simulate selection of women for more sensitive screening promoting early detection of cancers that would otherwise be diagnosed as interval cancers or as next-round screen-detected cancers. The aim of the study was to examine how AI could reduce radiologist workload and increase cancer detection.Methods In this retrospective simulation study, all women diagnosed with breast cancer who attended two consecutive screening rounds were included. Healthy women were randomly sampled from the same cohort; their observations were given elevated weight to mimic a frequency of 0•7% incident cancer per screening interval. Based on the prediction score from a commercially available AI cancer detector, various cutoff points for the decision to channel women to the two new work streams were examined in terms of missed and additionally detected cancer.Findings 7364 women were included in the study sample: 547 were diagnosed with breast cancer and 6817 were healthy controls. When including 60%, 70%, or 80% of women with the lowest AI scores in the no radiologist stream, the proportion of screen-detected cancers that would have been missed were 0, 0•3% (95% CI 0•0-4•3), or 2•6% (1•1-5•4), respectively. When including 1% or 5% of women with the highest AI scores in the enhanced assessment stream, the potential additional cancer detection was 24 (12%) or 53 (27%) of 200 subsequent interval cancers, respectively, and 48 (14%) or 121 (35%) of 347 next-round screen-detected cancers, respectively.Interpretation Using a commercial AI cancer detector to triage mammograms into no radiologist assessment and enhanced assessment could potentially reduce radiologist workload by more than half, and pre-emptively detect a substantial proportion of cancers otherwise diagnosed later.

La inteligencia artificial, una herramienta para la Salud Digital

“…Al respecto, detallamos como la IA aporta grandes beneficios mediante reconocimiento automático de patrones complejos de imágenes radiológicas, con evaluaciones cuantitativas derivadas de cualidades independientemente de la ubicación anatómica o patología involucrada, según lo establecido por Hosny et al, (2018). Adicionalmente, Dembrower et al, (2020) destacan su exactitud diagnóstica que supera la perceptibilidad de la visión humana, tomando esta ventaja en la detección del cáncer mamario; mientras Curioso & Brunette, (2020) presentan un algoritmo de aprendizaje profundo de análisis remoto de rayos X (eRx) que contribuye con el diagnóstico oportuno de tuberculosis. En diagnóstico microscópico, los trabajos de Torres et al, (2018) sobre algoritmos de aprendizaje automático nos revelan que su rendimiento diagnóstico, es equiparable a la microscopía óptica en el reconocimiento de parásitos maláricos en humanos, como P. falciparum y P. vivax.…”

“…Aplicando el anterior concepto, Dembrower et al, (2020) estudiaron una población de 7.364 mujeres que asistieron a dos rondas de detección consecutivas, de las que 547 fueron previamente diagnosticadas con cáncer de mama y 6817 correspondían a resultados normales, mediante un software de detección de cáncer clasificando exámenes en un flujo de trabajo sin radiólogo; resultando en una detección adicional de cáncer del 2,6% (1,1 a 5,4%, IC = 95%) que se habían pasado por alto en los análisis humanos, siendo una potencial herramienta preventiva al detectar una proporción sustancial de cánceres que de otro modo se diagnosticarían más tarde.…”

How does artificial intelligence in radiology improve efficiency and health outcomes?

“…The potential of AI for workflow optimization is also being explored in use cases, but this has not been put into clinical practice. For mammography screening, for example, studies have been performed to simulate an alternative workflow in which an AI risk score determines the number of radiology reads (none, single or double), reducing the total amount of reading time (level 3) [14,15]. For lung nodule detection on CT, it has been proposed to empower technicians with AI to leave only some of the workload to the radiologists (level 3) [16].…”

External validation of AI algorithms in breast radiology: the last healthcare security checkpoint?

“…Furthermore, these systems have been proposed as an option for independent reading. Along these lines, a recent report by Dembrower et al concluded that the use of AI CAD systems for triage mammograms without radiologist supervision could potentially reduce radiologists' workload by more than half and also pre-emptively detect a substantial proportion of cancers, which would otherwise be diagnosed later (16). In the short-term, the most realistic scenario regarding the use of AI CAD tools for breast screening diagnosis will probably be focused on their use as an independent second reader in order to reduce the radiologist's workload (17).…”