Ink Agents | Mammoscope | AI Reactions

“. . . a bit like a phone-a-friend IT solution on your ultrasound,”

Fetal medicine scientist, Professor Aris Papageorghiou, describing the auto-recognition tools built into GE Healthcare’s Voluson SWIFT women’s health ultrasound system.

I’m delighted to announce that Siemens Healthineers is now an Imaging Wire premium sponsor. Needless to say, Siemens Healthineers does so much for our industry and I’m very excited to share their message with you.

Every issue of The Imaging Wire is made possible by our sponsors, so make sure to keep up with the great work they’re doing.

Imaging Wire Sponsors

  • Bayer Radiology – Providing a portfolio of radiology products, solutions, and services that enable radiologists to get the clear answers they need.
  • GE Healthcare – Enabling clinicians to make faster, more informed decisions through intelligent devices, data analytics, applications and services.
  • Healthcare Administrative Partners – Empowering radiology groups through expert revenue cycle management, clinical analytics, practice support, and specialized coding.
  • Hitachi Healthcare Americas – Delivering best in class medical imaging technologies and value-based reporting.
  • Nuance – AI and cloud-powered technology solutions to help radiologists stay focused, move quickly, and work smarter.
  • Riverain Technologies – Offering artificial intelligence tools dedicated to the early, efficient detection of lung disease.
  • Siemens Healthineers – Shaping the digital transformation of imaging to improve patient care.

The Imaging Wire

Ink Agents

USC researchers found that common pigments and dyes, including those used in tattoos and food coloring, could advance contrast-enhanced medical imaging. Here’s the story:

  • Inspiration – Inspired by an art class and a tattoo artist, the researchers set out to discover if there are “imaging agents hiding in plain sight.”
  • Palette Properties – The team tested the fluorescence (light absorption and emission; supports sensitivity) and Raman (light scattering; supports specificity) properties of 30 commonly used tattoo pigments and food/cosmetic dyes. Seven colorants displayed fluorescence properties that were “comparable to or exceeded” the FDA’s three currently approved clinical dyes, while some of the would-be agents’ Raman signatures could support multiplexing.
  • The Study – The team tested the best-performing dyes and pigments by using them as imaging agents in mice with cancerous tumors. They found that the FDA-approved Green 8 dyes used in drugs and cosmetics have significant potential for imaging cervical and colon tumors, while the Orange 16 pigment found in tattoo inks has promising fluorescent properties and tumor targeting potential.
  • The Takeaway – By using a variety of these pigments and dyes loaded into liposomal nanoparticle-based agents, they could be used in contrast enhanced scans that show what cancers might be present based on which pigment/dye binds to different tumor markers. More research is obviously needed, but the team is optimistic and this makes for a much better story than most imaging agent studies.

Dual Scan Mammoscope

University at Buffalo researchers received a 4-year, $1.4m NIH grant to support the development of their ‘dual scan mammoscope’ (DSM), which uses a combination of ultrasonic and photoacoustic imaging to improve mammography screening for women with dense breasts.

  • How DSM Works – The portable system uses a laser to illuminate breast tissue, which generates acoustic waves that are measured by ultrasound technology, while performing two simultaneous scans (top and bottom of the breast).
  • DSM Benefits – The DSM’s main goal is to improve screening for women with dense breasts, who have much higher breast cancer risks and accuracy rates as low as 62%. DSM also boasts key advantages over mammography (no radiation, much less compression), MRI (cost and contrast), and photoacoustic imaging (can scan deeper than 7cm).
  • Up Next The team will use the grant to improve the DSM’s breast coverage, ultrasound capability, and its ability to image tissue closer to the chest wall, while also developing an algorithm to display 3D images that it captures. They also plan to partner with nearby imaging and cancer centers to test and refine the DSM’s image quality.

The Wire

  • AI Reactions: A Reaction Data survey of 45 imaging leaders revealed that imaging AI’s importance is growing within U.S. radiology departments (83% “important” in 2020 vs. 65% in 2017), and the majority of institutions that don’t use AI plan to adopt it within the next two years (31% in 0-12mo, 46% in 13-24mo). The respondents cited quality and productivity as the most common reasons for AI’s importance to their practice (36% and 28%), while the most commonly used AI applications focused on breast, lung, and cardiovascular imaging (37%, 12%, 12%).
  • Yale Ditches the Apron: The Yale New Haven Hospital system revealed that it will stop using X-ray aprons, making it the latest of several U.S. hospitals to ditch the protective gear. Although the shift away from X-ray aprons has been coming for some time due to their questionable effectiveness (worse images, more radiation exposure), the COVID emergency is apparently accelerating this shift given the aprons’ added transmission risks and cleaning requirements.
  • AI’s Geographic Bias: A Stanford team’s JAMA research letter revealed high levels of geographic bias in clinical AI training sets, suggesting that this could create bias in the deep learning algorithms that use them. Their review of 74 published AI studies (including 35 imaging AI studies) found that 76% of the training cohorts were from California, Massachusetts, and New York, while 37 states weren’t even represented in the training sets.
  • GE Voluson SWIFT: GE Healthcare unveiled its Voluson SWIFT women’s health ultrasound system, which launches with a new hardware design, user interface, and added personalization and workflow features. The Voluson SWIFT is highlighted by its embedded AI platform, which supports its new SonoLyst app (recognizes the 20 major OB/GYN views, cuts scan workflow by 73%), as well as GE’s existing Scan Assistant (guides clinicians, reduces patient scanning by up to 45%), SonoBiometry (reduces measurement time by 38%), and SonoCNS tools (reduces fetal brain measurement keystrokes by 75%).
  • Ultromics & Mayo: Mayo Clinic and Ultromics announced plans to leverage Ultromics’ EchoGo Core software to study how COVID-19 affects the cardiovascular system (n = 500 CV19+ patients). These results could help physicians triage COVID-19 patients using echocardiograms and predict / manage potential cardiovascular outcomes. Ultromics has clear momentum, as this high-profile collaboration comes shortly after a $10m funding round and a major grant from the UK NHS.
  • Multiphase CTA DL for LVO: A Brown University team developed a deep learning model to detect large vessel occlusion (LVO) strokes, enhanced by using multiphase CT angiography scans. The multicenter retrospective study (n = 540 w/ suspected stroke, 270 w/ LVO, 62 in test set), found that the deep learning model identified LVO far more effectively using multiple CTA phases (0.89 AUC, 100% sensitivity, 77% specificity) than with single-phase CTA (0.74 AUC, 77% sensitivity, 71% specificity).
  • Adverse Reaction Training: A Yale radiologist developed a training program to help clinicians and trainees recognize and treat adverse reactions to contrast agents. The hands-on program now trains 200 clinicians each year, using a mannequin to demonstrate symptoms, patient and colleague communication techniques, and treatment practices (e.g. nebulizer, epinephrine).
  • Synthetic Advantage: A new AJR study revealed some solid evidence in favor of adopting synthetic 2D mammography (SM) created from DBT scans, suggesting that SM+DBT screenings could reduce overall radiation exposure versus DM+DBT without sacrificing accuracy. A review of 13 previous studies (n = 201,304 women; 7,252 w/ breast cancer) found that SM and SM+DBT screenings matched or beat DM and DM+DBT in terms of sensitivity (76% & 83% vs. 74% & 84%) and specificity (91% & 92% vs. 86% & 90%), while SM+DBT can be performed with just one scan (vs. two scans for DM+DBT).
  • Datatrak & RadMD’s Clinical Alliance: Datatrak International and RadMD launched a partnership combining Datatrak’s imaging platform with RadMD’s radiologist network to create an end-to-end clinical imaging trial solution. Datatrak’s platform will feed clinical trial-ready data (remote capture, PHI anonymization, endpoint, adjudication) to RadMD’s radiologists who will “design imaging strategies, interpret image reads, and perform adjudication reviews.”
  • CV19-Net: A University of Wisconsin team developed a deep neural network, descriptively named CV19-Net, which accurately differentiates patients with COVID-19 using chest x-ray. Trained on CXRs from 2,060 CV-positive patients and 3,148 patients with non-COVID pneumonia (5,806 & 5,300 CXR scans from multiple hospitals / vendors), CV19-Net achieved a 0.94 AUC when tested against 500 random CXRs (vs. 0.85 AUC by three thoracic rads).
  • Siemens & WakeMed: Siemens Healthineers and WakeMed (North Carolina, a longtime Siemens client) announced a 10-year value partnership that will combine Siemens’ imaging technology/service with a range of collaboration, innovation, and training components. Through the value partnership, WakeMed will pioneer new cardiovascular intervention and imaging programs, including a new hybrid OR that combines the ARTIS Icono angiography system and Siemens’ recently-acquired Corindus CorePath GRX robot, and they will create a Corindus Training Academy.
  • False Claims in Florida: Advanced Imaging of Port Charlotte, Florida will pay $501k to settle a federal fraud case that alleged the imaging center submitted false claims to Medicare and Tricare. Advanced Imaging allegedly administered contrast without physician supervision and improperly billed for services performed by physicians who weren’t credentialed by Medicare.
  • GE’s AI DReAM: GE Healthcare announced the launch of the AI DReAM project, a GE-led collaboration with the French AI business and research community that’s focused on accelerating the country’s imaging AI development (AI for liver cancer, lung disease, and brain cancer are already in the works). As part of the project, GE Healthcare will develop an advanced visualization platform with 2D/3D annotation tools, a Software Development Kit (SDK), and a technology framework for medical app development.

The Resource Wire

– This is sponsored content.

  • Check out how Siemens Healthineers’ Tin Filtration is like selecting better quality sunglasses.
  • This Bayer Radiology case study details how its Certegra P3T Software automates contrast enhanced abdominal CT injection protocols based on patient characteristics and contrast concentration.
  • With orthopedic care growing with the aging population, orthopedic imaging is growing along with it. This Hitachi blog details how more orthopedic practices are bringing imaging in-house and what factors they should take into account as they decide how/whether to add imaging to their suite of services.
  • In its latest Q&A Nuance’s Diagnostics leader, Karen Holzberger, sat down with Dr. Irena Tocino from Yale New Haven Health System to learn about how Nuance solutions helped YNHHS overcome the challenges brought by the COVID-19 pandemic.
  • They say that in times of crisis, you get to know who your real friends and partners are. This Q&A session details how Healthcare Administrative Partners stepped up to guide their client Triad Radiology Associates through the challenges presented by the COVID-19 pandemic.
  • Check out Riverain Technologies’ on-demand webinar demonstrating how its AI solutions integrated into LucidHealth’s radiology workflow and sharing best practices on how to combine AI with radiologist expertise.
  • This GE Healthcare article details how the Vscan Extend POCUS system combined with DiA Imaging Analysis’ LVivo EF solution is helping more clinicians guide cardiac care on the COVID-19 frontlines.

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!