Please use this identifier to cite or link to this item: http://ahro.austin.org.au/austinjspui/handle/1/19644
Title: Peering Into the Black Box of Artificial Intelligence: Evaluation Metrics of Machine Learning Methods.
Authors: Handelma, Guy S;Kok, Hong Kuan;Chandra, Ronil V;Razavi, Amir H;Huang, Shiwei;Brooks, Duncan Mark;Lee, Michael J;Asadi, Hamed
Affiliation: The Florey Institute of Neuroscience and Mental Health, University of Melbourne, VIC, Australia
Interventional Neuroradiology Service, Department of Radiology, Austin Health, Heidelberg, Victoria, Australia
Department of Radiology, Beaumont Hospital, Dublin, Ireland
Department of Radiology, Belfast City Hospital, 51 Lisburn Rd, Belfast, Antrim BT9 7AB, Northern Ireland, UK
Royal College of Surgeons in Ireland, Dublin, Ireland
Interventional Radiology Service, Northern Hospital Radiology, Epping, Victoria, Australia
School of Medicine, Faculty of Health, Deakin University, Waurn Ponds, VIC, Australia
School of Information Technology and Engineering, University of Ottawa, Ottawa, ON, Canada
BCE Corporate Security, Ottawa, ON, Canada
Interventional Neuroradiology Service, Monash Imaging, Monash Health, Clayton, Victoria, Australia
Faculty of Medicine, Nursing and Health Sciences, Monash University, Clayton, Victoria, Australia
The Australian National University Medical School, Garran, Australia
Issue Date: 17-Oct-2018
EDate: 2018-10-17
Citation: AJR. American journal of roentgenology 2018: online first: 17 October
Abstract: Machine learning (ML) and artificial intelligence (AI) are rapidly becoming the most talked about and controversial topics in radiology and medicine. Over the past few years, the numbers of ML- or AI-focused studies in the literature have increased almost exponentially, and ML has become a hot topic at academic and industry conferences. However, despite the increased awareness of ML as a tool, many medical professionals have a poor understanding of how ML works and how to critically appraise studies and tools that are presented to us. Thus, we present a brief overview of ML, explain the metrics used in ML and how to interpret them, and explain some of the technical jargon associated with the field so that readers with a medical background and basic knowledge of statistics can feel more comfortable when examining ML applications. Attention to sample size, overfitting, underfitting, cross validation, as well as a broad knowledge of the metrics of machine learning, can help those with little or no technical knowledge begin to assess machine learning studies. However, transparency in methods and sharing of algorithms is vital to allow clinicians to assess these tools themselves.
URI: http://ahro.austin.org.au/austinjspui/handle/1/19644
DOI: 10.2214/AJR.18.20224
ORCID: 0000-0003-2475-9727
PubMed URL: 30332290
Type: Journal Article
Subjects: artificial intelligence
machine learning
medicine
supervised machine learning
unsupervised machine learning
Appears in Collections:Journal articles

Files in This Item:
There are no files associated with this item.


Items in AHRO are protected by copyright, with all rights reserved, unless otherwise indicated.