Abstract
Artificial Intelligence (AI) has taken radiology by storm, in particular, mammogram interpretation, and we have seen a recent surge in the number of publications on potential uses of AI in breast radiology. Breast cancer exerts a lot of burden on the National Health Service (NHS) and is the second most common cancer in the UK as of 2018. New cases of breast cancer have been on the rise in the past decade, while the survival rate has been improving. The NHS breast cancer screening program led to an improvement in survival rate. The expansion of the screening program led to more mammograms, thereby putting more work on the hands of radiologists, and the issue of double reading further worsens the workload. The introduction of computer-aided detection (CAD) systems to help radiologists was found not to have the expected outcome of improving the performance of readers. Unreliability of CAD systems has led to the explosion of studies and development of applications with the potential use in breast imaging. The purported success recorded with the use of machine learning in breast radiology has led to people postulating ideas that AI will replace breast radiologists. Of course, AI has many applications and potential uses in radiology, but will it replace radiologists? We reviewed many articles on the use of AI in breast radiology to give future radiologists and radiologists full information on this topic. This article focuses on explaining the basic principles and terminology of AI in radiology, potential uses, and limitations of AI in radiology. We have also analysed articles and answered the question of whether AI will replace radiologists.
Highlights
BackgroundReceived 05/05/2020 Review began 06/09/2020 Review ended 06/21/2020 Published 06/30/2020Women in the UK have one in eight chances of developing cancer of the breast during their lives [1]
The purported success recorded with the use of machine learning in breast radiology has led to people postulating ideas that Artificial Intelligence (AI) will replace breast radiologists
AI has many applications and potential uses in radiology, but will it replace radiologists? We reviewed many articles on the use of AI in breast radiology to give future radiologists and radiologists full information on this topic
Summary
Received 05/05/2020 Review began 06/09/2020 Review ended 06/21/2020 Published 06/30/2020. A deep neural network (DNN) is an ANN consisting of five or more layers of algorithms connected and organised according to the meaningfulness of the data, and this enables improved predictions from data [13,14,15,16] These layers store data from inputs and provide an output that is liable to change in an orderly manner once the AI system learns new features from the data [13,14,15,16]. We have seen algorithms in various domains that are comparable or even better than humans in various radiological tasks, especially in breast imaging, which is why it is our main focus in this article [2945] Because of this success of DL in breast imaging, people started discussing the possibility of automating image interpretation and bury radiologists. There is a need for the development of rigorous evaluation criteria of applications before they are licensed for use and radiologists should take this responsibility [29-47]
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
