“It is really a failure of the system”
Brigham and Women’s Hospital emergency medicine physician Jeremy Faust on how the RT-PCR test shortage is driving high CT volumes.
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
Don’t Rush CV19 Science
There’s very real urgency to identify and adopt the best ways to diagnose and treat CV19. However, a new paper in the Annals of Internal Medicine argues that there’s been too much of a rush to publish positive CT study results, leading to “the dissemination of premature conclusions with broad implications.”
- The Results Rush – Due to the shortage of RT-PCR tests, many have looked to CT to take on a key role in CV19 diagnosis. However, some of the early studies that found CT to be highly accurate for CV19 diagnosis and fit to be a primary CV19 diagnosis tool did so within a questionably-fast 24hr peer review turnaround time and didn’t follow research best practices.
- Good Science Takes Time – While acknowledging that these are extraordinary times, the team reminded readers that good science “takes time and careful consideration to prove its value.” The paper recommended that CV19 imaging studies should go through a thorough review process to make sure best practices are followed and conclusions are fair and appropriate. They also encouraged CV19 imaging studies to address the risk of transmission during the imaging process and detail the preventative processes that imaging teams should follow during CV19 exams. Seems logical.
Researchers at the famous Los Alamos National Laboratory developed a self-powered X-ray detector that they say could revolutionize medical, security, and research imaging.
- The Detector – The new detector is made with 2-dimensional perovskite thin films that convert X-ray photons to electrical signals without an outside power source. The perovskite thin film’s high levels of heavy elements (e.g. lead & iodine) allows it to absorb/detect X-rays that generally pass-through silicon undetected.
- Perovskite Advantages – The new detector achieves 100-times greater sensitivity than conventional silicon-based detectors, dramatically reduces radiation exposure, is able to detect high-energy X-rays (important for research), and has much lower manufacturing costs than silicon detectors. Not a bad list of advantages.
- CV19 CT News Goes National: A recent Bloomberg.com article described the use of CT for CV19 diagnosis as high-cost, high-risk (due to infections), not advised (by the ACR), and often only performed because of the U.S.’s CV19 test shortage. Even if that might not be entirely fair, there’s truth to each of these accusations, and a chunk of Bloomberg’s millions of readers now know the downsides of the CT for CV19 story.
- At Risk Inaction: A team of Tampa-based researchers found that even though women with higher-risks of breast cancer are usually advised to undergo additional screening and other follow ups (breast MRI, genetic risk assessments, risk-reducing meds), most of these women actually understand or follow this advice. The study of 66 women who received a letter notifying them of their risk status found that 30% did not remember receiving the letter, 39% didn’t even know they were high risk, and just 15% acted upon the recommendations after learning they were high risk (6 received MRIs, 4 attended a clinic appointment).
- Integrating AI During CV19 Era: The Royal College of Radiologists released new standards on how to implement imaging AI during the CV19 pandemic, emphasizing the importance of efficiency during this healthcare emergency. Here they are: 1. AI must be integrated into RIS and PACS workflows in a way that doesn’t add burden to radiologists; 2. Algorithm accuracy must be clearly displayed to clinicians; 3. Findings must be communicated to the RIS and PACS using widely-used standards (HL7 and DICOM); 4. AI analysis must be complete and available on the PACS before human reporters begin interpretation.
- Approaching Neurorad Level: A new paper in Radiology detailed an AI system for brain MRI-based differential diagnoses that achieved accuracy “approaching” neuroradiologist levels. The AI system used a combination of data-driven and domain-expertise methodologies (1. Lesion detection with DL; 2. Imaging feature extraction; 3. Combination of imaging & clinical features), achieving 91% accuracy in its top-three differential diagnoses (vs. 86% by academic rads, 57% by general rads, and 77% by neuroradiology fellows) with a 92-patient test set.
- CV19 Business Warnings: Healthcare companies are starting to officially announce that CV19 is going to hurt their performance. GE recently withdrew its 2020 guidance and revealed that it will miss its Q1 expectations due to the impact of CV19 (across its businesses), Hologic warned that COVID-19 had “significant negative impact” on its Q2 performance, and mega-practice Envision Healthcare absorbed a downgrade from Moody’s to “Negative” and began exploring ways to restructure its liabilities.
- PACSgear Enterprise Launches: Products still have to launch during the CV19 era and one of last week’s biggest non-CV19 launches, was Hyland Healthcare’s PACSgear Enterprise software. The latest generation of Hyland’s PACSgear server software leads with its enhanced suite of application modules to support the capture, integration, and sharing of a wide range of point-of-care images (as well as a long list of other upgrades).
- NYU Rads Step Up: Auntminnie.com detailed how NYU Langone Health radiologists stepped up to support their colleagues in the emergency department, ICU, and hospital floor amid the surge in CV19 cases. The NYU radiologist team made the most of the extra time they had from the CV19-related drop in imaging, taking on responsibilities for communicating CV19 patient status to their families.
- Ferrum Lands $9M: Maker of radiology mistake-detection AI tech, Ferrum Health, closed a $9m seed round, revealing plans to use the funding to improve its platform, add to its engineering staff, and grow its sales and marketing capabilities.
- CV19 Imaging Consensus Statement: RSNA Radiology published Fleischner Society’s consensus statement on CV19 imaging. The statement suggested (among other things) that: 1. Imaging is indicated for CV19 patients with worsening respiratory status; 2. Imaging isn’t indicated for suspected CV19 patients with mild features unless they are at risk of disease progression; 3. Imaging can be used in a resource-constrained environment to triage suspected CV19 patients with moderate-severe clinical features and a high pretest probability of disease.
- Artisanal Radiology: A new perspective from Indiana University’s Richard B. Gunderman, MD, PhD called for radiology educators and students to consider an “artisanal” style of radiology, rather than the “mass production” model expected today. Gunderman referred to outlier artisanal bread and craft beer companies, who succeeded while the mainstream leaders in their industries declined, suggesting that radiologists could stand out and deliver more value through an approach that emphasizes quality, detail, case-based customization.
- Huawei Rolls Out CV19 AI: Huawei has been expanding its Huawei Cloud AI-assisted Diagnosis solution for CV19 CT analysis across Asia (co-developed with HY medical, trained on 4k+ CV19 patients’ data), including recent integrations in China, the Philippines, and Malaysia. Huawei is apparently partnering with government health ministries to offer the solution to designated hospitals in the countries without charge.
- Purdue’s Optical Advancement: Purdue University researchers developed a new optical imaging technique that can help surgeons localize tumors and help researchers/doctors understand how certain diseases affect neuron activation in the brain. The new technique involves applying near-infrared fluorescent contrast agents and measuring light absorption, reportedly overcoming fluorescence imaging’s historical challenge with light scattering.
- 1Qbit’s CV19 AI: 1QBit announced the launch of its xrAI Chest Radiography tool for the diagnosis of lung abnormalities, including COVID-19. The new radiologist-developed tool will soon go-live through the Saskatchewan Health Authority, specifically to support the CV19 fight, and will be available across Canada.
The Resource Wire
– This is sponsored content.
- This case study details how University of Rochester Medical Center reduced its delayed diagnoses risk by 80% after adopting Nuance PowerScribe Follow-Up Manager.
- Riverain Technologies is dedicated to providing enterprise software tools to aid clinicians in the efficient, effective, early detection of lung disease. Learn more.
- This Neurology Live article highlights how focused ultrasound thalamotomy delivers lasting benefits to patients with essential tremor.
- Watch an informational video from Jeff Hersh, M.D., Chief Medical Officer of GE Healthcare on using ultrasound in a health crisis.
- CMS has extended the deadline for reporting 2019 MIPS data & relaxed the criteria for avoiding a penalty in 2021. Learn more in this blog post from Healthcare Administrative Partners.
- This yourstory.com article details how Qure.ai repurposed its AI technology to triage and monitor COVID-19 cases.