Medical imaging data is one of the richest sources of information about a patient’s well-being. At the same time, interpretation of extremely high-resolution medical images is a complex task .
With a staggering amount of unstructured data packed into X-rays, CAT scans, MRIs, and other testing modalities, it is no surprise that Artificial Intelligence (AI) is slated to be a valuable ally for radiologists, pathologists, caregiver institutions, policy makers and public health governance.
The Daily Dose team explored the possibilities and promise of artificial intelligence in diagnostic imaging during a tête-à-tête with Sukhdeep Sachdev, Global Chief Executive Officer, Leader Healthcare Group.
What is the timeline to realize the promise of Artificial Intelligence (AI) in Diagnostic Imaging?
Arab Health 2020 will serve as Leader Group’s platform to unleash the power of Artificial Intelligence (AI) in Diagnostic Medical Imaging. Leader Group partners with Centre for Advanced Research in Imaging, Neurosciences & Genomics (CARING) to launch CARPL - an AI-enabled diagnostic and analytics platform to support faster, reliable and cost-effective diagnoses through medical imaging.
What is CARPL?
CARPL is an AI-assisted global platform for development, validation and deployment of advanced radiologic analytics. It is a data processing platform that enables the deployment of multiple, multi-vendor AI algorithms into the radiology workflow. It is secure, seamless and a feather-light component of the health IT infrastructure. CARPL is the metaphoric octopus - supporting data mining, analytics, validation of AI algorithms, and more. It can be applied to enhance patient experience, improve healthcare delivery metrics, to create global disease level databases for population health management, to enable precision medicine through genomics and integrated diagnostics.
How does CARPL support KPIs for healthcare delivery?
CARPL places the ease and power of AI into the hands of radiologists, caregivers and caregiving institutions – restoring confidence to patients who may have experienced a mis-diagnosis, providing quality assurance to insurance providers who may have reimbursed additional tests due to false positives, and relief to caregiver institutions who may have encountered litigation arising from discrepancies in radiologic diagnosis.
What can CARPL achieve at granular and policy levels?
CARPL can reduce the previously impossible task of parsing through millions of medical scans to the timeframe of a few days – thus overhauling the approach to public health and empowering policy makers towards the monitoring of disease-burden in real-time at a fraction of the cost.
Is AI-assisted diagnostic imaging a threat to radiologists?
The radiologist is an integral part of healthcare delivery. No tool or technology can replace the human component and associated critical thinking skills. The radiologist and AI-assisted diagnostic imaging tools join hands to bridge gaps and deliver a superlative care experience.
The advent of AI as a support tool for diagnosticians has been heralded as a positive development by the thought leaders at The American College of Radiology. In addition, the American College of Radiology Data Science Institute (ACR DSI) has released use cases that support the deployment of artificial intelligence in medical imaging. The objective is to develop standardized AI for clinical decision support and diagnostics.
Having said that, in the era of artificial intelligence, it is beneficial for radiologists to master basic programming tools to sift and organise DICOM data. User-friendly simple tools that can be used in clinical and research practice are widely available.
What are the use cases for AI in medical imaging, and how can AI tools alter workflows towards improved detection of fatal conditions?
Applying AI to imaging data may help to identify thickening of certain muscle structures or monitor changes in blood flow through the heart and associated arteries. AI tools could also be used to automate measurement tasks, such as pulmonary artery diameter, aortic valve analysis, carina angle measurement. AI could be used to identify hard-to-see fractures, soft tissue injuries and dislocations. Certain fractures are often difficult to detect on standard images. AI assisted tools may be more likely to identify subtle variations that indicate an instability requiring surgery. Unbiased algorithms could be deployed to review images in trauma patients - to ensure that all injuries are accounted for and deliver the care required for a superlative outcome. These are just a few use cases that scratch the surface of the possibilities. A few years ago, applied AI for healthcare seemed a distant dream. As we step into 2020, the dream has transformed into reality.