“Radiologists remain ultimately responsible for patient care and will need to acquire new skills to do their best for patients in the new AI ecosystem,”
ACR Data Science’s J. Raymond Geis, MD on radiologists’ ethical responsibility to evolve as AI’s clinical role grows.
Imaging Wire Sponsors
- Carestream – Focused on delivering innovation that is life changing – for patients, customers, employees, communities and other stakeholders
- Focused Ultrasound Foundation – Accelerating the development and adoption of focused ultrasound
- Medmo – Helping underinsured Americans save on medical scans by connecting them to imaging providers with unfilled schedule time
- Nuance – AI and cloud-powered technology solutions to help radiologists stay focused, move quickly, and work smarter
- Pocus Systems – A new Point of Care Ultrasound startup, combining a team of POCUS veterans with next-generation genuine AI technology to disrupt the industry
- Qure.ai – Making healthcare more accessible by applying deep learning to radiology imaging
The Imaging Wire
Mapping Microbubbles
Scottish researchers unveiled a breakthrough ultrasound method that produces 10-times better images than current techniques and could even replace biopsies. The new approach overcomes ultrasound’s traditional limitations (fuzzy, greyscale images) and produces “near-microscopic quality” color images of blood flow. Here are some details to back this up:
Microbubbles – The team started with a focus on microbubbles, first establishing the physics of microbubbles before developing an AI algorithm to track their movement through the body. If the microbubbles travel through the body in unexpected ways, it could be a sign of cancer.
First Trial – The first human trials on the new ultrasound approach will start in late 2019 at Edinburgh’s Western General Hospital, initially focusing on prostate cancer patients since the prostate gland is typically removed and can then be compared to ultrasound images.
Going Forward – Although early cancer detection is a worthy target, the technique could be used across medicine given that most major diseases are associated with changes in blood flow.
Patient Navigation Works
A new study from RTI International and the CDC found that patient navigation (PN) services might be a cost-effective way to improve screening adherence and follow-up among women with abnormal screening mammography results and socioeconomic disadvantages.
The Study – The researchers used a breast cancer simulation model based on a cohort of 40-64yr old women within the National Breast and Cervical Cancer Early Detection Program (NBCCEDP) to measure the care outcomes and cost effectiveness of supportive services like PN.
PN Benefits – The researchers found that patients with PN support had more mammograms (4.23 vs. 4.14), lower lifetime breast cancer mortality (3.53% vs. 3.61%), and fewer missed diagnostic resolutions (0.017 vs. 0.025) than patients without PN.
QALY Effectiveness – Programs with PN support had a cost-effectiveness ratio of $32,531 per quality-adjusted life-years (QALY) compared to programs without a PN, well below healthcare’s standard $50k-$150k QALY threshold. Considering that a separate study recently found that DBT screening had a cost-effectiveness ratio between $195,026 and $270,135 per QALY compared to DM, the effectiveness of PN support is pretty notable.
The Wire
- Alphabet made another big healthcare leadership addition, hiring former FDA commissioner Robert M. Califf to lead health strategy and policy for the company’s Verily Life Sciences and Google Health divisions. Califf takes this role just a year after David Feinberg (former Geisinger / UCLA Health president & CEO) was hired to lead Google Health. The appointment also comes amid an effort to consolidate Alphabet’s healthcare operations, as Google officially absorbed Alphabet’s DeepMind Health division several weeks ago and Califf becomes the first executive to oversee both Verily and Google Health.
- Siemens Healthineers announced the FDA clearance of its AI-Rad Companion Chest CT software, which analyses CT images of the thorax, individually highlights the different structures of the chest, and creates structured exams. Siemens’ AI-Rad Companion Chest CT software gained CE clearance in late July and is the first solution from the AI-Rad Companion Platform to gain regulatory approval.
- Researchers at France’s CEA Saclay Nuclear Research Center are developing an 11.7T MRI that they suggest might lead to neurological breakthroughs and allow earlier diagnosis of neurodegenerative diseases. The “world’s most powerful MRI” uses a 16×16-foot, 286k-pound cylindrical super magnet that will allow 100-times sharper brain images than current systems. The super MRI’s first image isn’t expected until late 2020, and considering the slow rollout of 7T systems, clinical adoption is going to take a while.
- Bigger isn’t always better, though. NIH and Siemens Healthineers researchers developed a new lower-field MRI that “vastly improves” cardiac and lung imaging, and could also improve image-guided procedures, pacemaker/defibrillator imaging safety, patient cost and access, and installation and maintenance labor. This impressive list of advantages was achieved by modifying Siemens Healthineers’ 1.5T MAGNETOM Aera MRI to operate with a 0.55T field strength, while maintaining the system’s other advanced features.
- The U.S. ITC will review a judge’s findings that invalidated three Bracco patents related to its rubidium-82 generators, which if overturned could keep Jubilant from importing its Ruby cardiac PET imaging agent to the U.S. This is the latest in Bracco’s patent battle against Jubilant, which began with a formal complaint in March 2018 and continued over the last 1.5 years.
- The leading western radiology societies (ACR, ESR, RSNA, SIIM, EuSoMII, CAR, AAPM) released a united statement on imaging AI ethics, emphasizing that radiology AI should promote well-being, minimize harm, and fairly distribute benefits and risks. The statement also called for radiologists to evolve their AI-related knowledge and capabilities in order to “do their best for patients in the new AI ecosystem.”
- Mobile X-ray company, Trident USA Health Services, reached an $8.5 million settlement with the U.S. Department of Justice to resolve a kickback scheme involving skilled nursing facilities. Trident was accused of holding “swapping” arrangements with nursing facilities between 2006 and 2019, providing its Medicare/Medicaid-funded services at below-market rates in exchange for “lucrative” federal health program referrals. This is neither the first nor the last mobile imaging fraud story we’ll cover, as the segment’s vulnerable clientele and the fact that these scans can be done without physician authorization make it susceptible to schemes like this (here’s a previous example).
- An Intermountain Healthcare and Stanford team developed an AI algorithm that can accurately detect key findings in suspected pneumonia patients’ chest X-rays within 10 seconds (vs. up to 20 minutes or more). The team adapted Stanford’s CheXpert chest X-ray interpretation model with 6,973 additional images captured at Intermountain Healthcare’s EDs to better support its Utah-based population. They used the adapted CheXpert model on chest X-rays of 461 patients, categorizing the patients’ likelihood of having pneumonia with comparable accuracy as Intermountain’s radiologists.
- A team of Canadian researchers developed a new imaging technique called photoacoustic remote sensing microscopy that could “revolutionize surgical oncology” by improving clinicians’ ability to locate and measure the edges of tumors during surgery (vs. preoperative MRI or CT scans) and potentially eliminating follow-up surgeries.
- Imaging AI company Caption Health (previously known as Bay Labs) announced that its Caption Guidance ultrasound software (previously known as EchoGPS) received a breakthrough device designation from the FDA, following the completion of a successful study. Caption Guidance combines with ultrasound systems to allow clinicians with various skill levels to quickly and accurately perform ultrasound exams, leveraging the solution’s guidance, interpretation, and quality assessment capabilities. With the help of its new CEO’s star power (CEO Andy Page previously led 23andMe), this Caption Health announcement generated more buzz than any previous efforts from Bay Labs, suggesting that we’ll hear a lot more from them as Caption Guidance gets closer to launch.
- Imaging agent developer, Lantheus Medical Imaging (via its holding company), will acquire oncology medicine and AI company, Progenics Pharmaceuticals, for nearly $520m in stock. Lantheus and Progenics will have a more diverse diagnostic and therapeutics portfolio, while combining Lantheus’ manufacturing and operational strengths with Progenics’ pipeline and portfolio.
- The FDA issued a warning about 11 cybersecurity vulnerabilities, referred to as “URGENT/11,” that could allow hackers to remotely take control medical devices (imaging systems, infusion pumps, anesthesia machine, and maybe more) without detection. In addition to granting hackers control of the devices, the vulnerabilities could also be ‘wormable,’ meaning the could be used to spread malware across a network.
- The ACR took a stand against the proposed 2020 physician payment schedule, warning that the upcoming changes to Evaluation and Management (E/M) Services coding will have a “devastating impact” across radiology and other specialties. Although the E/M changes would actually result in $5b additional reimbursements, this increase would require cuts in other areas due to CMS’ budget-neutral mandates, particularly threatening radiologists who rarely bill E/M codes. ACR estimated that diagnostic and interventional radiologists would experience 8% and 6% cuts as a result of these changes.
The Resource Wire
- Qure.ai’s qXR chest X-ray tool and qER heat CT tool were highlighted in this MumbaiMirror article on medical AI.
- This Healthcare IT News article details how the University of Rochester Medical Center’s Backstop program leverages Nuance’s mPower Clinical Analytics and PowerScribe Follow-up Manager solutions to track and complete more follow-up imaging recommendations.
- The Focused Ultrasound Foundation’s 2019 State of the Field Report is now available, detailing FUSF’s initiatives and achievements and the state of the focused ultrasound market.
- This Carestream case study compares images of foot trauma captured using the OnSight 3D Extremity System to images captured on 2D X-rays.
- By partnering with Medmo, imaging centers can keep their schedules full and their equipment busy. Here’s where to learn more and get started.
