Artificial Intelligence, Machine Learning and Radiomics in Radiology
The use of artificial intelligence (AI) in radiology – radiomics – has been getting a lot of attention, fuelled by the availability of large datasets, substantial advances in computing power, and new deep-learning algorithms. This has led to a rapid rise in the potential use of artificial intelligence in various radiological imaging tasks. Machine learning in radiology, a subset of artificial intelligence, is expected to have a substantial clinical impact, with the imaging examinations routinely obtained in clinical practice providing an opportunity to improve decision support in medical image interpretation.
On this page, we present a collection of articles showcasing recent research on these topics. Studies undertaken confirm the expectation that AI in radiology will bring significant changes. In an article in Academic Radiology, doctors at New Jersey Medical School discuss this impact and the benefits it could bring, predicting that artificial intelligence will lead to a fundamental change in practice in many professional fields. Deep learning (DL), part of a broader family of machine learning methods, has recently become a remarkably powerful tool for image processing.
The challenges and potential pitfalls to this tool are discussed in a study published in the Journal of the American College of Radiology, where it is concluded that for deep learning in radiology to succeed, appropriately annotated, large data sets are needed. A further study in Clinical Radiology looks at how a machine-learning based approach can be applied to classify chest radiographs as normal or abnormal. The anticipation of AI's role in clinical radiology has been met with anxiety about job security by some, and optimism about the potential to streamline monotonous functions by others. Irrespective of how they feel, researchers in an article published in Clinical Imaging strongly feel that radiologists should ensure they are at the forefront of this inevitable marriage of man and machine, molding AI into a powerful tool that will solidify their position as the fulcrum in clinical medicine.
You can find out more about these studies and others in our special collection on Artificial Intelligence, Machine Learning and Radiomics, which is free to access online until 31 December 2018.
Deep Learning in Radiology
By: M. P. McBee et al.
Toward Augmented Radiologists: Changes in Radiology Education in the Era of Machine Learning and Artificial Intelligence
By: S. H. Tajmir & T. K. Alkasab
Canadian Association of Radiologists Journal
Artificial intelligence in clinical imaging: An introduction
By: A. Starikov et al.
Artificial intelligence and deep learning – Radiology's next frontier?
By: R. C. Mayo & J. Leung
Machine learning "red dot": open-source, cloud, deep convolutional neural networks in chest radiograph binary normality classification
By: E. J. Yates, L. C. Yates & H. Harvey
Artificial intelligence in fracture detection: transfer learning from deep convolutional neural networks
By: D. H. Kim & T. MacKinnon
European Journal of Radiology
The past, present and future role of artificial intelligence in imaging
By: M. I. Fazal et al.
Convolutional auto-encoder for image denoising of ultra-low-dose CT
By: M. Nishio et al.
Journal of Forensic Radiology and Imaging
Journal of the American College of Radiology
Artificial Intelligence: Threat or Boon to Radiologists?
By: M. Recht & R. N. Bryan
Machine Learning in Medical Imaging
By: M. L. Giger
Journal of Vascular and Interventional Radiology
Magnetic Resonance Imaging
Radiomics in radiooncology – Challenging the medical physicist
By: J. C. Peeken et al.