Review
Radiology, Nuclear Medicine & Medical Imaging
Brendan S. Kelly, Conor Quinn, Niamh Belton, Aonghus Lawlor, Ronan P. Killeen, James Burrell
Summary: Radiology AI projects involve integrating multiple medical devices, wireless technologies, data warehouses, and social networks. The rise of AI research in radiology has increased cybersecurity threats in healthcare, making them a major risk in 2021. This review provides an introduction to cybersecurity concepts in medical imaging and discusses approaches to enhance security through detection and prevention techniques. It also suggests potential risk mitigation strategies for radiology AI projects.
EUROPEAN RADIOLOGY
(2023)
Article
Multidisciplinary Sciences
Jane Scheetz, Philip Rothschild, Myra McGuinness, Xavier Hadoux, H. Peter Soyer, Monika Janda, James J. J. Condon, Luke Oakden-Rayner, Lyle J. Palmer, Stuart Keel, Peter van Wijngaarden
Summary: The survey revealed that the majority of medical specialists believe that artificial intelligence will improve healthcare outcomes and impact workforce demands in the next decade. Key benefits of artificial intelligence include improved disease screening and streamlining of monotonous tasks, while concerns revolve around medical liability and the involvement of technology companies in healthcare.
SCIENTIFIC REPORTS
(2021)
Article
Chemistry, Multidisciplinary
Rowa Aljondi, Salem Saeed Alghamdi, Abdulrahman Tajaldeen, Shareefah Alassiri, Monagi H. Alkinani, Thomas Bertinotti
Summary: The AI system tested in this study showed better accuracy and diagnostic performance than radiologists, making it a potential support diagnostic tool for breast cancer detection in clinical practice and reducing false-positive recalls.
APPLIED SCIENCES-BASEL
(2023)
Article
Oncology
Stylianos Kotsyfakis, Evangelia Iliaki-Giannakoudaki, Antonios Anagnostopoulos, Eleni Papadokostaki, Konstantinos Giannakoudakis, Michail Goumenakis, Michail Kotsyfakis
Summary: This scoping review examines the application of machine learning techniques to hematological malignancy imaging. The findings suggest that there are diverse ML methods being used for diagnosis, segmentation, and prognostication in this field. However, the studies reviewed have limitations such as high risk of bias and lack of independent validation. To improve the accuracy of ML-based models in managing hematological malignancies, future research should focus on standardized reporting, independent validation, and comprehensive prospective studies.
FRONTIERS IN ONCOLOGY
(2022)
Article
Radiology, Nuclear Medicine & Medical Imaging
Mohammad H. Rezazade Mehrizi, Simon H. Gerritsen, Wouter M. de Klerk, Chantal Houtschild, Silke M. H. Dinnessen, Luna Zhao, Rik van Sommeren, Abby Zerfu
Summary: In the field of diagnostic radiology workflow, providers of AI solutions promote their products by proposing a variety of value propositions related to improving work efficiency and medical service quality, while using multiple strategies to legitimize and support these value propositions. However, these companies often provide limited evidence to show how their solutions deliver such systematic values in clinical practice.
EUROPEAN RADIOLOGY
(2023)
Article
Radiology, Nuclear Medicine & Medical Imaging
Maria Camila Velez-Florez, Adarsh Ghosh, Daniela Patton, Raymond Sze, Janet R. Reid, Susan Sotardi
Summary: This study aims to assess the needs for the development of an AI curriculum during pediatric radiology training and continuing education. A focus group study was conducted to understand the perceptions, competence, and expectations of radiology trainees and attending radiologists regarding AI. The results revealed heterogeneity in perspectives, with variations in AI knowledge, previous training, learning preferences, expectations, and concerns. The participants expressed a preference for case-based teaching and highlighted the need for improved training in interpreting and applying AI literature.
ACADEMIC RADIOLOGY
(2023)
Article
Radiology, Nuclear Medicine & Medical Imaging
Mohammad Hosein Rezazade Mehrizi, Peter van Ooijen, Milou Homan
Summary: The study systematically analyzed the landscape of AI applications in diagnostic radiology, finding that the majority of AI applications are narrow in terms of tasks, modality, and anatomic region. Most AI functionalities focus on supporting perception and reasoning in the radiology workflow.
EUROPEAN RADIOLOGY
(2021)
Review
Radiology, Nuclear Medicine & Medical Imaging
Ling Yang, Ioana Cezara Ene, Reza Arabi Belaghi, David Koff, Nina Stein, Pasqualina (Lina) Santaguida
Summary: Stakeholders generally believe that AI can enhance radiology practice and view the replacement of radiologists as unlikely. Most stakeholders express the need for education and training on AI, as well as collaborative efforts to enhance AI implementation. Further research is required to obtain perspectives from non-Western countries, non-radiologist stakeholders, on economic considerations, and medicolegal implications.
EUROPEAN RADIOLOGY
(2022)
Review
Radiology, Nuclear Medicine & Medical Imaging
Katharine A. Dishner, Bala McRae-Posani, Arka Bhowmik, Maxine S. Jochelson, Andrei Holodny, Katja Pinker, Sarah Eskreis-Winkler, Joseph N. Stember
Summary: Researchers reviewed publicly available MRI datasets in the field of radiology and summarized the important features for AI researchers, serving as a reference for advancements in AI for radiology.
JOURNAL OF MAGNETIC RESONANCE IMAGING
(2023)
Review
Medicine, General & Internal
Hayley Higgins, Abanoub Nakhla, Andrew Lotfalla, David Khalil, Parth Doshi, Vandan Thakkar, Dorsa Shirini, Maria Bebawy, Samy Ammari, Egesta Lopci, Lawrence H. Schwartz, Michael Postow, Laurent Dercle
Summary: This article discusses the potential of artificial intelligence techniques in medical imaging, particularly in the management of metastatic cutaneous melanoma. By analyzing information from medical images, AI can provide insights into the biology of tumors, improving patient care and clinical treatment.
Article
Radiology, Nuclear Medicine & Medical Imaging
Summary: A survey conducted among European Society of Radiology members examines the practical clinical experience of radiologists with AI-powered tools. The majority of radiologists experienced no reduction in workload, but AI algorithms were found to be reliable for different use case scenarios.
INSIGHTS INTO IMAGING
(2022)
Review
Radiology, Nuclear Medicine & Medical Imaging
Brendan S. Kelly, Conor Judge, Stephanie M. Bollard, Simon M. Clifford, Gerard M. Healy, Awsam Aziz, Prateek Mathur, Shah Islam, Kristen W. Yeom, Aonghus Lawlor, Ronan P. Killeen
Summary: This systematic review examines the advances in artificial intelligence (AI) applied to clinical radiology. The study found that most research in this field uses supervised learning and focuses on segmentation tasks. The UNet architecture is commonly used for performance comparison. The results indicate the potential application of AI in clinical radiology.
EUROPEAN RADIOLOGY
(2022)
Article
Radiology, Nuclear Medicine & Medical Imaging
Minerva Becker
Summary: Although AI-powered tools may change the way radiologists work, they will not replace them completely. Radiologists have unique advantages in multidisciplinary and patient-centered consulting tasks. To adapt to this trend, training for radiologists should focus on clinical backgrounds, communication skills, and integration with other disciplines.
INSIGHTS INTO IMAGING
(2023)
Article
Radiology, Nuclear Medicine & Medical Imaging
Francesca Coppola, Lorenzo Faggioni, Daniele Regge, Andrea Giovagnoni, Rita Golfieri, Corrado Bibbolino, Vittorio Miele, Emanuele Neri, Roberto Grassi
Summary: Italian radiologists had a mostly positive attitude towards the implementation of AI in their working practice, with concerns focused on the potential damage to their professional reputation rather than replacement by AI. Most radiologists believed that specific policies should regulate the use of AI.
Article
Radiology, Nuclear Medicine & Medical Imaging
Chanel Fischetti, Param Bhatter, Emily Frisch, Amreet Sidhu, Mohammad Helmy, Matt Lungren, Erik Duhaime
Summary: This paper investigates the current state of radiologic education within medical trainee curricula and explores the potential impact of artificial intelligence on the current and future models of radiologic education.
ACADEMIC RADIOLOGY
(2022)
Article
Cardiac & Cardiovascular Systems
Jeffrey B. Geske, C. Noelle Driver, Vidhushei Yogeswaran, Steve R. Ommen, Hartzell Schaff
AMERICAN HEART JOURNAL
(2020)
Article
Anesthesiology
C. Noelle Driver, Mariana L. Laporta, Sergio D. Bergese, Richard D. Urman, Fabio Di Piazza, Frank J. Overdyk, Juraj Sprung, Toby N. Weingarten
Summary: A study revealed that postoperative respiratory depression (RD) events are common in surgical patients, with risk increasing with higher PRODIGY scores. The peak occurrence of initial RD events happens in the afternoon to early evening, while the peak time for all RD events is in the early morning. Additionally, patients with higher PRODIGY risk scores experienced a greater number of RD episodes.
ANESTHESIA AND ANALGESIA
(2021)