How innovative technologies such as AI, robotics, and automation are shaping a bright future for radiologists and their patients.
by Christopher Khang
Radiology, also known as diagnostic imaging, has redefined the diagnosis process. Where historically medical decisions and treatments were made based solely on visual examinations and evaluations of symptoms, today, a series of pictures and images provide a clear cognition of what is going on inside a patient’s body, to enable prudent treatment planning. Radiology has become a driving force in advancements and is vital in medical diagnosis, and disease detection and management.
With the healthcare industry continuing rapidly on its path towards precision health, the evolving role of radiology is only set to grow as healthcare professionals and patients seek greater information and insights into the human body.
Radiology seems to be a field increasingly burdened with heavy clinical research, and administrative demands. On the clinical side, it is estimated that some 95 per cent of patients who enter a hospital require imaging, and the demand continues to grow even as fewer young physicians are entering the field.1 In fact, estimates suggest that over 3.5 billion diagnostic X-rays and over 40 million MRIs are performed around the world every year.2
As the demand for radiology services increases, so does the urgent need to provide faster, more accurate and safer diagnoses for patients.
Radiologists, being experts in image capture technology, require key diagnostic tools and intuitive solutions that will enable them to monitor treatment and predict outcomes.
COVID-19 and the Rise of Radiology
When it became clear that COVID-19 would become the top priority to detect and treat inpatients around the globe, hospitals and healthcare systems were forced to abandon their typical annual plans for operational growth and budget goals to deal with the crisis at hand. Radiology was no exception in this turmoil.
In the initial, acute phase of the pandemic, chest x-rays emerged as a key tool3 for diagnosing COVID-19, and radiology staff experienced a surge in examination volume. Diagnostic imaging modalities employed for direct COVID-19 patient management and computed tomography (CT) were perceived to have been under increased procedural pressure while other elective or non-urgent diagnostic and screening services were paused in some global settings.
Reports and experiences3 indicate that there has since been extensive re-organisation within diagnostic imaging and radiotherapy departments in response to the pandemic, a necessity due to changes in workload and working practice guidelines.
Before the pandemic, diagnostic services were already burdened with high demand, shortage of staffing, and inefficient workflows, leading to alarming rates of staff burnout. Now, in the face of COVID-19, the need for greater efficiency and responsiveness has never become more urgent.
Digital transformation in radiology is now having to shift into a higher gear in response to COVID-19, offering new opportunities - identifying needs and delivering healthcare from prevention and health promotion to curative interventions and self-management.
AI in the Digitalisation of Radiology
Advances in medical imaging constantly increase our knowledge of diseases, their impact on patients and their treatments, creating a surge in the amount of data generated.
Research4 shows that a typical hospital generates enough data per year to fill 20 million four-drawer filing cabinets, but 97 per cent of that data never gets used. This is where solutions such as Artificial Intelligence (AI) can play an increasingly vital part in transforming radiology.
Building on automation, AI has the ability to turn large amounts of data into insights that will support more precise diagnosis, targeted treatment, and greater patient satisfaction. Smart improvements from AI, including deep learning algorithms, can help radiologists make their diagnostic determinations faster and with a high degree of confidence due to the clarity and sharpness of the images. AI-powered imaging can also make up for the additional time needed between each patient examination to execute disinfection protocols.
Clinicians are now increasingly realising that AI could be a key tool in managing the demands of the modern-day radiologist. For example, in a prostate cancer study survey conducted by Sectra5 in three key markets, radiologists agreed that AI application benefitted daily workflow. The survey found that 68 per cent of radiologists agreed that “automatic characterisation (AI) and scoring lesions according to internationally accepted criteria would be valuable to me.”
Recognising these pressing needs, GE Healthcare has been developing AI tools that assist radiologists as well as departmental users in their workflow. In Malaysia, public hospitals are starting to pivot towards AI-enabled software for their MRI machines. Thailand and Vietnam have also introduced AI-enabled CT machines to help eliminate manual variations. Achieving a better-quality scan now no longer requires more time in the scanner or physical manoeuvring of the patient on the scanner.
The company is also partnering with academic medical institutions in more comprehensive ways to integrate AI into clinical practice1 – and imaging is just part of the focus.
AI has the potential to revolutionise health systems and help improve patient experience while alleviating the risk of physician burnout. Excessive manpower is freed up through the automation of repetitive tasks.
Challenges Over AI and Cybersecurity Threats
With digitalisation comes complexity, and hospitals need to ensure that technology issues do not get in the way of the medical staff delivering essential care. Two pressing issues are fears over the use of AI as well as potential cybersecurity threats.
As AI gradually integrates into radiology in several different ways, it is undoubtedly a hot topic today. AI has demonstrated that it can surpass human capabilities in biotechnology and imaging. The idea of non-human intelligence getting the better of humans may seem scary, but physicians and scientists increasingly see it as a valuable ally in their daily work.
In the long term, however, with AI-assisted image analysis improving by the day, radiology as we traditionally know it will cease to exist, but not disappear. Instead, it will be significantly different, and at the same time, more important than ever.
A recent commentary in Academic Radiology titled “Towards Augmented Radiologists”6 supports an optimistic view of imaging AI. The radiologist authors at Massachusetts General Hospital acknowledge there will be a short-term learning curve for radiologists – both residents and their supervising radiologists. But in the long-term, “artificially intelligent software assistants” will transform the teaching of radiology residents and offer new opportunities for attending radiologists in their practice as well.
In the future, more and more reliable and quality patient data from around the world will need to use AI in other fields of medicine; and teach it to recognise as many cases and patterns as possible. For instance, triage could be an area where smart technologies could have a major impact,7 but without the data, AI cannot unlock its potential.
Effective cybersecurity is also critical in helping to ensure the security of medical devices and healthcare networks, as well as patient privacy and health information.
However, as the world turned its attention to the healthcare urgencies created by COVID-19, there are now increased opportunities for cyber attackers targeting the healthcare industry. There is a growing risk that malicious threat actors could modify, or control connected devices and access patient data, creating risks for patient privacy or otherwise adversely impacting patient care.
Since 2019, the healthcare sector has seen a shift from breaches caused by internal actors to primarily external actors, according to Verizon’s most recent Data Breach Investigations report.8 Still, healthcare breaches increased 55.1 per cent in 2020 where the average cost per healthcare record breached increased from $429 in 2019 to $499 in 2020, costing healthcare organisations around $13.2 billion last year alone.
It is important that radiology administrators understand that cybersecurity is a shared responsibility. Modern healthcare infrastructure and cybersecurity protections are most effective when departments such as radiology, clinical engineering (biomed), and IT have an early seat at the table to collaborate with, educate, and inform the cybersecurity team about any potential vulnerabilities specific to connected medical devices within radiology.
Medical device manufacturers also need to be vigilant in understanding device risk factors and defining, developing and embedding risk-based security controls in these devices to help protect the security of hospitals and healthcare networks. GE Healthcare relies on a high level of collaboration with its customers as well as other manufacturers on this topic to elevate the industry and its contribution to enabling healthcare providers’ progress in improving patient outcomes.
The Future of Digitalisation
It is now time for radiologists to work smarter, not harder. Innovative technologies such as AI, robotics, and automation are crucial enablers for expanding precision medicine, transforming care delivery, and improving patient experience.
Addressing the challenges to patient care using more automated workflows and AI-powered imaging will allow radiologists to provide more precise and personalised care. AI tools in radiology are continually being developed that will supplement the radiologist’s expertise and enable them to be more involved in clinical information flow and patient care.
Collaborative discussions and participation in cybersecurity efforts by radiology administrators will also result in a better understanding of the cybersecurity infrastructure of the health facility, as well as an expanded understanding of the unique cybersecurity needs for medical devices and enterprise imaging platforms.
The next step in the digitalisation of radiology is machine learning: the ability for a computer to learn without being programmed. Machine learning identifies complex patterns automatically and helps radiologists make intelligent decisions on radiology data such as conventional radiographs, CT, MRI, and PET images and radiology reports. In many applications,9 the performance of machine learning-based automatic detection and diagnosis systems have shown to be comparable to that of a well-trained and experienced radiologist.
As radiology departments push towards taking advantage of digitalisation to image the body with automatic segmentation, active monitoring, and protocol management, they should keep in mind the main goal: improving clinical quality.10 This focus may be closer than ever to realisation, where radiologists, now empowered with digitalisation, will truly be able to make an image that is worth a life.
- GE Healthcare Systems. (2021). Developing AI to Optimize Radiology. Retrieved from https://www.gehealthcare.com/article/developing-ai-to-optimize-radiology/.
- Venkataraman, S. (2020). Digitization in Radiology: How Digital Tools are Converting Challenges into Opportunities. Medium. Retrieved from https://medium.com/carre4/digitization-in-radiology-how-digital-tools-are-converting-challenges-into-opportunities-8b59ca2e248b.
- Akudjedu, T. N., Mishio, N. A., Elshami, W., Culp, M. P., Lawal, O., Botwe, B. O., ... & Franklin, J. M. (2021). The global impact of the COVID-19 pandemic on clinical radiography practice: A systematic literature review and recommendations for future services planning. Radiography.
- Kellner, T. (2017). AI Healthcare Expert: Doctors And Machines Make A Brilliant Match. General Electric Company. Retrieved from https://www.ge.com/news/reports/ai-healthcare-expert-doctors-machines-make-brilliant-match?ga=2.175041122.170888882.
- Sectra. (2017). Where radiologists see the added value ofmachine learning. Retrieved from https://medical.sectra.com/resources/where-radiologists-see-the-added-value-of-machine-learning/
- Tajmir, S. H., & Alkasab, T. K. (2018). Toward augmented radiologists: changes in radiology education in the era of machine learning and artificial intelligence. Academic radiology, 25(6), 747-750.
- Ibrahim, S. (2021). Digitising Clinical Data: An uphill road for Switzerland. SWI swissinfo.ch. Retrieved from https://www.swissinfo.ch/eng/business/sanit%C3%A0_digitizing-clinical-data--an-uphill-road-for-switzerland/46363068.
- Verizon. (2021). 2021 Healthcare Data Breaches and Security. Retrieved from https://www.verizon.com/business/resources/reports/dbir/2021/data-breach-statistics-by-industry/healthcare-data-breaches-security/.
- Wang, S., & Summers, R. M. (2012). Machine learning and radiology. Medical image analysis, 16(5), 933-951.
- Artificial intelligence: shaping the future of healthcare. (2018). In GE Signa Pulse. Retrieved from https://www.gesignapulse.com/signapulse/autumn_2018/MobilePagedReplica.action
About the Author
Christopher Sung Wook Khang is President & CEO of GE Healthcare ASEAN based in Singapore. He is primarily driving consistent and sustainable growth and establishing a strong organisational structure in 10 countries, ASEAN.
Since joining as President & CEO of GE Korea in January, 2012, Chris has driven sustainable growth by promoting GE’s global excellence in technology and innovation, engaging government and strategic private customers, and closely cooperating with the GE business teams in Healthcare, Aviation, Renewables and Power. In 2019, Chris took on direct responsibility for GE Healthcare as President & CEO and has driven strong business growth through the development of commercial strategy and focus on operational rigour.