Imaging + Genetics | Coronavirus Imaging Continued

“Medical AI systems are still too new to have been challenged in medical malpractice lawsuits, so it is unclear how courts will determine responsibility and what kind of transparency should be required.”

Scientific American’s Sara Reardon on AI’s eventual day in court.

We’re thrilled to announce that Riverain Technologies is now an Imaging Wire premium sponsor. Riverain Technologies is doing amazing work saving lives through the early detection, diagnosis, and management of lung disease, and we can’t wait to share their message with you.

The Imaging Wire is made possible by Riverain Technologies and the rest of our sponsors, so make sure to keep up with the great things they’re doing.

Imaging Wire Sponsors

  • Focused Ultrasound Foundation – Accelerating the development and adoption of focused ultrasound.
  • GE Healthcare – Providing point of care ultrasound systems, from pocket-sized to portable consoles, designed to support your clinical needs and grow along with your practice.
  • Healthcare Administrative Partners – Empowering radiology groups through expert revenue cycle management, clinical analytics, practice support, and specialized coding.
  • Nuance – AI and cloud-powered technology solutions to help radiologists stay focused, move quickly, and work smarter.
  • Qure.ai – Making healthcare more accessible by applying deep learning to radiology imaging .
  • Riverain Technologies – Offering artificial intelligence tools dedicated to the early, efficient detection of lung disease.

The Imaging Wire

Coronavirus Imaging Continued

Novel coronavirus was at the top of international and imaging news again this week, bringing new insights into how radiologists are diagnosing the deadly virus, and new efforts from industry players to support the related surge in diagnostic demand (here’s last week’s coverage too).

  • Coronavirus’ Imaging Features – The journal Radiology detailed the CT imaging features that radiologists should look for when detecting and diagnosing coronavirus. Using CT scans from 21 coronavirus patients, a Mount Sinai-led team found that the virus nearly always presented with bilateral ground-glass opacities and consolidative pulmonary opacities, while looking for nodular opacities, crazy-paving pattern, and a peripheral distribution of disease in CT scans can also help with early diagnosis.
  • Coronavirus AI – Lung-focused Chinese AI firm, Infervision, launched what is believed to be the first algorithm intended to help detect and monitor Coronavirus (or at least the first algorithm branded for Coronavirus). The Coronavirus AI solution is already in use in Chinese hospitals (including Wuhan’s Tongji Hospital and Shenzhen’s Third People’s Hospital of Shenzhen), helping offset the huge spike in image reading volume and serving as a way to preserve Coronavirus lab test kits.
  • United Imaging Pitches In – We heard a lot about the imaging OEMs doing their part to support the fight against novel coronavirus last week, but United Imaging went about 100 steps further, revealing plans to deliver over 100 CTs and X-ray machines across China. United Imaging donated over $1.4 million worth of imaging systems to Wuhan, China hospitals, while apparently selling many other systems across its home country.

Imaging + Genetics

The combination of MR imaging with genetic/cellular tumor analysis could lead to improved cancer treatment planning, particularly for brain cancer, which is hard to diagnose and then treat/remove.

  • The Study – A Translational Genomics Research Institute and GE Research team correlated the genetic and protein fingerprints of brain cancer cells with brain MRI scans to get a full picture of the “tumor microenvironment” (the cells in and around tumors). This was a complex study for those of us without an expertise in genomics, but the simplified version is: the study revealed that the spatial and cellular context that imaging provides makes genetic/cellular analysis of cancer cells more effective (and vice versa).
  • The Impact – This imaging and genetic/cellular combo could be used to help physicians decide how much cancer tissue to remove, how much radiation dosage to use, and what drugs might be most effective.

Share The Imaging Wire

Thank you all for supporting The Imaging Wire and Visage Imaging‘s referral program benefiting the New South Wales Rural Fire Service and WIRES Wildlife.

Please take a moment to consider your colleagues who might find this newsletter valuable and let them know about it. Once again, we’re donating $2 to these Australian Bushfires charities for each new subscriber that you refer.

If you’re a subscriber, please consider using your personal referral link to introduce your industry friends to The Imaging Wire (You can find your personal link in your latest newsletter).

If you haven’t received a referral from one of your colleagues yet, just use this link to sign up.

The Wire

  • A study in Academic Radiology found that the top 20 U.S. hospitals’ imaging chargemasters fall short of the cost transparency they were intended to provide. Although all of the hospitals had publicly available chargemasters that notified patients that there would be differences in out-of-pocket costs, only 20% detailed what those out-of-pocket costs would be or provided tools to estimate those costs, and just 10% included CPT codes (limiting their ability to compare hospitals).
  • Qynapse announced the FDA clearance of its QyScore imaging software, used to help diagnose and measure neurodegenerative diseases like Alzheimer’s, Parkinson’s, and MS (it gained CE clearance in 2017). QyScore automatically quantifies a range of brain MRI markers and produces patient reports that can be compared to a database of healthy subjects.
  • A Canadian team developed a machine learning algorithm that can detect infarction in patients with acute ischemic stroke (AIS) using non-enhanced CT images with similar precision as diffusion-weighted MRI. The study used Non–contrast CT images from 257 AIS patients who also underwent diffusion-weighted MRI, training the algorithm using imaging data from 157 of the patients (CT images and DW MRI lesion labels) and using the 100 remaining patients to test the algorithm. Testing revealed that the lesion volume detected by the model correlated with expert-contoured lesion volume in acute DW MRI scans, with an 11 mL mean difference.
  • An orthopedic surgeon filed a whistleblower lawsuit against Orlando Health, alleging that he was fired for performing surgeries at outside facilities and making referrals to outside imaging centers. The surgeon argued that Orlando Health increasingly asked him to keep his procedures and referrals in-network before he was eventually fired in 2018 for refusing these requests, violating the Stark Law and the federal Anti-Kickback Statute.
  • Payers took a public stand against the new federal healthcare cost transparency rule, complaining that it would cost them 26-times more than originally estimated ($13.63m per insurer, mainly due to setup and maintenance). Even the AHA sided with them (to an extent), as the hospital lobby disagreed with forcing the release of negotiated rates between payers and providers.
  • Portugal became the first country to launch a national telehealth plan, combining and expanding upon the most successful parts of its regional telehealth systems to create a “articulated and synergistic system, which is more than the sum of its parts.” Although the system expands far beyond imaging, it does support teleconsultations between healthcare providers for sharing imaging and lab results (among other applications).
  • Independent radiology practice coalition, Strategic Radiology (SR), announced a partnership with Radius to provide private radiology cloud services that SR member groups can use to exchange studies for quality-improvement initiatives and cross-practice service coverage. Radius will be part of the new SR Connect platform (joining Mach7), linking each of the member groups through a Radius-hosted enterprise platform, and allowing aggregation and routing of HL7 and DICOM information between the members and other affiliated organizations.
  • IBM Research and the CHDI Foundation used functional MRI and AI algorithms to map the progression of Huntington’s disease (HD) and demonstrate how early HD progression diagnosis could eventually be performed using a single fMRI brain scan. Leveraging data from previous HD research, the team created an AI system that used fMRI brain activity measurements to estimate a person’s brain structure network and identify whether the patient is stable or showing cognitive decline.
  • Saskatchewan witnessed its MRI waitlist grow from 5,055 people in 2015 to 10,018 in 2019, prompting some to call for the Canadian province to increase its MRI capacity and do more to ensure appropriate referrals. Saskatchewan has been trying to address this issue for years, including expanding MRI hours in some regions and allowing private MRI clinics for the first time in 2016, but these efforts don’t appear to be improving the situation.
  • A recent study from Altarum found that “low-value” healthcare services cost commercial insurers $5.5 billion in 2015 alone, including ~$193 million in low-value imaging procedures. Lower back pain imaging was at the top of the “low-value imaging” list ($170m), followed by CT for rhinosinusitis diagnosis ($12m), CAC scoring for known CAD ($8m), repeat DXA scans ($2m), and MRI for arthritis monitoring (<$1m).
  • European Radiology detailed a new AI model that combined B-mode and color Doppler ultrasound breast scans to accurately classify BI-RADS levels based on the probability of malignancy, suggesting that it could automate this process in the future. The researchers trained a pair of CNNs (one B-mode, one B-mode & color Doppler) on images of 103,212 masses (plus a 2,748 held-out set) that were originally interpreted by 20 experienced radiologists, and then tested it against a set of 605 biopsy-proven masses. The dual model (B-mode & color Doppler) categorized the masses as benign or malignant with an impressive 0.982 AUC, above the radiologists’ 0.948 AUC.

The Resource Wire

– This is sponsored content.

  • Riverain Technologies is dedicated to providing enterprise software tools to aid clinicians in the efficient, effective, early detection of lung disease. Learn more.
  • This GE Healthcare white paper details how its suite of point of care ultrasound AI tools simplify complex patient assessments, enable faster clinical decisions, and calculate precise results.
  • Learn how and why Seattle Children’s Hospital, Duke University Health System, and HCA Healthcare chose to ditch the disk by adopting Nuance’s PowerShare Network.
  • The pressures driving radiology practice consolidation are significant, but there’s a strong case for staying independent. Healthcare Administrative Partners explains how independent practices can thrive by building relationships, relying on outside support, and leveraging capital.

You might also like

You might also like..

Select All

You're signed up!

It's great to have you as a reader. Check your inbox for a welcome email.

-- The Imaging Wire team

You're all set!