The AI Gap | Stockholm3 | Automating the Mundane

“We’re at the start of something big.”

Ambra Health CEO, Morris Panner, during a HIMSS conversation about imaging’s evolution.

Imaging Wire Sponsors

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

The Imaging Wire

The AI Understanding Gap

It often seems like everyone in radiology has their mind made up about AI, but a new study revealed diverse levels of actual AI understanding, and found that imaging team members with lower AI understanding are generally more pessimistic about its future impact.

  • The Study – University of London researchers held interviews (12 radiologists, 6 radiographers) and a focus group (8 radiographers) with members of five different NHS breast imaging units to understand their knowledge and attitudes about imaging AI.
  • The Awareness Gap – The radiologists had far greater awareness of AI and its potential applications than their radiographer colleagues. This is partially because the radiologists learned about AI from a broad range of outside sources (professional networks, conferences, and AI developers), while the radiographers learned about AI from within their personal networks.
  • Rad Optimism – The radiologists were also more optimistic about AI, expecting it to help them eliminate repetitive tasks and believing that they will be able to influence future AI innovations. They also weren’t concerned about AI undermining their professional control / autonomy or about AI creating unacceptable risks / responsibilities.
  • Radiographer Pessimism – On the flip side, the radiographers were concerned that AI could “de-skill” their roles and possibly undermine their job security, while still creating more risks / responsibilities.
  • Team Agreement – Despite these significant differences, both groups were open minded about using new technologies and agreed that AI innovations could help alleviate current workforce shortages.
  • The Takeaway – There are surely plenty of radiologists that don’t share these interviewees’ AI optimism and plenty of radiographers who understand AI and see how it could help. However, this study tells an interesting story about the link between information exposure, understanding, and optimism. It also highlights the need to build radiographers’ AI awareness, especially considering how quickly modality-connected AI is evolving.

ETMC and Bayer’s Contrast Reduction

See how East Texas Medical Center reduced its abdominal CT contrast volumes by 30% after adopting Bayer Radiology’s contrast dose management software.

– Sponsored.

Creating Your AI Platform Strategy

Adopting a platform strategy can simplify the deployment and management of imaging applications and AI algorithms, but there’s a lot to consider. In this eBook, Blackford Analysis and its clients detail how AI platforms can benefit clinical and IT teams, and share guidelines to consider when selecting a platform.

– Sponsored.

The Wire

  • Stockholm3 Screening: A prostate cancer screening method combining the novel Stockholm3 blood test with MRI-targeted biopsies could reduce unnecessary exams and catch just as many clinically significant cancers. That’s from a new study out of Sweden (n = 12,750 men, 2,293 w/ high risk) that found the Stockholm3/MRI-biopsy approach reduced overdiagnosis (-69%), MRI exams (-36%), and biopsies (-8% & -51% ) compared to PSA-based screening with systematic or MRI-targeted biopsies.
  • Optellum and J&J’s Lung Cancer Alliance: Imaging AI startup Optellum announced its inclusion in Johnson & Johnson’s Lung Cancer Initiative, integrating Optellum’s Virtual Nodule Clinic platform (identifies / defines patients’ LC risk in CTs) into various J&J lung cancer research programs. J&J has been actively assembling its lung cancer diagnostics and treatment portfolio in recent years (e.g. Auris & NeuWave acquisitions, Veracyte partnership), making this a promising alliance for Optellum.
  • Reducing Breast MRI False Positives: A new study in Radiology detailed a prediction model that analyzes MRI findings and clinical characteristics to help reduce breast MRI false positives among women with extremely dense breasts. The predictive model analyzed data from 454 women who received positive breast MRI results (375 of which were false), showing that it would have reduced false positives by 45.5% and avoided 21.3% of benign biopsies without missing any cancers.
  • Qure.ai Quantifies: Qure.ai unveiled qER-Quant (FDA 510k & CE approved), a brain CT quantification tool used to measure the severity of brain injuries (TBI, hemorrhagic stroke, and hydrocephalus) and track progression over time. qER-Quant combines with Qure.ai’s original qER tool (head CT triage/detection) to give the company a relatively complete head CT portfolio, while making Qure.ai one of the few triage AI players to expand into quantification.
  • Neuroimaging’s Trainee Impact: Rising neuroimaging volumes are having a disproportionate impact on US radiology trainee workloads. Between 2002 and 2018, Medicare neuroimaging claims increased significantly for both radiologists and radiology trainees (+86.1% vs. +162.5%), driven in part by major increases in CT (102.9% vs. 196.8%) and MRI exams (59.9% vs 106.6%).
  • Ultromics’ Series B: Cardiac ultrasound AI startup Ultromics secured a $33m Series B round (total now $58.9m) that it will use to drive adoption of its EchoGo Core and EchoGo Pro echocardiogram analysis software. EchoGo Core and EchoGo Pro are already in use at select organizations (Mayo Clinic, OHSU) and are backed by some favorable studies, potentially making the time right for Ultromics to target more widespread commercialization.
  • Imaging Optimism: The AHRA’s latest Medical Imaging Confidence Index (MICI) revealed that US imaging managers / directors (n = 152; score range: 0-200) feel relatively confident about how they’ll perform in Q3 2021. The new report revealed a “high” 117 overall confidence score (vs. 103 in Q3 2020), driven by their confidence in scan/IR volume growth (136 vs. 125) and imaging’s growing role as a profit center (142 vs. 131). The imaging leaders had reduced (but still high) confidence about their operating cost stability (113 vs. 121), while their confidence about access to capital (101 vs. 67) and whether they’ll receive adequate reimbursements (98 vs. 88) improved from “low” to “neutral.”
  • Automating Tube Measurements: A new Journal of Digital Imaging study detailed an AI model that automates the common (but relatively mundane) task of confirming endotracheal tube positioning in CXRs. The vRad team trained the model with a large CXR dataset annotated with endotracheal tubes and carina bounding boxes. Against a pair of test datasets, the model was able to generate bounding boxes on new CXRs and calculate the distance between the tube and carina with similar variability as radiologists (0.70cm & 0.68cm w/ AI vs. 0.70cm for radiologists).
  • Fujifilm & ScreenPoint’s 3D Update: Fujifilm’s ASPIRE Cristalle DBT mammography system is now available in the US with the latest version of ScreenPoint Medical’s Transpara 2D/3D lesion detection software. Fujifilm previously offered Transpara’s 2D version, making the new 2D/3D version a logical addition to Fujifilm’s DBT platform.
  • 18F-FDG PET/CT for Pancreatic Cancer Response: UCLA researchers found that 18F-FDG PET/CT can help predict chemotherapy response at the early stages of pancreatic cancer treatment. The researchers evaluated 28 patients with pancreatic ductal adenocarcinoma who underwent 18F-FDG-PET/CT scans before and during chemotherapy (scans at baseline, week 4, week 11). Patients who showed metabolic and radiographic tumor responses in their 4-week scans had median survival rates of 36.2 months and 25.4 months, while patients who showed responses at week-11 had median survival rates of 27.4 months and 58.2 months.
  • Siemens Distributing SyMRI NEURO: SyntheticMR’s SyMRI NEURO software is now available directly through Siemens Healthineers across much of the world. SyMRI NEURO, which provides clinicians with more neuro MRI data and shortens scan times, was previously available through the Siemens Healthineers Digital Marketplace.

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

  • 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 (not CDs).
  • Take the AiCE challenge and see why half the radiologists in a recent study “had difficulty differentiating” images from Canon Medical Systems’ Vantage Orian 1.5T MR using its AiCE reconstruction technology compared to standard 3T MRI images.
  • This study by Hospital of Valenciennes, Paris used their emergency X-rays to show how effective Arterys Chest MSK AI is for finding X-ray abnormalities.
  • 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.
  • 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.

You might also like

You might also like..

Select All

You're signed up!

It's great to have you as a reader. Check your inbox for a welcome email.

-- The Imaging Wire team

You're all set!