fbpx

Imaging in H1 | AI Code | Going Corporate


“It regularly informs your reporting in tangible ways rather than feeling like a boondoggle. Feels good.”

Part of a Twitter thread from Dr. Luke Oakden-Rayner, describing the benefits of comprehensive AI tools that can detect a wide range of findings.


Imaging Wire Sponsors

Arterys | Bayer Radiology | Canon Medical Systems
Fujifilm Healthcare Americas | GE Healthcare |
Healthcare Administrative Partners | Novarad | Nuance
Riverain Technologies | Siemens Healthineers
United Imaging | Zebra Medical Vision



The Imaging Wire


Imaging in H1 2021

The first half of 2021 is now a wrap, and it was another big one for medical imaging. Here are some of the top storylines from the last 6 months and some things to keep in mind as we head into the year’s second half:

  • Approaching Post-COVID – We’re not out of the woods yet (especially in poorer countries), but COVID infections and hospitalizations are way down and the pandemic is casting a much lighter shadow on many healthcare providers and vendors.
  • Security Takes Center Stage – Cybersecurity emerged as one of the world’s most pressing healthcare issues after ransomware attacks crippled operations across several major health systems. It’s safe to say that most medtech discussions will focus on security going forward.
  • AI Winter – After years of optimistic stories about imaging AI’s potential impact, 2021 brought a surge of stories about what’s wrong with AI. So far this year, we’ve seen repeated criticisms of AI’s generalizability, potential for bias, clinical impact, regulatory processes, and clinical adoption (among others). The good news is, we’re still very early in the AI ballgame and these critiques don’t take away from all the AI progress we’re seeing – even if the critical articles make for better headlines.
  • Population Pivot – Meanwhile, H1 2021 brought a major increase in population health AI activity, including major commercial launches from Zebra-Med and Cleerly, new initiatives from the NHS and the WHO, and an increased focus from academia.
  • Consolidation Continues – The radiology practice and imaging center consolidation trend continued, including some big moves from Akumin, continued activity from RP and CDI/RAYUS, and an influx of PE funding at SimonMed (that should lead to more consolidation).


COPC’s Case for Novarad

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.

– Sponsored.



The First AI Codes

The American Medical Association issued the first imaging AI CPT codes on July 1st, including Category III codes for incidental vertebral fracture detection and quantitative ultrasound characterization, allowing clinics to start billing for these applications in January 2022 and potentially leading to the first AI reimbursements (the new AI codes are on page 3).

  • The VCF AI Code – Highlighted among the AMA’s CPT update is a new code (0691T) for using AI to automatically analyze existing chest CTs to detect vertebral fractures, potentially paving the way for AI-based population health osteoporosis programs. This CPT code represents quite a milestone for AI (it’s the first radiology-specific AI code) and for Zebra Medical Vision, which is the only company with an FDA/CE-cleared product for population-scale incidental VCF detection.
  • Ultrasound AI Codes – The AMA also issued a pair of CPT codes for quantitative ultrasound tissue characterization, which can be used in conjunction or separately from existing ultrasound exam codes (0690T w/ other US exam codes, 0689T separately). These codes could also represent a major step for key ultrasound exams (e.g. breast, thyroid) and ultrasound AI decision support players (e.g. Koios Medical).
  • Societal Influence – These new codes might come as a surprise to some readers, but they are a direct result of the major radiology societies’ (ACR, RSNA, ARRS, AUR) work with the AMA CPT Editorial Panel, suggesting that future AI codes will follow this path.



Siemens’ Dual Source Cardiac CT

Explore untapped potential in dual source CT for cardiac imaging. This new Siemens Healthineers whitepaper showcases the clinical benefits of fast native temporal resolution.

– Sponsored.


The Wire

  • Corporate Practice Shift: By the end of 2020, 50% of U.S. physician practices were owned by hospital systems or corporate entities (e.g. PE groups) and 70% of U.S. physicians were employed by these entities. That’s from a new Physician Advocacy Institute study that attributed this corporate / hospital employment shift to a 32% increase in acquisitions in 2019/2020, and pressures created by corporatization and the COVID pandemic.
  • Annalise.ai’s Comprehensive CXR AI: A new Annalise.ai-funded study had 20 radiologists review 2,568 patients’ CXR cases with and without AI (3 months apart), revealing that the radiologists achieved far higher AUCs when they had AI support (0.713 vs. 0.808). The AI model also “significantly improved” the radiologists’ classification accuracy with 102 of 127 clinical findings, and didn’t decrease their accuracy with the other findings. As this Twitter thread details, the Annalise CXR model’s performance is even more notable given the clinical value of comprehensive diagnostic AI tools (127 findings!) and its robust training dataset (821k images w/ 127 labels).
  • Midstate’s CT Expansion: Connecticut’s Midstate Radiology Associates (18 locations, >40 rads) continued its regional expansion, acquiring local practice Radiology Group, P.C. (2 locations, 3 rads). Midstate has grown considerably in recent years (in 2015 it had just 2 locations, 11 rads), executing a number of similar acquisitions without PE support.
  • PSMA-11-PET Advantage: Dutch researchers found that PSMA-11-PET can identify prostate cancer patients who are suitable for stereotactic ablative radiotherapy (SABR) earlier than choline-PET, allowing them to avoid or delay androgen deprivation therapy (ADT). The researchers performed PSMA-11-PET and choline-PET on 50 patients with oligometastatic recurrent prostate cancer (40 w/ PSMA-11-PET), finding that the PSMA-11-PET group qualified for SABR at lower PSA levels (1.8 vs. 4.2 ng/mL median), while PSMA-11-PET-guided SABR resulted in longer PSA response durations (34 mo. vs. 14.7 mo. median) and ADT-free survival periods (32.7 mo. vs. 14.9 mo. median).
  • Siemens’ Magnetom Free.Max FDA: Siemens Healthineers announced the FDA clearance of its small and lightweight MAGNETOM Free.Max 0.55T MRI, touting its potential to bring whole-body MRI into new clinical environments (UC clinics, orthopedics, ICU, ED) and the benefirts of its wide 80cm bore (supporting obese or claustrophobic patients). The MAGNETOM Free.Max owes its clinical expansion potential to its very low helium (1 liter, no quench pipe) and low installation requirements (“no harder to install than a CT,” <3.5 tons, <80” high).
  • A Case for Regular HCC Screening: A new JAMA study found that performing regular ultrasound screenings on patients with higher hepatocellular carcinoma risks (HCC) leads to earlier HCC detection and extends quality of life. The study evaluated 59k patients with HCC over a 14-year period, finding that patients screened every 6 months before their diagnosis had more quality-adjusted life-years than patients who weren’t screened during the 36 months before their diagnosis (men = +4.6 QALYs; women = +2.4 QALYs).
  • HHS’s PACS Alert: An HHS alert revealed that as of June 2021, 130 U.S. health systems are using vulnerable PACS systems that hackers could breach and gain access to data from 8.5m cases and 2m patients. The HHS alert urged healthcare providers to adopt a series of security protocols and patch over twenty PACS servers with known vulnerabilities.
  • Quantitative MRI Variations: National Institute of Standards and Technology researchers found wide variations in quantitative MRI T1 measurements across different scanner brands and field strengths, warning that this undermines diagnostic consistency, and calling for more rigorous quality control standards. The researchers assessed T1 measurement variations across 27 MRI scanners at 9 sites (1.5T = two OEMs & 9 scanners; 3T = 3 OEMs & 18 scanners) using an inversion recovery sequence and a variable flip angle sequence, finding significant bias and variation and no consistent pattern between vendors.
  • SHINE’s $150m: Radioactive isotope company SHINE Medical Technologies closed a massive $150m Series C-5 round (total now $394.7m) that it will use to support its efforts to develop fusion-based technologies. SHINE’s fusion tech strategy goes well beyond medical imaging, but the new funding should also support its shorter-term plan to use nuclear fusion to produce medical isotopes like Mo-99 without a nuclear reactor.
  • Shifting Spine Injections to the Clinic: A new JACR study provided a compelling argument for performing image-guided spine pain injection procedures at clinic-based interventional suites. The review of 276 procedures (190 in clinics, 86 in hospitals) revealed that the two settings handled cases with similar complexity without complications. However, clinic-based procedures were performed with shorter fluoroscopic times (112 sec vs. 163 sec), procedural times (17 min vs. 28 min), and wait times (20 min vs. 38 min), while hospital-based procedures cost the health system over 14x more.
  • Intelerad Acquires HeartIT: Intelerad continued to add to its enterprise imaging portfolio, acquiring cardiac image viewing and reporting software company, Heart Imaging Technologies (aka HeartIT). Intelerad has adopted an aggressive M&A strategy and shown an increased focus on cardiac imaging informatics over the last few years, buying Clario Medical in 2018 (zero-footprint worklist), Radius in mid-2020 (cloud tech), Digisonics in late-2020 (cardiovascular & OB/GYN PACS), and LUMEDX in early 2021 (cardiovascular information systems).
  • Quantifying CMR PAT: UK researchers developed the first deep learning system that can automatically quantify pericardial adipose tissue (PAT, fatty tissue surrounding heart) using standard MRI scans, suggesting that PAT could become a cardiovascular disease biomarker (it was previously difficult to measure). In tests against independent and external datasets, the CMR PAT tool achieved high correlations with reference standard CCT PAT measurements, while showing that larger CMR PAT areas were significantly associated with diabetes.
  • California Kickbacks: California imaging center CEO, Sam Sarkis Solakyan, found out the hard way that crime doesn’t pay (at least not in the long run) after being convicted of 12 counts of fraud for operating an MRI “cross-referral scheme.” Solakyan paid recruiters and clinicians over $8.6m to refer workman’s comp patients to his nine imaging centers where he generated over $250m in kickback-related imaging procedure billings.

Duke’s Case for ClearRead CT

This Riverain Technologies case study details how Duke University Medical Center integrated ClearRead CT into its chest CT workflows, reducing read times by 26% and improving nodule detection by 29%.

– Sponsored.


The Resource Wire

  • Not sure how to navigate the upcoming surprise billing laws? This Healthcare Administrative Partners post details how the No Surprises Act might affect your practice and what processes will be required to handle these changes.
  • The future of radiology is here. Join Nuance on July 14th to experience the true potential of AI and its impact throughout the patient journey.
  • In this quick video, Sharp Memorial Hospital interventional radiologist, Jim Lyon, MD, describes the image quality and dose advantages of Canon Medical Systems’ Alphenix Sky+ system.
  • As part of United Imaging’s Equal Healthcare For All mission, they embed their most innovative technology in all MI systems. This forward innovation approach means that even their most financially accessible scanner can do a total body PET exam in five minutes, without compromising imaging quality.
  • It’s clear that structured reporting is a must for CVIS platforms, but they aren’t all created equal. This Fujifilm Healthcare article reveals what physicians and sonographers view as the “non-negotiable” CVIS structured reporting features.
  • Check out this case study (page 30) showing how GE’s Arterys-powered ViosWorks deep learning solution significantly shortens myocardial hypertrophy CMR exam and analysis times.
  • This Zebra-Med post details how AI is revolutionizing population health programs through automation, workflow integration, and significantly expanding early disease detection.
  • See how East Texas Medical Center reduced its abdominal CT contrast volumes by 30% after adopting Bayer Radiology’s contrast dose management software.