Ziad Obermeyer, MD, in the opening lecture at SIIM 2023.

Imaging Wire Sponsors

Annalise.ai  •  Arterys  •  Bayer  •  Blackford  •  CARPL.ai  •  Change Healthcare  •  Enlitic  •  GE HealthCare  •  Hyperfine  •  Intelerad  •  Medality  •  Merge by Merative  •  Nuance  •  Riverain Technologies  •  Siemens Healthineers  •  Subtle Medical  •  United Imaging  •  Us2.ai  •  Visage Imaging

Artificial Intelligence

AI Reinvigorates SIIM 2023

AUSTIN – Before AI came along, the Society for Imaging Informatics in Medicine (SIIM) seemed to be a conference in search of itself. SIIM (and before it, SCAR) built its reputation on education and training for radiology’s shift to digital image management. 

But what happens when the dog catches the truck? Radiology eventually fully adopted digital imaging, and that meant less need to teach people about technology they were already using every day.

Fast forward to the AI era, and SIIM seems to have found its new mission. Once again, radiology is faced with a transformative IT technology that few understand and even fewer know how to put into clinical practice. With its emphasis on education and networking, SIIM is a great forum to learn how to do both. 

That’s exemplified by the SIIM keynote address on Wednesday, by Ziad Obermeyer, MD, a physician and researcher in machine learning at UC Berkeley who has published important research on bias in machine learning. 

While not a radiologist, Obermeyer served up a fascinating talk on how AI should be designed and adopted to have maximum impact. His advice included:

• Don’t design AI to perform the same tasks humans do already. Train algorithms to perform in ways that make up for the shortcomings of humans.
• Training algorithms on medical knowledge from decades ago is likely to produce bias when today’s patient populations don’t match those of the past.
• Access to high-quality data is key to algorithm development. Data should be considered a public good, but there is too much friction in getting it. 

To solve some of these challenges, Obermeyer is involved in two projects, Nightingale Open Science to connect researchers with health systems, and Dandelion Health, designed to help AI developers access clinical data they need to test their algorithms. 

The Takeaway 

The rise of AI – particularly generative AI models like ChatGPT –  has given SIIM a shot in the arm from a content perspective, and the return of in-person meetings plays to the conference’s strength as an intimate get-together where the networking and relationship-building is almost as important as the content. Please follow along with the proceedings of SIIM 2023 on our Twitter and LinkedIn pages. 

Us2.ai AI Echo for Cardiac Amyloidosis

Transthyretin (ATTR) amyloidosis is emerging as an under-recognized form of heart failure, especially in the elderly. In this video, Marianna Fontana, MD, PhD, of UCL describes how Us2.ai echo AI technology was used to assess ATTR.

Improving Radiology Accuracy and Scale with Cloud AI

See how AI and the cloud combine to alleviate IT challenges and amplify radiologist performance in this Arterys white paper.

The Wire

• Inflation Wipes Out Salary Gains: Physician salaries rose in 2022 compared to 2021, but high inflation wiped out those gains. So says the latest data from MGMA in its survey of almost 190,000 providers across 6,800 organizations. The survey found “modest” gains in payment for doctors (2.13% to 2.54%) that failed to keep up with 6.5% inflation in 2022. Diagnostic radiology has seen 5.25% wage growth over 3 years, about in the middle of the pack for specialists.

• ChatGPT vs. Google Bard: In a face-off between large language model AI platforms published in Radiology, OpenAI’s ChatGPT came out top compared to Google’s Bard for accuracy in answering non-expert questions on lung cancer prevention and screening. In a set of 120 questions, ChatGPT (version 3.5) answered more correctly than Bard (70.8% vs. 51.7%). To improve accuracy and avoid the “hallucination effect,” the authors recommended that information from medical societies be added to the data used to train LLMs.

• RadNet’s $175M Offering: Imaging services provider RadNet says it will issue $175M in common stock in an offering that it plans to use to pay down $100M in debt and for general corporate purposes. Separately, CFO Mark Stolper last week said the company had generated $18M in revenue from its AI division, largely in line with previous forecasts. Much of that is from its Enhanced Breast Cancer Detection (EBCD) service, which charges patients for AI analysis of their images. 

• Ultra-High-Res CBCT Spots Brain Fistula: In a case study in JAMA Neurology, German researchers describe how they used ultra-high-resolution cone-beam CT myelography to visualize a cerebrospinal fluid venous fistula in a 78-year-old man presenting with headache and memory problems. Images were acquired with Siemens Healthineers’ Artis Icono flat-panel angiography system at 0.14mm spatial resolution, which enabled visualization of very small anatomy – in fact, the authors said the UHR-CBCT technique “provides the highest spatial resolution in spinal imaging currently available.”  

• FDA Panel Endorses Leqembi: An FDA advisory committee endorsed the efficacy and clinical benefit of Leqembi, Esai’s recently approved drug for early diagnosis of Alzheimer’s disease. The move is a shot in the arm for Esai, which hopes Leqembi avoids a fate similar to another Alzheimer’s drug that targets beta amyloid, Aduhelm, but that has mostly been withdrawn from the market. Leqembi could spur  use of neuroimaging to monitor treatment, although CMS said patients have to be part of a clinical registry.

• AI Makes Users Lonely: Will using AI make you lonely? That’s the suggestion of a new study in Journal of Applied Psychology that summarized the experiences of workers in 4 different countries who used AI. Such workers reported higher levels of loneliness and insomnia, although the researchers pointed out that they couldn’t be sure that AI was actually causing the association. They suggested that AI developers add social features to their algorithms to simulate human-like interactions.

• Hybrid Gamma Camera Installed: UK device developer Serac Imaging Systems has made the first US installation of its Seracam portable hybrid gamma-optical camera, at Ohio State University. Seracam combines gamma camera and optical images for better visualization of anatomy and function, and is designed to be moved to a patient’s bedside. Ohio State researchers will perform clinical testing of Seracam in a six-month study of 25 patients, and will compare the technology to conventional gamma cameras.

• CCTA Before LAAC: When interventional cardiologists use pre-procedural CCTA scans to plan echocardiography-guided percutaneous left atrial appendage closure (LAAC) procedures, patients have significantly better outcomes. That’s the takeaway from a recent JACC study that compared 219 LACC procedures, including 92 where ICs had access to pre-procedural CCTAs. After adjusting for confounders, CCTA planning was associated with far higher success rates following the procedure (93.5% vs. 81.1%; adjusted OR: 2.12) and at 45 days (83.7% vs 72.4%; adjusted OR: 2.76). 

• PE Firm Buys Premier Radiology: Teleradiology provider Premier Radiology Services of Florida has been acquired by Grovecourt Capital Partners, a PE firm that invests in “lower-middle market” opportunities. Premier offers teleradiology services in 48 states, with 100 radiologists reading close to 2M images a year. Grovecourt said it intends to expand in teleradiology through organic growth as well as M&A; the deal demonstrates private equity’s continued appetite for radiology acquisitions.

• AR Firm Medivis Raises $20M: Augmented reality developer Medivis has raised $20M in a Series A financing designed to grow the company and commercialize its surgical navigation technology, including its SurgicalAR platform. Interestingly, the funding round was led by an investor group that includes Mayo Clinic, Disney chief executive Bob Iger, and NBA star Kevin Durant. Medivis launched SurgicalAR at HIMSS 2019; the software was cleared by the FDA a few months later.  

• Apollo Touts SIIM Highlights: Apollo Enterprise Imaging is highlighting new cloud-based enhancements to its arcc enterprise imaging software at SIIM 2023. The new version of arcc supports storage of medical images and other clinical data on cloud storage locations hosted by Apollo on Amazon Web Services (AWS). Apollo makes arcc available in modules that include a universal viewer with EMR interface, a clinical content repository and IT management platform, and a mobile app. 

Monitoring AI Solutions in Healthcare

Clinical applications for healthcare AI are rapidly expanding, but many barriers are still preventing widespread adoption. This Nuance post explores a critical set of questions: what happens after an AI model goes into production, and how to know if it continues to perform as expected?

Visage 7 Named #1 Universal Viewer

Visage Imaging’s Visage 7 was named the #1 Universal Viewer in the 2023 Best in KLAS Software and Services Report. Find out what’s behind this award-winning solution and learn more about Visage’s One Viewer philosophy.

How to Grow Your Practice When It’s Impossible to Recruit and Retain

Today’s employment environment in radiology has never been more challenging. But there are tools you can use to keep your radiologists happy and engaged, according to this white paper from Medality.

The Resource Wire

• 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%.

• Healthcare providers can’t feel confident in the impact they make for their patients without feeling their IT environments are resilient and reliable. This eBook from Merge breaks down 3 essential ingredients of a resilient imaging IT environment: hybrid cloud, reliable uptime, and zero-trust security. 

• 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.

• When Middlesex Health set out to adopt imaging AI, the Connecticut-based community hospital made the unique decision to start with non-interpretive AI solutions. See how that decision led them to Subtle Medical, and the impact it had with all of the hospital’s imaging stakeholders.

• Did you know that healthcare generates 30% of all global data created every year? Learn more facts like this – and how Enlitic’s Curie platform can help deal with the data deluge – in this infographic. 

• Trying to figure out how your IT resources can handle increased AI adoption? This Blackford paper details how the cloud is helping radiology organizations scale their computing resources to support multiple AI applications or algorithms.

• How is fast, quantitative SPECT/CT supporting quicker assessment of treatment response in patients with metastatic prostate cancer? Find out in this case study from New York-Presbyterian Hospital/Weill Cornell Medical Center, provided by Siemens Healthineers. 

• Curious how certain your AI is about its own finding? Annalise.ai’s confidence bar displays the likelihood of each finding and the AI model’s level of certainty, helping clinicians perform their interpretations with greater confidence.

• Imaging’s cloud evolution didn’t happen all at once. This Change Healthcare animation details the history of digital imaging architectures, and how cloud-native imaging improves stability and scalability, ease of management, patient data security, and operating costs.

• 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.

• When RWJBarnabas Health wanted to boost its percentage of radial-access interventional cardiology procedures being performed, they leveraged data from HealthView Analytics, a business intelligence platform by Intelerad subsidiary Lumedx.

• “It completely changes the way we think about MRI imaging.” Take a look at this video interview with Mass General’s Chief of Neurocritical Care to see how clinicians can use Hyperfine’s Swoop Portable MRI to eliminate care disruptions in the ICU by keeping critically ill patients in the unit throughout the neuroimaging process.

• These Theranostics thought leaders share their perspectives on the discipline’s beginnings and discuss why it’s now primed to change how we treat cancer, and soon, other diseases. Personalized medicine is here, and this video from GE HealthCare explores its anticipated growth.

• After setting ambitious regulatory and commercialization goals, Lunit leveraged CARPL.ai’s platform and operational guidance to complete the clinical trials needed for its INSIGHT CXR and MMG AI tools’ FDA clearances.