The medical imaging industry is rapidly evolving to bring new tools to market to facilitate medical imaging demands. At ECR 2022, we saw a new focus on artificial intelligence (AI) and digitization and the efficiencies they can bring to radiology.
The number of diagnostic images acquired in recent years has skyrocketed, with an increase of 100% in the last 10 years, while the number of radiologists has grown by 2% in the same period. This has created a heavy workload and stress.
At the same time, the COVID-19 pandemic has put even more pressure on radiology staff. A 2021 study conducted in the UK showed how the crisis had changed the landscape of clinical practice in terms of work patterns and workload, causing workplace-related anxiety and stress among radiologic technologists.
At ECR we saw vendors demonstrate innovative solutions to address many of these challenges. Diagnostic accuracy and standardization were the main trends of ECR. Many medical artificial intelligence (AI) vendors and equipment manufacturers showcased updated hardware and solutions, with a focus on systems integration. Unlike enterprise healthcare solutions, medical AI is often sold as a one-time cost along with hardware.
Philips showcased its connected intelligent imaging systems and integrated radiology workflow solutions, which help address operational inefficiencies in radiology departments. This included the SmartSpeed ​​MRI acceleration protocol.
In the meantime, Canon Medical Systems unveiled the Vantage Fortian MRI Scanner in ECR 2022, which has accelerated scanning technology.
Siemens Healthineers showed its research on the Using an AI Algorithm to Classify Mammography Exams; the tool was able to speed up efficiency by reducing patient recovery rates.
Finally, GE Healthcare demonstrated a comprehensive breast care clinic, recreating the path a woman would take through a screening center, from the waiting room to the pathology lab. The concept included options that allowed women to customize the experience to their own preferences, such as selecting a background for the mammography room from multiple landscape photos.
Medical imaging AI continues to evolve into 2022 and beyond
We were impressed by the scale of ECR ​​2022 and the positivity of the atmosphere. There was also a good mix of startups, mega-corporations and industry players in between. I’ve often heard it said that crowds were down compared to “normal” (ie pre-COVID) years, but there was still a good buzz.
Imaging AI, one of our abiding interests, was omnipresent at ECR 2022: all the major medical imaging equipment manufacturers are now developing AI-based software for their machines. And there are also a lot of software startups in this space, a fact that was fully on display at ECR.
To what extent medical imaging AI has already penetrated the market is a key question. Omdia conducted a survey of radiographers and radiologists (n = 95) which found that image cleanup was the AI ​​application used by the majority of imaging professionals (59% of those who had used image AI), with the physiological measurement the second (used by 48%).
However, 52% had never used medical imaging AI apps and 20% were unsure. Only 28% said they had actually used AI. This is certainly a lot of progress compared to three years ago, but a lot remains to be done for the widespread deployment of imagery AI.
The AI ​​paradox
So AI was everywhere in ECR, but it’s a relatively fringe phenomenon in terms of clinical practice. What is the answer to this paradox?
We discussed this topic several times at ECR and learned that there are some fundamental issues that need to be addressed before the potential of AI is fully realized in radiology.
First, there is the crucial issue of legal liability. Software developers don’t want to be sued. Human radiologists are already prepared to take legal responsibility in this situation: it is part of their job and the rules are well established. It will take work to address the disruptive influence of AI systems in this space.
The second problem is time: Medical imaging AI is still so new that most clinics or hospital departments simply haven’t had the time to implement commercial AI systems yet. People are still getting used to the idea.
Our survey provides a detailed profile of medical imaging professionals’ attitudes toward medical imaging AI. Overall, the results are encouraging that there is an openness to medical imaging AI and that the fear of “being replaced” is not widespread. This paves the way for the continued deployment of medical AI systems.
Where will we be with medical imaging AI in five years, say at ECR 2027? By then, we imagine that around 80% of radiographers and radiologists will be using AI systems. Confidence will have been gained by AI developers and the list of applications, both in development and proposed, will have expanded, and implementation will proceed at a rapid pace.
Rapid digitization of radiology
Radiology is poised for significant change in the next 10 years. Increasing labor shortages, increasing demand for radiology services, and exponential growth in imaging data drive the adoption of a digitization strategy in radiology.
From a technological point of view, the main pillars of the digitization strategy include the following:
- internet of medical things
- Connected medical devices and health sensors
- Cloud
- Software
- AI and data analysis
- 3d print
- Virtual reality
This year’s ECR 2022 welcomed key developments in the digitization of radiology. Major trends and themes include the following:
- Applications in AI-driven decision making
- Workflow optimization
- Adoption of digital platforms
- Rise of service-based models in radiology
In the next few paragraphs, we will outline the current state of developments and industry sentiment.
Applications in AI-driven decision making
The exponential growth of medical imaging data and the shortage of radiologists remain key challenges in radiology. The focus of the current AI-enabled applications featured at ECR is to empower radiologists with tools for faster and more accurate clinical decision-making.
The current focus of applications is primarily on image capture and analysis. As for image capture, the algorithms focus on reducing image noise and improving image quality. Additional features include automated acquisition, patient positioning to reduce human error, and technology to reduce patient radiation exposure.
Although the goal of the industry remains the development of AI for assisted diagnosis, such as early detection of tissue abnormalities and early detection of disease, currently most AI-enabled applications are focused on “soft” diagnostic applications. “, such as the filtering of non-suspicious images to allow radiologists to focus only on suspicious cases.
Workflow optimization continues to be at the epicenter of efforts to achieve clinical and operational efficiency, and this will continue for the foreseeable future. Innovation extends across the touchpoints of the radiology patient journey, from patient record acquisition to patient selection and image acquisition.
Adoption of digital platforms
Digital platforms have become key workflow command centers. Leading player platforms such as Philips and GE Healthcare act as command centers encompassing diverse solutions from multiple vendors that aim to simplify operations across multiple modalities, create intelligent workflows by reducing workload and increasing productivity of the labor force.
At ECR 2022, vendors showcased platforms that have been enriched with artificial intelligence to provide real-time contextual insight and operational insights into imaging departments. The platforms are essentially new business propositions for the industry where radiologists, technologists and OEMs collaborate to provide an ecosystem of applications that solve critical problems in the radiology department.
Given the complexity and challenges ahead, as well as the influence of value-based collaborative systems driven by cost containment, Omdia analysts believe the platforms are the first examples of emerging digital models in the industry. variety of service-based models that will enter the market in the next five years.
The future of digital radiology is a connected, interoperable and collaborative one, empowered with real-time and predictive information that supports collaboration in care throughout the patient journey.
Felix Beacher, Dionysia Patrinou, and Sharjeel Ahmad are market research analysts with Omdia, a division of Informa. You can contact them at [email protected].
The comments and observations expressed are those of the authors and do not necessarily reflect the opinions of tiaminnie.com.
Copyright © 2022 AuntMinnie.com