COVID-19 Imaging: New Discoveries, More Support

“We don’t want to do it. We don’t think we should have to do it. If we have to do this, then we want to do it in a fair and rational and thoughtful way.”

Cassie Sauer of the Washington State Hospital Association on creating guidelines for deciding who lives and dies if coronavirus overwhelms the state’s healthcare system.


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

COVID-19 Imaging: New Discoveries, More Support

COVID-19’s expansion across the U.S. and Europe continued, filling more hospital beds and justifiably dominating the headlines. Things seem bad, but these latest headlines came with some hope, as we learn more about the tools we have to fight COVID-19 and how manufacturers are ramping up their support for providers on the coronavirus frontlines.

  • COVID-19 Imaging ProgressionA new longitudinal study out of China (n = 90 patients, 366 CTs) detailed how the imaging characteristics of patients with COVID-19 pneumonia change as the disease progresses. The study found that CT scores and the number of zones involved progressed rapidly, peaking during days 6-11 of the illness, and remaining high over the following days. Ground-glass opacity was the most prominent CT abnormality pattern during the early days of the disease (45% to 62%), while mixed pattern abnormalities were most common during days 12-17 (38%), and 94% patients still had signs of the disease in their CT scans even after they were discharged.
  • Nuance Adds-On – Nuance is providing its healthcare customers with free add-on licenses to help them scale up during the COVID-19 emergency. The new initiative includes 90-day add-ons to its PowerMic Mobile, Dragon Medical One, Dragon Medical for Epic Haiku and Canto, PowerScribe Mobile Radiologist, and PowerScribe for remote reading.
  • CV19 CT AI – A team of Chinese clinicians and AI experts detailed an algorithm that can help diagnose COVID-19, particularly in cases where tests are in short supply or faster diagnosis is necessary. The proof of concept deep learning tool used 1,119 CT images from patients with COVID-19 and typical viral pneumonia, achieving total accuracy of 89.5% with specificity of 0.88 and sensitivity of 0.87 against an internal set (external set: 79.3%, 0.83, 0.67).
  • AI Pneumonia Differentiator – A new deep learning model detailed in the journal Radiology was able to detect COVID-19 and differentiate it from typical pneumonia and other lung diseases using volumetric chest CTs. The algorithm (n = 4,356 CT exams, 3,322 patients) was able to identify COVID-19 with a 0.96 AUC and spot community acquired pneumonia with a 0.95 AUC, suggesting that it may help overcome concerns about CT-based COVID-19 misdiagnosis.
  • Telemedicine Opportunity – Healthcare Administrative Partners detailed how relaxed telemedicine rules during the COVID-19 Public Health Emergency could create an opportunity for radiologists to use telemedicine in their practice. This is particularly true for interventional radiologists and radiation oncologists, who can at least temporarily be reimbursed for patient interactions that might have otherwise gone unbilled (especially if they follow HAP’s guidelines).
  • More US Evidence A new study out of China added more evidence supporting lung ultrasound’s utility for COVID-19 assessment and management, due to its safety, repeatability, absence of radiation, low cost, and use at the point of care (here’s a previous Italian study). The researchers examined 20 patients with COVID-19 using a 12-zone method, finding the following ultrasound characteristics: 1. Thickening of the pleural line with pleural line irregularity; 2. B lines in a variety of patterns; 3. Consolidations in a variety of patterns; 4. A lines during the recovery phase; and 5. Pleural effusions are uncommon.
  • GE & Philips Increase Capacity – GE and Philips announced increased efforts to help in the fight against COVID-19. GE increased its manufacturing capacity for imaging systems, monitors, and ventilators (now producing ventilators around-the-clock) and expanded its workforce and supply chain to support these moves. Philips’ manufacturing capacity in China is back to 80% and the company is working on increasing its global production of imaging systems, patient monitors, and ventilators.
  • COVIR-19 and E/M – The ACR and other groups took another shot at the recent changes to Evaluation and Management (E/M) reimbursements that could cost radiologists $5.6b over 10 years, urging congress to waive budget neutrality requirements as they craft COVID-19 relief packages. With the current structure, the E/M changes create $5b in additional provider reimbursements, but this increase will largely go to other specialties and would therefore require cuts in other areas due to CMS’ budget-neutral mandates. Although this may not feel relevant to COVID-19, the ACR suggests that imaging/radiology has been negatively impacted by the crisis and therefore can’t handle these E/M changes.

The Wire

  • Korean Collusion: Siemens Korea and Canon Medical Systems Korea were fined a combined 54 million won ($43,029) for rigging a $1.23m CT scanner contract that Siemens Korea won with a local hospital. Reports allege that Siemens was concerned that the deal would be cancelled if Canon dropped out, so Siemens “supported” Canon’s participation.
  • Presurgical Breast MRI Benefits: New research in Radiology highlighted the benefits of presurgical breast MRI, finding that it led to more successful surgical outcomes without affecting mastectomy rates. The study reviewed outcomes from 420 women with DCIS who underwent breast MRI, finding that it revealed 67 additional lesions (16%), led to 57 changes in surgical plans (13%, 31 changes were appropriate), and was associated with lower risk of positive resection margin (0.39 odds ratio) and repeat surgery (0.33 OR).
  • Auto Imaging Scheme: A New Jersey man pleaded guilty for his role in a four-year healthcare fraud scheme that involved convincing people who were involved in accidents to fabricate/exaggerate their injuries and sending them to participating medical imaging centers and chiropractors who billed their insurers for medically unnecessary services. The imaging centers and chiropractors paid the man $500 for every referral.
  • No Love from MedPAC: The Medicare Payment Advisory Commission (MedPAC) suggested that congress shouldn’t increase payments to radiologists and other specialists in 2021. MedPAC pointed to the 18.6% increase in median physician compensation between 2014 and 2018, but suggested that these increases have disproportionately gone to certain specialties (like radiology), while PCPs missed out.
  • Canon AiCE for MR: Canon announced the FDA approval of its Advanced intelligent Clear-IQ Engine (AiCE) image reconstruction technology for use with its Vantage Galan 3T MR system, expanding AiCE to Canon’s MR portfolio for the first time. Canon has actively expanded AiCE throughout its CT portfolio over the last two years and the technology’s MRI expansion represents a key milestone.
  • AI Therapy Predictor: Columbia University researchers developed a CT-based algorithm to predict patient response to three systemic cancer therapies and help radiologists predict treatment efficacy at early stages. The algorithm was developed using CT scans from patients with advanced non-small cell lung cancer (NSCLC) and tested with CT images from clinical trial patients receiving nivolumab, docetaxel, gefitinib (n = 92, 50, 46). By comparing baseline and first treatment CTs, the algorithm was able to label each patient as treatment-sensitive or treatment-insensitive with respective AUCs of 0.77, 0.67, and 0.82.
  • Siemens’ Pakistan Partnership: Siemens Healthcare and Pakistan’s National University of Medical Sciences (NUMS) are partnering with a pair of other German companies to expand modern medical facilities and practices across the country. The alliance will upgrade existing hospitals and create new facilities, with Siemens serving as the initiative’s official medical technology partner.
  • New Brain Cancer Agent: German researchers developed a new xenon-based MRI agent capable of detecting some cancers before they metastasize in the brain. The team used a synthetic molecule, temporarily filled with xenon, to detect the formation of new blood vessels and visualize tumor-forming cells, finding that it’s 16,000-times more efficient than some fluorine-based alternatives and uses far lower agent concentrations.
  • Cerveau & Alector’s PET Partnership: Cerveau Technologies will license its investigational PET brain imaging agent (MK-6240) to Alector, allowing Alector to use MK-6240 as an exploratory biomarker for its research studies. Cerveau is creating a network of research partners to help develop and verify MK-6240’s use in assessing the status and progression of neurofibrillary tangles (NFTs) in the brain, which could be used for diagnosing and monitoring neurodegenerative diseases.
  • ACR’s AI Comments: The ACR weighed in on the federal Office of Management and Budget’s (OMB) plan for artificial intelligence oversight, generally agreeing with the OMB’s 10 principals, while making a number of comments intended to support and protect radiologists. The response is quite detailed, although most of the comments focused on ensuring that the unique needs of healthcare are represented, encouraging collaboration between the government and healthcare stakeholders, and requiring data quality and algorithm transparency.
  • AI Osteoarthritis Predictor: New research in Radiology AI detailed an algorithm that can predict knee osteoarthritis severity in X-rays as well as musculoskeletal radiologists. The CNN model (trained w/ 32,116 images, tuned w/ 4,074, tested w/ 4,090) achieved a 0.70 F1 score and 0.71 accuracy score compared to radiologists’ osteoarthritis assessments.

The Resource Wire

This is sponsored content.

  • In this quick video, Einstein Healthcare Network’s Peter Wang, MD describes how Nuance’s embedded clinical guidance delivers streamlined radiology workflow and improves report quality.
  • The 2020 CPT code updates brought a number of changes to diagnostic radiology, interventional radiology, and nuclear medicine. This Healthcare Administrative Partners blog details the 2020 MPFS CPT code changes and what they mean to your practice.
  • Vessel suppression from Riverain Technologies’ ClearRead CT software was recently found to significantly improve nodule detection, interreader agreement, and reading time with oncologic chest CT scans.

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.