Artificial Intelligence and Machine Learning in Radiology: Opportunities, Challenges, Pitfalls, and Criteria for Success.

@article{Thrall2018ArtificialIA,
  title={Artificial Intelligence and Machine Learning in Radiology: Opportunities, Challenges, Pitfalls, and Criteria for Success.},
  author={James Thrall and Xiang Li and Quanzheng Li and Cinthia Cruz and Synho Do and Keith J. Dreyer and James A. Brink},
  journal={Journal of the American College of Radiology : JACR},
  year={2018},
  volume={15 3 Pt B},
  pages={
          504-508
        }
}
Artificial intelligence in medical imaging and radiation oncology: Opportunities and challenges
TLDR
Artificial intelligence can combine many and disparate data sources and types into decision-making or interrogate very large sets of retrospective data to learn patterns where manual investigation would not be feasible or would not readily identify them, for example in early sub-clinical disease detection.
Interpretative applications of artificial intelligence in musculoskeletal imaging: concepts, current practice, and future directions
TLDR
Artificial intelligence and ML advances for image interpretation can increase the value that MSK radiologists provide to their patients, referring clinicians, and organizations by increasing diagnostic accuracy while decreasing turnaround times, enhancing image processing and quantitative analysis, and by potentially improving patient outcomes.
Artificial Intelligence and Machine Learning in Radiology
TLDR
The relevance of appropriate imaging data as a bottleneck for innovation is described in detail to provide insights into the many obstacles for technical implementation, and give additional perspectives to radiologists who often view AI solely from their clinical role.
Engineering and clinical use of artificial intelligence (AI) with machine learning and data science advancements: radiology leading the way for future
TLDR
The role of artificial intelligence (AI) and its contribution to data science as well as various learning algorithms in radiology are reviewed and the potential applications in image interpretation and radiological advances for AI are explored.
Artificial Intelligence: A Primer for Breast Imaging Radiologists.
  • M. Bahl
  • Computer Science
    Journal of breast imaging
  • 2020
TLDR
Key terminology and concepts are reviewed, common AI models and methods to validate and evaluate these models are discussed, emerging AI applications in breast imaging are described, and challenges and future directions are outlined.
Artificial intelligence and the radiologist: the future in the Armed Forces Medical Services
TLDR
Early widespread incorporation of information technology and enabled services in the authors' hospitals, adequate funding, regular upgradation of software and hardware, dedicated trained manpower to manage the information technology services and train staff, and cyber security are issues that need to be addressed.
Artificial Intelligence in Medical Imaging
TLDR
The current research status of artificial intelligence (AI) in Oncology is discussed, reviewing the basics of machine learning (ML) and deep learning (DL) techniques and their emerging applications on clinical and imaging cancer workflow.
Reviewing the relationship between machines and radiology: the application of artificial intelligence
  • R. Ahmad
  • Medicine
    Acta radiologica open
  • 2021
TLDR
Radiomic information extracted through machine learning programs form images that may not be discernible through visual examination, thus may improve the prognostic and diagnostic value of data sets.
Key Success Factors for Successful Implementation of AI Based Segmentation Algorithms in Clinical Radiology Practice
TLDR
The development of artificial intelligence tools based on machine learning and deep learning capable of performing several tasks such as image classification or regression, organ segmentation or feature extraction, has soared over the past few years.
...
...

References

SHOWING 1-10 OF 22 REFERENCES
Deep Learning for Health Informatics
TLDR
A comprehensive up-to-date review of research employing deep learning in health informatics is presented, providing a critical analysis of the relative merit, and potential pitfalls of the technique as well as its future outlook.
Pulmonary Nodule Detection in CT Images: False Positive Reduction Using Multi-View Convolutional Networks
TLDR
It was showed that the proposed multi-view ConvNets is highly suited to be used for false positive reduction of a CAD system.
The Potential of Radiomic-Based Phenotyping in Precision Medicine: A Review.
  • H. Aerts
  • Medicine, Biology
    JAMA oncology
  • 2016
TLDR
The semantic features approach to tumor phenotyping, accomplished by visual assessment of radiologists, is compared with a computational radiomic approach that relies on automated processing of imaging assays to capture important information for diagnostic, prognostic, and predictive purposes.
How much data is needed to train a medical image deep learning system to achieve necessary high accuracy
TLDR
This research will present a general methodology for determining the training data set size necessary to achieve a certain target classification accuracy that can be easily applied to other problems within such systems.
Fully Automated Deep Learning System for Bone Age Assessment
TLDR
A fully automated deep learning pipeline to segment a region of interest, standardize and preprocess input radiographs, and perform BAA was proposed and deployed in the clinical environment for more accurate and efficient BAAs at much faster interpretation time (<2 s) than the conventional method.
Lung Pattern Classification for Interstitial Lung Diseases Using a Deep Convolutional Neural Network
TLDR
A comparative analysis proved the effectiveness of the proposed CNN against previous methods in a challenging dataset, and demonstrated the potential of CNNs in analyzing lung patterns.
Dermatologist-level classification of skin cancer with deep neural networks
TLDR
This work demonstrates an artificial intelligence capable of classifying skin cancer with a level of competence comparable to dermatologists, trained end-to-end from images directly, using only pixels and disease labels as inputs.
The Cancer Imaging Archive (TCIA): Maintaining and Operating a Public Information Repository
TLDR
The management tasks and user support model for TCIA is described, an open-source, open-access information resource to support research, development, and educational initiatives utilizing advanced medical imaging of cancer.
Multi-scale Convolutional Neural Networks for Lung Nodule Classification
TLDR
A hierarchical learning framework--Multi-scale Convolutional Neural Networks (MCNN)-- is proposed to capture nodule heterogeneity by extracting discriminative features from alternatingly stacked layers to sufficiently quantify nodule characteristics.
Deep Learning for Medical Image Analysis
TLDR
Different novel methods based on deep learning for brain abnormality detection, recognition, and segmentation for analyzing medical images using deep learning algorithm are explored.
...
...