fbpx

Unexpected Imaging | DLIR Advantage


“Finally, I see a typical pneumonia on chest CT.”

A tweet from Saurabh Jha, MBBA (aka RogueRad) on how his chest CT reads are changing as the COVID emergency subsides.


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


COVID’s Unexpected Imaging Impact

We’ve covered plenty of studies detailing how the COVID emergency reduced overall imaging volumes, but a new Mass General paper revealed that it also brought them significant and unexpected per-patient ED imaging increases.

  • The Study – The MGH team compared its emergency imaging during COVID’s April 2020 peak against the same period in 2019, measuring overall and per-patient imaging utilization.
  • ED Shift – COVID resulted in a big drop in MGH’s overall ED patients (5,686 in April 2020 vs. 9,580 in 2019), largely due to a major decline in ED patients without respiratory complaints (3,493 in 2020 vs. 8,207 in 2019). As you might expect, patients presenting with respiratory complaints nearly doubled (2,193 in 2020 vs. 1,373 in 2019).
  • Per-Patient Results – Interestingly, patients who came into the hospital’s ED without respiratory complaints during April 2020 generated 33% more imaging studies and 24% more imaging wRVUs (per-patient) versus the same month in 2019. Despite a 24% absolute decline in ED imaging exams and a 36% drop in ED imaging wRVUs across all patients, MGH’s non-respiratory imaging increases actually caused its overall wRVU/patient and studies/patient rates to increase by 7% and 23%.
  • A Per-Patient Theory – The authors speculated that these unexpected increases could be due to COVID’s influence on both clinician and patient behavior. The COVID pandemic likely caused clinicians to place a greater emphasis on diagnostic certainty, resulting in imaging orders that might not have happened in other years. Meanwhile, because COVID exposure fears kept many patients from seeking care, it’s possible that the non-respiratory patients who did go into the ED had more acute conditions.


Improving Efficiency with Arterys Lung AI

See how Arterys Lung AI allowed radiologists to reduce their thoracic CT interpretation times by between 35% and 55%, while increasing their nodules reported per study.

– Sponsored.



HAP’s Guide to Staying Private

Independent and staying that way? Healthcare Administrative Partners just released a helpful set of guidelines that radiology practices can follow to stay private despite ongoing consolidation pressures.

– Sponsored.


The Wire

  • TrueFidelity Advantage: A new EJR study detailed how deep learning image reconstruction (DLIR) can help radiologists predict whether pancreatic cancers could be resected (aka removed). The researchers reconstructed contrast-enhanced CTs from 47 patients with pancreatic cancer using filtered back projection (FBP), adaptive statistical iterative reconstruction (ASiR-V), and GE Healthcare’s TrueFidelity DLIR technology. Four radiologists identified resectable cancers far more accurately using TrueFidelity (0.91 AUC vs. 0.75 w/ FBP, 0.81 w/ASiR-V) and posted far higher confidence scores with the GE DLIR solution.
  • Physician Shortage Coming, but Maybe Not for Radiology: The AAMC forecast that the US could face a shortage of between 37,800 and 124,000 physicians by 2034 due to population growth/aging and upcoming physician retirements. Like last year’s report, the new AMMC forecast suggests that AI efficiencies could reduce future demand for radiologists and pathologists.
  • DL Auto-Segmenting for RT: Canadian researchers found that deep learning-based CT auto-segmented contour (DC) models effectively support radiotherapy planning. In the study (n = 551 cases, 203 surveys), the DCs identified organs at risk for a range of cancers (central nervous system, head and neck, prostate). The DCs required minimal editing (score ≤ 2 on a 5-point scale) and achieved high satisfaction ratings ( ≥ 4.1 on a 5-point scale).
  • Zebra & NHS’ Population Health Plan: Zebra Medical Vision will receive a share of the NHS/NHSX’s Long Term Plan funds (£140m over 4yrs), which is intended to accelerate the NHS’s AI adoption. With this NHS grant, Zebra-Med will work with the NHS (plus Zebra’s existing NHS, academic, and health society partners) to clinically validate population health AI and develop an osteoporosis screening and care pathway.
  • A Case for Risk-Based Breast MRI Screening: South Korean researchers found that supplemental breast MRI screening effectively catches interval cancers among women with a personal history of breast cancer and key additional risk factors (family BC history, estrogen receptor– and progesterone receptor–negative primary cancers, moderate or marked background parenchymal enhancement). Among these women, breast MRI caught 1.5 interval cancers per 1,000 screenings (n = 6,603 exams), most of which were early stage.
  • The First AAA Tracer: A study presented at SNMMI 2021 detailed a new PET radiotracer (64Cu-DOTA-ECL1i) that can detect abdominal aortic aneurysms (AAAs) before ruptures occur by targeting AAA’s CCR2 biomarker. The researchers found 64Cu-DOTA-ECL1i to be safe and effective for imaging CCR2 in human patients, while predicting AAA ruptures in animal tests, and showing that it could help develop future AAA treatments.
  • The Road to Multimedia Reporting: HIMSS and SIIM released a white paper outlining the status and future potential of interactive multimedia reporting (IMR), which would combine a full range clinical media (e.g. images, video, sound, text, anatomic maps, tables/graphs) collected from and accessed by a similarly full range of departments (e.g. radiology, cardiology, dermatology, ophthalmology). This multispecialty content is already being created and the HIMSS-SIIM team is confident in IMRs’ significant benefits, so we have a good starting point. However, significant technical, workflow, and adoption barriers will have to be overcome in order for IMRs to become a standard.
  • Low-Field MRI’s Lung Potential: A new study highlighted low-field MRI’s potential for lung disease evaluations. NIH researchers performed 0.55T MRI and CT scans on 24 patients with common lung abnormalities, rating MRI’s image quality as “good” or “excellent” in the majority of scans and finding that it sufficiently spotted many common lung diseases. However, MRI fell short of CT in discerning ground-glass opacities and tree-in-bud nodules.
  • AMA Reining in PAs: The AMA recommended a series of new policies intended to streamline the prior authorization (PA) process, specifically targeting peer-to-peer (P2P) reviews and public health emergencies. The new policies call for: 1) Making P2P PAs actionable within 24hrs; 2) Requiring reviewing P2P physicians to have expertise in the condition / disease and its treatments; 3) Requiring P2P reviewers to follow evidence-based guidelines; 4) Suspending all PA requirements and extending all existing approvals during national health emergencies; 5) Automatically approving all necessary incidental surgeries / procedures if performed during an original procedure.
  • Early Alzheimer’s PET Tracer: A SNMMI 2021 presentation detailed how the novel 18F-MK6240 PET radiotracer was able to measure increases in brain tau before patients showed symptoms of Alzheimer’s disease. The researchers performed 18F-MK6240 PET scans on 106 healthy and Alzheimer’s patients, finding that 18F-MK6240 uptake was higher among the Alzheimer’s patients in both baseline and 12mo follow-up scans.
  • Precontrast T1 MRI Unnecessary for MS: Skip the precontrast T1-weighted MR scans when examining patients that might have multiple sclerosis. That’s from a new German study that had two neuroradiologists review pre- and post-contrast T1 and T2 spinal MR images from 265 patients with suspected MS (once w/ both T1 & T2, once w/ only T2), finding no significant difference in diagnostic confidence or detection rates between the two approaches.
  • Cardiac SPECT AI: A SNMMI 2021 presentation detailed a SPECT-based deep learning model that can predict patients’ risk of major cardiac events. Developed with a multicenter SPECT MPI dataset (n = 20,401 patients), the algorithm predicted patients’ 4.7-year cardiac event risk with a 0.75 AUC (vs. stress TPD’s 0.70 & ischemic TPD’s 0.68) and found that patients with the highest AI risk scores were 10.2-times more likely to experience a cardiac event than patients with the lowest scores.
  • Pulmonary Fibrosis PET Tracer: Pairing PET with a 68Ga-labeled fibroblast activation protein inhibitor (FAPI) could be able to noninvasively identify and monitor pulmonary fibrosis. Wisconsin-based researchers targeted the protein in two groups of mice (an induced pulmonary fibrosis group and controls), finding that the mice with pulmonary fibrosis had a much higher uptake of the radiotracer.

United Imaging’s Software for Life

Because of United Imaging’s Software Upgrades for Life program, every time United Imaging launches a new solution it can automatically be installed in every compatible system at no cost.

– Sponsored.


The Resource Wire

  • 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.
  • CD burning issues? Check out this one-minute video showing how Novarad’s CryptoChart image sharing solution allows patients to easily access and share their medical images using personalized, highly secure QR codes.
  • Watch Jared Christensen, MD, MBA explain how Duke University Health uses Riverain Technology’s ClearRead CT Vessel Suppress and ClearRead CT Detect in its daily practice.
  • Discover how to make your next AI workflow integration a success in tomorrow’s webinar featuring AI leaders from DASA, Signify Research, and GE Healthcare.
  • See why the time is right for imaging AI-enabled population health in this Hardian Health profile featuring Zebra-Med’s CEO, Zohar Elhanani.
  • With Turbo Suite Excelerate by Siemens Healthineers, you can reduce MRI exam times by up to 50%. See how it’s possible in these videos featuring example hip, knee, and brain scans.
  • Hear Regional Health CIO and CMIO, Stephanie Lahr, MD, explain how Nuance PowerShare enables Regional Health to share diagnostic images across a large geographical region quickly and efficiently.

Today’s issue was brought to you by Jake Fishman and Jason Barry.