Lång K et al, from the MASAI study in The Lancet Oncology on AI for screening mammography.

Imaging Wire Sponsors

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

Breast Imaging

Mammography AI’s Leap Forward

A new study out of Sweden offers a resounding vote of confidence in the use of AI for analyzing screening mammograms. Published in The Lancet Oncology, researchers found that AI cut radiologist workload almost by half without affecting cancer detection or recall rates.

AI has been promoted as the technology that could save radiology from rising imaging volumes, growing burnout, and pressure to perform at a higher level with fewer resources. But many radiology professionals remember similar promises made in the 1990s around computer-aided detection (CAD), which failed to live up to the hype.

Breast screening presents a particular challenge in Europe, where clinical guidelines call for all screening exams to be double-read by two radiologists – leading to better sensitivity but also imposing a higher workload. AI could help by working as a triage tool, enabling radiologists to only double-read those cases most likely to have cancer.

In the MASAI study, researchers are assessing AI for breast screening in 100k women in a population-based screening program in Sweden, with mammograms being analyzed by ScreenPoint’s Transpara version 1.7.0 software. In an in-progress analysis, researchers looked at results for 80k mammography-eligible women ages 40-80. 

The Transpara software applies a 10-point score to mammograms; in MASAI those scored 1-9 are read by a single radiologist, while those scored 10 are read by two breast radiologists. This technique was compared to double-reading, finding that:

• AI reduced the mammography reading workload by almost 37k screening mammograms, or 44%
• AI had a higher cancer detection rate per 1k screened participants (6.1 vs. 5.1) although the difference was not statistically significant (P=0.052)
• Recall rates were comparable (2.2% vs. 2.0%)

The results demonstrate the safety of using AI as a triage tool, and the MASAI researchers plan to continue the study until it reaches 100k participants so they can measure the impact of AI on detection of interval cancers – cancers that appear between screening rounds.

The Takeaway

It’s hard to overestimate the MASAI study’s significance. The findings strongly support what AI proponents have been saying all along – that AI can save radiologists time while maintaining diagnostic performance. The question is the extent to which the MASAI results will apply outside of the double-reading environment, or to other clinical use cases.

Imaging AI’s Next Wave

We may be entering a third wave of imaging AI’s rapid evolution, one that brings a shift from narrow point solutions to comprehensive multi-finding AI systems. Hear this discussion with Annalise.ai Chief Medical Officer Rick Abramson, MD, exploring how this transition could take place, how radiologist and VC perspectives on AI are changing, and how AI might continue to evolve in the future.

5 Questions about Lung Cancer and AI

Lung cancer causes more deaths globally than colon, breast, and colon cancer combined. But technologies like AI and low-dose CT screening can help. Learn more in this Q&A with Ankur Sharma, MD, head medical affairs digital radiology at Bayer.

Demystifying Radiology AI: A View from Two Perspectives

In our latest episode of The Imaging Wire Show, we take a look at radiology AI from two perspectives: A hospital that’s currently using AI and one that’s about to take the plunge. Our guests are Sohail Contractor, MD, of the University of Louisville Hospital in Kentucky and Amar Shah, MD, of Sinai Chicago in Illinois. Thanks to AI leader Qure.ai for bringing everyone together.

The Wire

• Informa to Buy HIMSS Show: In a massive consolidation in the healthcare B2B meeting industry, Informa is planning to buy the annual healthcare IT conference run by HIMSS. Informa will take over management of the HIMSS meeting, which drew 35k attendees at its most recent edition in Chicago, while HIMSS will “oversee developing expert content and programming” starting with the 2024 show in Orlando. The HIMSS show draws dozens of imaging IT vendors, including PACS and AI companies.

• AI Grows in the Netherlands: A study in European Radiology shows the number of radiology departments in the Netherlands using AI has risen steadily in just three years, to 33% in 2022 from 20% in 2020. What’s more, the number of unique AI products has grown 5X over the same period (34 vs. 7), and Dutch departments had an average of three AI products running last year versus one in 2020. Barriers to wider AI adoption include a lack of financial resources and IT integration difficulties. Sound familiar?

• Societies Spar over CT Colon Screening: The American College of Physicians (ACP) has issued new guidelines on colon cancer screening that specifically recommend against the use of CT colonography. The move prompted the ACR to call the guidance “a step backward” that could hurt underserved communities. ACP in its 2019 guidelines didn’t recommend CT colonography, but it also didn’t advise against its use. The battle illustrates CT colonography’s continued struggle to be accepted as a mainstream screening tool.

• MRI Radiomics Hones in on Crohn’s: A radiomics model that combined AI analysis of non-contrast MRI scans with clinical features like lab tests outperformed pediatric radiologists for diagnosing Crohn’s disease, a GI condition typically first diagnosed with children and young adults. In a study of 135 patients in AJR, radiologists said their combo of AI and clinical values had higher accuracy than three pediatric radiologists (93.5% vs. range of 83.7-88.1%). Wider deployment of the radiomics technique could boost detection of Crohn’s, which affects 750k people annually in the US.

• Sonio Gets FDA Nod for Fetal AI: In a major milestone, French prenatal ultrasound AI developer Sonio has received FDA clearance for its Sonio Detect software. Sonio Detect guides sonographers through prenatal scans in real time to ensure exam completeness, and then supports diagnosis by detecting potential abnormalities. With the clearance and armed with $14M in a recent funding round, Sonio is now poised to enter the US market. 

• COVID Crushed Cancer Diagnoses: A new analysis by American Cancer Society researchers in The Lancet Oncology found that the number of adult cancer diagnoses fell by 15% in April 2020 during the COVID-19 pandemic. Analyzing 2.4M cancer diagnoses, researchers found the proportion of stage I diagnoses fell the most compared to 2019 (from 40% to 31.4%), which led to a higher proportion of late-stage diagnoses (20% to 26.5%). The findings confirm previous studies showing the pandemic’s impact on cancer screening and diagnosis. 

• Supercon Experiment Replicated: Researchers in China claim to have replicated a South Korean experiment in achieving room-temperature superconductivity with LK-99, a new type of material. The Chinese experiment involves a flake of LK-99 responding to an external magnetic field; the original experiment has also reportedly been corroborated by researchers from the US as well as by a second Chinese team. If proven, room-temperature superconductivity could have a major impact on many fields of electronics and energy, including MRI.

• SCOT-HEART Radiation Dose: The landmark SCOT-HEART trial showed that use of CCTA in people with chest pain can reduce risk of heart attack or death. But how much radiation did they get? In a Radiology sub-analysis of 2.9k SCOT-HEART subjects, researchers found that those in the CCTA arm indeed got more cumulative radiation dose than standard care over five years (8.1 mSv vs. 0 mSv). But CCTA still has major clinical benefits, and new technologies have reduced radiation since the first SCOT-HEART patients were scanned in 2010. 

• Qure.ai Gains Another NHS Site: Qure.ai’s solutions are being installed at another hospital in the NHS system in the UK. Milton Keynes University Hospital NHS Foundation Trust (MKUH) will install Qure’s qER software for all non-contrast CT scans at the hospital to detect abnormalities such as intracranial hemorrhages (ICH). The Qure algorithm is one of the first that MKUH is making following its installation of the Blackford Platform from Blackford Analysis. Qure has been landing a number of NHS installations lately.

• Ultromics’ HFpEF Performance: A new Mayo Clinic-led study showed that Ultromics’ AI can improve HFpEF detection using a single echo view. The researchers applied Ultromics’ EchoGo Heart Failure solution to 1,284 patients’ echo exams, achieving 87.8% sensitivity and 81.9% specificity (above typical clinical performance without AI). They then applied the solution to 701 and 776 patients who had non-diagnostic HFA-PEFF and H2FPEF scores, finding that the AI model correctly reclassified 73.5% and 73.6% of these indeterminate cases.

• Ultromics Earns NTAP: And the hits just keep coming for Ultromics. This week the company said CMS included its EchoGo Heart Failure software in the NTAP program, with Medicare reimbursement starting October 1 at a rate of $1,023 per acute hospital inpatient stay. The technology is indicated as an echo diagnostic aid for HFpEF. Ultromics in July said CMS established a new HCPCS outpatient code that will apply to EchoGo use. 

• Outpatient Vascular Care: Good, Bad or Ugly? The New York Times and ProPublica have both recently detailed what appeared to be big problems with outpatient peripheral arterial disease treatments, suggesting that these procedures largely prioritize profit over patient care. However, an editorial by cardiologist Anish Koka, MD suggests that this coverage ignored “the real issues” with outpatient atherectomies, potentially at the expense of one well-respected physician’s reputation and many patients’ trust.

• Rivanna Get $30.5M for MSK Imaging: Rivanna has received $30.5M in funding from the US government’s Biomedical Advanced Research and Development Authority (BARDA) to further develop its Accuro XV musculoskeletal imaging system. Accuro XV combines AI with ultrasound-based bone and 3D imaging for soft tissues, and is designed to help detect and assess skeletal fractures and soft-tissue injuries. This is Rivanna’s second round of BARDA funding, and will be used to secure 510(k) clearance for AI-enabled fracture detection.

Image Quality and Molecular Imaging

How important is image quality in molecular imaging? Improvements in PET/CT and SPECT/CT are addressing historical challenges and empowering clinicians with more accurate and detailed information. Learn more in this white paper from GE HealthCare.

From CPACS to CVIS for Comprehensive Cardiac Care

Two prominent IT systems, CPACS and CVIS, have emerged as pivotal tools for managing the vast array of cardiology-related medical data. Learn the key differences between them in this Intelerad article.

See How Subtle Supports Accelerated MRI

Looking for a way to do faster MRI scans without losing image quality? Learn about the impact that Subtle Medical’s SubtleMR software has on accelerated MRI sequences in this case review image series.

The Resource Wire

• What are the factors that contribute to errors and misreads in radiology – and how can they be prevented? Find out how to improve your practice’s quality and accuracy in this white paper from Medality. 

• Get to know Clearpath and its novel approach to empowering patients by giving them digital access to their images and medical records in this video interview from SIIM 2023. 

• Adding cardiac PET to your PET/CT practice may present challenges, but the potential clinical and operational benefits make it worthwhile. Learn how to get started in this webinar from Siemens Healthineers. 

• 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.

• Trying to decide between public and private cloud? Watch this brief Change Healthcare animation detailing the operational benefits of public cloud hosting for imaging.

• If you’re in the business of using or providing AI, there’s a good chance you spend a lot of time managing AI evaluations. But are your evaluations as efficient or effective as they could be? Check out this Imaging Wire Show with Riverain Technologies CEO, Steve Worrell, detailing the best practices for mitigating AI adoption risks, today and into the future.

• When Sao Paolo’s Diagnosticos da America SA (DASA, the world’s fourth largest diagnostics company) set out to evaluate Qure.ai’s QXR solution for their pediatric chest X-ray workflows, they leveraged CARPL.ai’s platform to streamline their evaluation. See how it worked here.

• SOIN Soluciones Integrales of Costa Rica turned to Merge enterprise imaging solutions from Merative when it wanted to modernize the imaging environments of 50 hospitals across the country. Download this PDF white paper to find out how they did it.

• Arterys’ Cardio AI solution recently added a new Atrial Volumes feature that allows cardiologists and radiologists to easily quantify volumes for both left and right atria in cardiac MRI images.

• Join the conversation in this webinar recording and hear from PACS administrator Griff R. Van Dusen of Memorial Health System how Nuance PowerScribe One’s next-generation reporting experience helps streamline workflow and improve report quality so radiologists can get more done in less time. 

• Knowing an individual’s mammographic breast density is key to understanding their risk for cancer. A new article from Visage Imaging explains how an AI-based solution can improve breast density assessment. 

• What’s the latest news from United Imaging? Driven by a focus on R&D, United has increased its brand influence and market share worldwide. Get the details in the company’s first annual report since going public. 

• Strain imaging is the most sensitive parameter for determining myocardial deformation and systolic left ventricular function. In this video, see how echo AI from Us2.ai was used to analyze myocardial global longitudinal strain. 

• Still sharing patient images with CDs? It may be time to move on. Find out how PocketHealth can help you make the transition to fully digital image exchange in this article.

• Radiology follows a structured workflow to ensure accurate and timely imaging studies. Where workflow can break down – and where Enlitic’s Curie|ENDEX technology can help – is explained in this article.