A Laser Ultrasound First | Breast MRI Radiomics Breakthrough

“The only people who don’t want this fixed are the people who benefit from these excessive fees.”

Senate health committee Chairman Lamar Alexander on the folks opposing his surprise medical billing legislation.


Happy holidays everyone!

This is the final issue of the year, so keep an eye out for the next one on January 2nd. Huge thanks to all of our readers and sponsors who made this last year so special. Wishing all of you an excellent holiday season and a fantastic 2020.


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


A Laser Ultrasound First

An MIT team scanned human subjects with laser ultrasound technology for the first time, representing a milestone for the new imaging technique. Here are some details.

About Laser US – The technique leverages one laser to generate sound waves that bounce through a patient’s body and a second laser that remotely detects the reflected waves, which are translated into images.

First Image – The researchers scanned several volunteers’ forearms from ~1.5-feet away, observing common tissue features (muscle, fat, bone) down to about 6cm below the skin, and achieving image quality comparable to conventional ultrasound.

Laser Ultrasound’s Promise – Although traditional ultrasound works well for most patients, laser ultrasound could serve as a viable option for assessing patients who can’t/shouldn’t have a probe pressed against their skin (e.g. infants, burn victims, and accident survivors). Laser Ultrasound’s impact may also extend beyond these unique cases, as it avoids the image variability problem created by probe contact.

Up Next – Noting that “we’re at the beginning of what we could do with laser ultrasound,” the researchers plan to improve the system’s performance (specifically fine tissue features) and reduce its design down to “portable” sizes.



Breast MRI Radiomics Breakthrough

Penn Medicine researchers used MRI radiomics to predict 10-year risk of breast cancer recurrence, potentially improving individualized treatment. Here are some details.

The Study – The team set out to determine if imaging and radiomics could allow more personalized tumor characterization, extracting 60 MRI radiomic features/biomarkers from 95 women with primary invasive breast cancer. Using 10-year follow up status from these same women, the group generated “signal enhancement ratio” (SER) maps, extracting a range of imaging data to understand the relationship between imaging features, conventional biomarkers (e.g. gene mutations or hormone receptor status), and patient outcomes.

The Results – The algorithm was able to successfully predict recurrence-free survival after 10 years, finding that women with high tumor heterogeneity at the time of diagnosis (high diversity of cells) had a greater risk of cancer recurrence.

Biopsy Augmented – Although the team stopped short of saying that this method would replace biopsy, they did suggest it could augment the current “gold standard” by providing more details on a patient’s disease and guiding personalized treatment. That said, they also had some pointed criticism for biopsies, noting that “only taking out a little piece of a tissue from one part of a tumor” does not give a full understanding of a person’s disease and their potential therapy response.

Up Next – The researchers will next expand their analysis to a larger group of patients and explore how specific markers are more predictive of particular outcomes.


The Wire

  • Radiology Partners ramped its acquisition activity, adding Nashville-based Advanced Diagnostic Imaging (84 radiologists, 1.7m annual interpretations, 31 imaging centers, 5-states, a top-30 U.S. practice) and Radiology Associates of Appleton in Wisconsin (~20 radiologists), expanding RP to 24 states, over 1,500 radiologists, and nine top-30 practices. RP had a fast start to 2019 with five partnerships in the first half of the year (plus its Canopy acquisition), but these are its first practice partnerships since RP added $700 in capital in July.
  • Zebra Medical Vision announced a global co-development and commercialization agreement with DePuy Synthes (a Johnson & Johnson company) that will expand Zebra’s imaging AI solutions to orthopaedics as part of DePuy Synthes’ VELYS Digital Surgery solutions. Targeting the orthopedic specialty’s continued reliance on 2D images, the companies plan to leverage a Zebra AI solution that creates 3D models from 2D X-ray images and could “revolutionize” pre-surgical planning.
  • A Cedars-Sinai-led team developed a machine learning-based approach that combines clinical and imaging data to predict myocardial infarction (MI) and cardiac death better than current standard clinical risk assessment methods (0.82 vs. 0.77 AUC). The ML model quantified 1,912 asymptomatic subjects’ coronary artery calcium (CAC) and epicardial adipose tissue (EAT) data from cardiac CT scans, along with a range of clinical co-variates (e.g. plasma lipid panel, aortic calcium), finding that subjects with higher ML scores had a higher risk of events over a 14.5 ± 2 year follow-up period (10.38 vs. 2.94).
  • Medical image storage and access startup, Lyfebin, was in the news last week for exposing over 93,000 medical imaging files. TechCrunch identified the files in an unprotected Amazon Web Services (AWS) storage bucket and not the “secure environment” that Lyfebin promotes. Lyfebin assured that the AWS bucket was a test environment and the images did not reveal patient information (and threatened to sue TechCrunch), but TechCrunch countered that the largely-anonymized files did include some patient-linked info (physician names, patient gender, and date-of-birth).
  • Bayer announced the FDA clearance of its new MEDRAD Stellant FLEX CT Injection System with Certegra Workstation for Contrast Enhanced Mammography (CEM, 150 mL and 200 mL sizes), expanding the Stellant FLEX CT system’s clearance beyond CT applications. Bayer emphasized the new system’s dosing accuracy benefits over manually administered CEM injections and the clinical benefits of CEM versus standard mammography (tumor enhancement, strong performance with dense breasts), as well as the system’s efficiency features (easy visualization of fluids, syringe size barcode identification, lot and expiry date info, contrast barcode reader, streamlined training program).
  • Image quality issues forced a pair of Southern California and North Carolina imaging facilities to lose their mammography accreditations. LA-area Radiology Disc of Encino had its breast imaging accreditation revoked after poor audit results in April and August (August: 21 of 30 screening exams didn’t meet clinical image standards), while Raleigh Radiology Blue Ridge lost its mammography accreditation after an ACR review found insufficient image quality (they are filing for reinstatement).
  • Terason and AmCad BioMed announced a partnership to provide Terason’s ultrasound technology with AmCad BioMeds’ Obstructive Sleep Apnea (OSA) solution, AmCAD-UO. AmCAD-UO uses AI to analyze and reconstruct upper airway ultrasound images, helping physicians assess and diagnose obstructive sleep apnea. The partnership comes several months after Terason signed a similar AI deal to support DiA Imaging Analysis’ cardiac ultrasound analysis solutions with its point-of-care ultrasound devices.
  • A LocumTenens.com survey found that the average radiologist salary rose by 12% between 2015 and 2018 to $434,541 (third highest after orthopedic surgeons and cardiologists), while the average physician salary increased by 16.5% over the same period to $363,924.
  • The CMS’ proposed Oncology Care First Model has the ACR concerned that it is too “oncology centric” and doesn’t reflect the multi-disciplinary efforts and costs associated with cancer care, potentially incentivizing providers to underutilize imaging. The Oncology Care First Model would combine all cancer care payments in a single 6-month package (potentially including radiation therapy, imaging, and labs) that begins after chemo treatment is completed.
  • New research from Georgetown and Loyola University Medical Center found that integrating peer review into PACS resulted in a fivefold increase in “significant discrepancies” reported, while highlighting the role of system design in peer review effectiveness (e.g. PACS integration/simplification, randomization, and blinding). The researchers introduced a PACS tool that randomly prompts radiologists to review the interpretations of anonymous peers (replacing the RADPEER website), causing significant discrepancy reporting rates to increase from 0.19% ± 0.46% to 0.93% ± 1.45% per radiologist over four months (vs. no change in the control group).
  • A national audit from the Royal College of Radiologists revealed “significant deficiencies in care and imaging of severely injured patients” in the UK, calling for increased staffing, improved equipment, larger and modernized facilities, improved IT, and a commitment to improve the NHS after Brexit. Using data collection tools sent to UK radiology departments with an emergency department (n= 85 departments participated, 2,161 scans used), the audit found that just 240 scans (11%) had a documented primary survey report and 53 of those scans were unavailable to clinicians after 20 minutes, while the final time to consultant report took 1hr for 1,033 scans (48%) and up to 2hrs for an additional 540 scans (25%).
  • A new study provided insights into how American academic radiology departments handle outside studies. A survey of 91 SCARD members found that 70 (76.9%) of their departments added outside studies to their PACS, but only 34 (37.4%) required formal internal reports on all outside studies and 74 (81.3%) allowed “some” outside studies into their PACS without requiring an internal report. Most of the departments billed insurers (73.6%) but not patients (67%) for second opinion consultations. With minimal exceptions, results from a related AAARAD survey (n = 35) paralleled the SCARD results.

The Resource Wire

  • This Nuance video shares how Orlando’s Arnold Palmer Children’s Hospital used PowerShare to Ditch The Disk and improve patient care.
  • Did you know that imaging patients are most likely to no-show for their procedures on Mondays and Saturdays? By partnering with Medmo, imaging centers can keep their schedules full, despite the inevitable Monday no-shows.

Leave a Reply

Basic HTML is allowed. Your email address will not be published.

Subscribe to this comment feed via RSS

Get every issue of The Imaging Wire

Delivered right to your inbox. We will never sell or share your information.