“These results set the stage for a new paradigm of screening . . .”
Eric A. Klein, MD on the multi-cancer early detection (MCED) blood test’s latest study results.
I’m very happy to announce the launch of The Imaging Wire Show, an ongoing video interview program featuring some of the biggest names and ideas in medical imaging.
The Imaging Wire show kicks off with Dave Wilson from Fujifilm Healthcare Americas, sharing insights from his diverse imaging career, how a Canadian guy can become a Manchester United and Miami Dolphins fan, and new details on Hitachi’s transition to Fujifilm (which becomes official today).
Check it out and stay tuned for more episodes in the coming weeks.
Imaging Wire Sponsors
Arterys | Bayer Radiology | Canon Medical Systems | GE Healthcare
Healthcare Administrative Partners | Hitachi Healthcare Americas
Novarad | Nuance | Riverain Technologies | Siemens Healthineers
United Imaging | Zebra Medical Vision
The Imaging Wire
Cancer Blood Test Validated
New research in the Annals of Oncology revealed that the multi-cancer early detection (MCED) blood test can accurately detect 50 types of cancer before symptoms occur, representing a major milestone in MCED’s potential path towards becoming a population scale cancer screening method.
About MCED – The methylation-based MCED test sequences cell-free DNA (cfDNA), using machine learning to detect cancer signals and predict cancer origin.
The Study – The researchers performed MCED tests on 4,077 participants, including 2,823 patients with cancer.
The Results – The MCED test detected cancer with 99.5% specificity and 51.5% sensitivity, but achieved much higher sensitivity with later-stage cancers (stage III = 77%; stage IV = 90.1%) and with the twelve deadliest cancers (stage I-III = 67.6% vs. 40.7% for all I-III cancers). The MCED tests also predicted cancer origin with 88.7% accuracy.
MCED’s Potential – Considering that there are only five established cancer screening tests (colonoscopy, mammography, Pap test, PSA, low-dose CT), a blood test like MCED could significantly improve and expand cancer screening.
Imaging’s Role – The study didn’t mention MCED’s imaging implications, although the authors were clear that MCED should serve as a supplement to the current screening tests. It’s also quite likely that positive MCED screening results would often lead to follow-up imaging.
Next Steps – There’s other major MCED studies taking place, which will help scientists understand what it will take for MCED to become an official screening method.
CVIS’ Cloud Advantages
This Diagnostic and Interventional Cardiology article details the unique advantages of cloud-based CVIS systems (off-property access, team collaboration), with insights from one Mississippi-based cardiologist on the benefits of Hitachi’s VidiStar CVIS.
Imaging Wire Q&A: Regulating Service with Bayer & MITA
In the latest Imaging Wire Q&A we sat down with Bayer Radiology’s Dennis Durmis and MITA’s Peter Weems to discuss the medical device service debate and how ongoing legislation and regulation efforts could impact patients, clinicians, and OEMs. Here are some of the big takeaways:
Servicing Origins – Service regulation became a greater issue for imaging OEMs after cases involving improper service from unregulated third parties grew more serious, while the recent expansion of some Right-to-Repair bills to include medical devices has made this issue even more urgent.
How Serious – Durmis and Weems discussed cases where third-party servicers bypassed X-ray system radiation controls, repaired devices using twist ties, and used non-validated parts found at hardware stores.
The Current Rules – The FDA currently only regulates medical device servicing when it’s performed by the manufacturer, but not third parties.
A Regulated Proposal – Durmis and Weems (and presumably most OEMs) propose that all medical device servicers are required to register with the FDA and follow the same quality, safety, and regulatory requirements as manufacturers.
Next Up – MITA and the medical device OEMs continue to engage Congress regarding unregulated third-party servicing, while some state-level bills could expand third parties’ access to proprietary technical materials.
Canon and UCD’s Ultra High Resolution CT Experience
This presentation from Dr. Brian Goldner, MD details UC Davis Sacramento’s experience with Canon’s Ultra High Resolution CT and how it can be applied to cardiothoracic interpretations.
- Drone Ultrasound: A new Alberta Health paper highlighted drone-delivered teleultrasound’s medical potential, suggesting that drones are already “poised to revolutionize healthcare” (specifically emergency and remote care) and would benefit from ultrasound’s portability, real-time functionality, and remote guidance. To support this theory, the researchers used a drone to deliver a Philips Lumify ultrasound to an untrained volunteer 100km away who was able to perform a full ultrasound self-assessment using Zoom for guidance.
- Prostate MRI’s Growth Disparities: A new AJR study revealed notable disparities in prostate MRI’s rapid growth. The study (n = 82k men undergoing prostate biopsy) found that overall pre-biopsy MRI utilization rates increased significantly between 2008 and 2015 (0.5% to 9.2%), although black men were less than half as likely to undergo a pre-biopsy MRI than white men and men from the Western US were one-third as likely to receive a pre-biopsy MRI than in the Northeast.
- Synapsica’s Series A: Indian AI startup, Synapsica Healthcare, announced a $4.2m Series A round that it will use to develop its radiology workflow automation platform and support its global expansion. The Synapsica Radiolens platform is intended to simplify radiology report creation (removes repetitive / mundane tasks) and improve report quality (adds illustrations, evidence of disease, etc.).
- NM Dosage Stays High: U.S. radiologic technologists’ occupational dosage exposure consistently declined for general exams between 1980 and 2015 (0.60 mSv to undetectable), but remained high among techs who regularly performed nuclear imaging exams (especially cardiac and PET exams). That’s from a new Radiology Journal study (n = 58,434 technologists) that also found far higher median annual radiation dosage among technologists who regularly performed nuclear medicine exams than those who didn’t (1.2 mSv vs. 0.01 mSv).
- GE Healthcare’s New Leader: GE Healthcare announced that Integra LifeSciences’ Peter Arduini will replace Kieran Murphy as its president and CEO in 2022. Murphy will continue to lead GE Healthcare through the end of 2021 and serve as a strategic advisor during the transition. GE heralded Arduini’s accomplishments as Integra’s president and CEO and during his previous 15-year career at GE Healthcare.
- CCTA DL, Efficient & Effective: A new EJR study detailed a coronary CT angiography deep learning model that could help physicians detect obstructive coronary artery disease more efficiently and effectively. When used autonomously, the algorithm processed 165 patients’ CCTA exams in just 2.3 minutes (69% to 76% faster than rads without AI) and detected obstructive CAD with a 0.90 AUC (higher than less-experienced rads without AI). When radiologists used the DL model, they processed the exams in just 3.7 minutes each (vs. 8.5 minutes for rads without AI), while diagnosing obstructive CAD more accurately than radiologists who weren’t using AI.
- CDI Becomes RAYUS: Center for Diagnostic Imaging (>400 rads, imaging centers in 22 states) announced its relaunch as RAYUS Radiology, calling it a major step in a growth plan that will focus on acquisitions, opening new imaging centers, and employing more radiologists. CDI/RAYUS began its new growth plan with its acquisition of InHealth Imaging (3 centers in WA’s Puget Sound region), although CDI has been an active acquirer since taking on PE funding in early 2019.
- Rethinking AI Ground Truth: A new BIR paper argued that AI validation studies shouldn’t rely on individual radiologists’ interpretations as the ground truth (prone to inconsistent reads, low reproducibility), instead suggesting that these studies should use multiple readers’ collaborative interpretations. To prove this, the researchers had 6 radiologists individually review 1,100 CXRs and then had two reader panels produce collaborative interpretations, achieving far higher inter-reader agreement rates and greater reproducibility with the collaborative interpretations.
- NHS CXR AI Lung Cancer Screening Pilot: The UK NHS is trialing a Behold-based lung cancer screening system that analyzes chest X-ray for signs of lung cancer. The system could reportedly cut NHS radiologists’ missed lung cancer cases by 60% (from 38,500 to 15,750 annually), while also potentially reducing patients’ time to diagnosis (refers patients to follow-up CT in <2 hours vs. 14-24 days).
- Ultrasound Scoliosis Pre-Screening: A new UMB Journal study found that ultrasound-based pediatric scoliosis screening can effectively support X-ray referral decisions and reduce unnecessary radiation exposure. The study screened 442 children with suspected scoliosis using ultrasound and ultrasound combined with body measurements, predicting scoliosis with 0.735 and 0.832 respective AUCs. Based on this performance, the authors suggest that adding an ultrasound step to scoliosis screening workflows could cut unnecessary follow-up X-ray exams by 50%.
- LG’s AI X-Ray Package: LG Electronics unveiled a digital X-ray detector package that combines its new TFT-based detector with a companion AI device featuring Vuno’s chest X-ray AI tool (flags CXR scans w/ abnormalities). LG plans to launch the package globally, although it didn’t provide details about how it would support AI integrations into X-ray devices or radiology workflows.
- Predicting CT Delays: UCSF researchers developed a ML model that can predict CT interpretation delays and identify factors that contribute to delays. Using data from 15,117 non-business hour CT cases (85% training, 15% testing), the model predicted which scans would require more than >245 minutes in “total time” with an 0.85 AUC and predicted “interpretation times” longer than 57 minutes with a 0.71 AUC. Time of day, CT scan descriptions (e.g. “CTA chest pulmonary embolism protocol”), and resident training year were the greatest delay factors.
Einstein & Bayer’s Injection System Upgrade
See how Einstein Healthcare Network reduced its syringe expenses, enhanced its syringe loading, and improved its contrast documentation when it upgraded to Bayer Radiology’s MEDRAD Stellant FLEX CT Injection System.
The Resource Wire
- Cardiovascular disease is the number one global cause of death, but it’s also preventable, which is one of the reasons Zebra-Med views AI-powered cardiovascular screening as the next frontier in population health.
- It says a lot when a solution works so well for a radiology department that they decide to perform a study to quantify its benefits. In this Imaging Wire Q&A, University Hospital of Zurich’s Thomas Frauenfelder discusses his experience and study on Riverain Technologies ClearRead CT.
- Learn how GE Healthcare’s growing suite of homegrown and partner AI tools can boost radiology efficiency and effectiveness in its latest post.
- This prospective study out of France’s Centre Hospitalier de Valenciennes revealed that Arterys’ MSK AI module improved junior and senior radiologists’ fracture interpretations, while suggesting that the tool might be particularly useful for ED physicians that handle frontline X-ray reads.
- United Imaging is going “all-in” this summer in Nashville. The first stop is the HealthTrust University Conference starting July 26th, where they will showcase four modality portfolios and their unique ways of helping healthcare providers achieve investment transparency and reduce total cost of ownership. But what does “all-in” mean? Stay tuned as United Imaging reveals more over the next few weeks.
- Know how your practice measures up? In this post, Healthcare Administrative Partners details the key benchmarking quality metrics and how they can help radiology practices improve.
- Concerns about being sued were recently found to be radiologists’ top workplace challenges, but it doesn’t have to be that way. This Nuance blog details how adopting the right mix of structured reporting, clinical decision support, and follow-up tracking can help radiologists improve patient outcomes and reduce medico-legal exposure.
- See how Novarad’s CryptoChart solution allowed Central Ohio Primary Care (COPC, 70 practices, 400 physicians) to make the transition to digital imaging sharing in this Healthcare IT News case study.
- Make quick and well-informed decisions by having a clear overview of your clinical and operational performance data with Siemens Healthineers’ teamplay.