Ethics of Artificial Intelligence in Radiology: Summary of the Joint European and North American Multisociety Statement
AI has great potential to increase efficiency and accuracy throughout radiology, but it also carries inherent pitfalls and biases. Widespread use of AI-based intelligent and autonomous systems in radiology can increase the risk of systemic errors with high consequence and highlights complex ethical and societal issues. This statement highlights that ethical use of AI in radiology should promote well-being, minimize harm, and ensure that the benefits and harms are distributed among stakeholders. View the full statement.
AI-LAB Reference Architecture Framework
For a healthcare institution to include artificial intelligence (AI) workflows in medical imaging, an architecture must be designed and implemented to support it. This architecture supports the workflows of inference, model validation, model training, and study annotation, and scales according to the needs of the institution.
Artificial Intelligence and the Practice of Radiology: An Alternative View
The advent of computers that can accurately interpret diagnostic imaging studies will upend the practice of radiology. The two currently unanswered questions are just how much upending there will be and how long it will take to happen.
The Radiologist’s Imitation Game
The recently released films The Imitation Game, Her, and Ex Machina have brought a philosophical quandary to the forefront of popular culture: can machines think or, at least, display behavior indistinguishable from that of a thinking human?
Rest Assured, João, You are Safe From Artificial Intelligence
Dr. Siegal responds to a first-year radiology resident concerned for the future of radiologists as the functions of artificial intelligence grow.
Evaluating Artificial Intelligence Systems to Guide Purchasing Decisions
Many radiologists are considering investments in artificial intelligence (AI) to improve the quality of care for our patients. This article outlines considerations for the purchasing process beginning with performance evaluation.
Artificial Intelligence: A Private Practice Perspective
With scale and a focus on innovation, our practice has had the opportunity to be an early adopter of AI technology. We have gained experience identifying use cases that provide value for our patients and practice; selecting AI products and vendors; piloting vendors’ AI algorithms; creating our own AI algorithms; implementing, optimizing, and maintaining these algorithms; garnering radiologist acceptance of these tools; and integrating AI into our radiologists’ daily workflow.
Creation of an Open Framework for Point-of-Care Computer-Assisted Reporting and Decision Support Tools for Radiologists
Decreasing unnecessary variation in radiology reporting and producing guideline-concordant reports is fundamental to radiology’s success in value-based payment models and good for patient care. In this article, we present an open authoring system for point-of-care clinical decision support tools integrated into the radiologist reporting environment referred to as the computer-assisted reporting and decision support (CAR/DS) framework.
Artificial Intelligence Using Open Source BI-RADS Data Exemplifying Potential Future Use
AI and radiologists working together can achieve better results, helping in case-based decision making. Continued evaluation of the metrics involved in predictor handling by AI algorithms will provide new insights into imaging.
Radiologists in Virginia are trialing a beta version of an application that uses artificial intelligence to detect findings on CT exams.
AI-LAB Evaluation Standards
The ACR has defined a set of standards AI models should adhere to in order to be compatible with AI-LAB. All models within AI-LAB are validated using the same evaluation metrics, allowing users to compare the performances of different AI models directly. The code to calculate the metrics is publicly available in order to increase transparency, maintain research integrity, and encourage public contribution.
The Artificial Intelligence Ecosystem for the Radiological Sciences: Ideas to Clinical Practice
For the most part, individual AI software developers are currently working with individual radiologists at single institutions to create AI algorithms. These developers are using a single institution’s prior imaging data for training and testing the algorithms, and the algorithm output is specifically tailored to that site’s perspective of the clinical workflow. Will theybe generalizable to widespread clinical practices?
Big Data Management, Access, and Protection
It’s hard for a day to go by without some reminder of the explosion of big data, artificial intelligence, machine learning, and data science. Although sometimes used interchangeably, these terms refer to the ability to leverage massive amounts of data to produce algorithms that enhance or substitute for human cognition or executive tasks prone to disruption by human foibles.
ACR Registries Serve Multiple Purposes
One of the singular advantages of working in the digital age is the ability to collect data that can be aggregated and shared for multiple purposes, including improvement of the quality of care offered to our patients and improvement of our management abilities. To help accomplish these goals, the ACR has taken a leadership role in the development of data registries that support the practice of radiology.
ACR Select® helps providers determine the most appropriate imaging exams at the point of care.
Structured for Care
An academic tertiary care center implements structured reporting, achieving 94 percent compliance among radiologists.
Structured for Success
One North Carolina practice establishes a business case for structured reports.
Setting the Standard
In today's value-driven health care environment, standardized language in structured reports allows for improved patient care.
The Perfect Radiology Report
For the greatest impact on patient care, radiologists must write clear and concise reports.
Artificial Intelligence and Radiology: Collaboration Is Key
If we can work together with other entities to facilitate the transition of AI from theory to practical application in radiology, benefits will abound for all involved, with a more rapid introduction of AI to radiology in ways that are best suited to aid radiologists (as opposed to replacing them), ultimately allowing for improved care for patients.
IT: From Complement to Substitute
Technologies that improve worker productivity are key drivers of economic growth. A handful of powerful general-purpose technologies have greatly accelerated the normal rate of growth.
Adding Value Isn’t an Option Anymore
As practicing radiologists, we believe the benefits of our services are self-evident. However, there is a relative paucity of literature that clearly documents our value from the perspective of the health care practitioners who send their patients to us for our services.
AI tools are moving from rare to mainstream. To leverage the power of AI, we’re here to help you understand how it works and optimize its performance. Focused topic areas offer a variety of on-demand videos developed by AI experts in medical imaging and the opportunity to learn, understand and incorporate best practices into your work.
The Fellowship in Informatics provides a radiology resident or early-career professional with hands-on experience in the field of informatics, including one-on-one mentoring. The fellow will be introduced to initiatives of the Data Science Institute®, ACR's AI-LAB™ , and other ACR Informatics projects as part of a three part program. Applications for 2023 will be accepted from February 1 - March 31, 2023.
DSI's Travel Grant for Members-In-Training provides an opportunity for those interested in the emerging disciplines of informatics and artificial intelligence within radiology to join DSI at for the DSI Summit and SIIM annual meeting. Applications will open early fall 2023.
The Informatics Value Proposition for Radiologists
Translating AI to Clinical Practice: Overview of ACR Data Science Institute Initiatives
The Value of Structured Use Cases for Optimizing Machine Learning Practices
Effective Validation of AI Models Prior to Clinical Use
Evaluating AI for Use in Your Clinical Practice
Monitoring AI to Ensure It Continues Working in Our Practices
|Video Series by Bibb Allen, Jr., MD, FACR on how ACR is responding to the development of AI in radiology.|
Will Watson Replace Radiologists
Do Deep-Learning Machines Still Need Human Radiologists?
Presentations by Keith Dreyer, DO, PhD, FACR, FSIIM, chief data science and information officer in the radiology departments of Massachusetts General Hospital and Brigham and Women’s Hospital
In the second part of our conversation on AI, machine learning, and computer-aided diagnosis, we are joined by David Louis, MD, pathologist-in-chief at Massachusetts General Hospital. In this episode, Louis discusses how pathology has approached computer-aided diagnosis, how pathologists are confronting the fears and challenges of potential AI implementations, and his outlook for an exciting future in diagnostic medicine.
Mark Michalski, MD, executive director of the Center for Clinical Data Science at Massachusetts General and Brigham Hospital, joins us as we discuss the history of AI, how it's being used today, how it might change radiology, and how radiologists can leverage this new technology to provide even more value and better patient care.