Wilhelm Conrad Roentgen discovered X-ray technology more than a century ago, pioneering the role of medical imaging in diagnosis, treatment, and monitoring of various clinical conditions. While the X-ray remains relevant today, over the years many advanced diagnostic procedures such as Computed Tomography (CT), Mammography, Ultrasound, Magnetic Resonance Imaging (MRI), Positron Emission Tomography (PET), and Bone Mineral Densitometry (BMD) have been added to the list.
The reports they create provide vital input for a clinician as they set the course for further diagnosis, treatment, prognosis as well as follow-ups. Thus, the accuracy and timing of a diagnostic imaging report is critical in a patient’s clinical management and cure.
Here’s a look at how Intel (INTC) is working with medical device manufacturers to apply advanced technologies to innovate their product basket and enhance patient care.
While the field of radiology has made significant advancements, a radiologist still needs to visually assess medical images for the detection and characterization of diseases. This, coupled with more health awareness, better accessibility to radiology tech, a huge emphasis and reliance on diagnostic reports -- all have generated massive amounts of data and layers of complexity. This has resulted in overworked radiologists.
According to the American Society of Radiologic Technologists, healthcare organizations generate close to 600 million diagnostic imaging procedures annually. Broadly, an estimated 90% of all healthcare data comes from imaging technology, yet 97% of it goes unanalyzed or unused.
This is where artificial intelligence (AI) and machine learning have the potential to transform health care by deriving crucial insights from the vast amount of data generated each day at clinics and hospitals.
To enable radiologists to read images more quickly and to lower the total cost of ownership for imaging devices by up to 25%, Intel and GE Healthcare announced a first-of-its-kind digital development lab for healthcare imaging technology. The Joint Performance Acceleration Lab (JPAL) has been formed “to create solutions that will offer greater hospital efficiency through increased asset performance, reduced patient risk and dosage exposure—with faster image processing—and expedited time to diagnosis and treatment.”
Using Intel AI technologies, GE Healthcare optimized algorithms for its Critical Care Suite to scan X-ray images and detect pneumothorax within seconds, speeding up detection by 3.3x.
Each year, approximately 74,000 Americans get impacted by a collapsed lung, known as a pneumothorax, which can turn deadly if not diagnosed quickly and accurately. Thus, a quick diagnosis is vital for patients.
AI techniques such as object detection and segmentation can assist radiologists in faster and more accurate identification. Intel and Philips have teamed up to investigate how servers powered by Intel processors could be used to efficiently perform deep learning inference on patients' X-rays and CT scans.
They tested two healthcare imaging proof of concepts; the results showed that Philips was able to achieve a speedup of 188x for their bone-age-prediction model and a 37x speedup for their lung-segmentation model over the baseline measurements using the Intel Distribution of OpenVINO toolkit and other software optimizations.
Intel and Siemens are working to provide real-time cardiovascular disease diagnosis. Using second generation Intel Xeon Scalable processors for AI inference, Intel and Siemens Healthineers demonstrated the ability to deliver MRI inferencing results in real time.
Using an AI model of the heart potentially saves time for cardiologists because “they do not have to manually segment different ventricles, myocardium, and blood pool cavities.” Cardiovascular disease (CVD) is among the leading cause of death in the U.S. It is estimated that approximately every 40 seconds, an American will have a myocardial infarction while the annual total cost of CVD assessed at $351.2 billion.
In April this year, the FDA published a discussion paper that describes the foundation for a potential approach to premarket review for artificial intelligence and machine learning driven software modifications. The agency is “considering a total product lifecycle based regulatory framework for these technologies that would allow for modifications to be made from real-world learning and adaptation, while still ensuring that the safety and effectiveness of the software as a medical device is maintained.”
A 2018 study that surveyed 200 U.S. healthcare decision-makers in April 2018 on their attitudes about AI highlights that 83% of the respondents said it will improve accuracy of medical diagnoses. In terms of revenue, the global AI radiology market is expected to reach $3.5 billion by 2027 owing to advancement in algorithms for better image recognition, characterization and monitoring of disease.
The broader market for AI in the healthcare is expected reach $27.6 billion by 2025, and Intel’s initiatives aren’t limited to medical imaging. It is working on precision medicine, drug discovery, reducing patient anxiety through virtual reality, patient data and management, home-based monitoring and predictive analytics.
The author has no position in any stocks mentioned. Investors should consider the above information not as a de facto recommendation, but as an idea for further consideration.
The views and opinions expressed herein are the views and opinions of the author and do not necessarily reflect those of Nasdaq, Inc.
- How to Invest In Edge Computing: Why Exploding Data Demand And Creation is Driving This Trend
- How Nasdaq is Advancing its Mission to Safeguard the Financial System and Its Participants
- Why Is Cleveland-Cliffs (CLF) Up 11.6% Since Last Earnings Report?
- What Data Analytics Will Look Like in 2021 - And How to Capitalize On It