A tweet from oncologic imaging ‘black belt’ Anton Becker, MD, PhD

Imaging Wire Sponsors

annalise.ai  •  Arterys  •  Bayer  •  Blackford  •  Canon Medical  •  CARPL.ai  •  Change Healthcare  •  Enlitic  •  GE Healthcare  •  Hyperfine  •  Intelerad  •  Novarad  •  Riverain Technologies  •  Siemens Healthineers  •  United Imaging  •  Us2.ai

Artificial Intelligence

Multimodal NSCLC Treatment Prediction

Memorial Sloan Kettering researchers showed that data from routine diagnostic workups (imaging, pathology, genomics) could be used to predict how patients with non-small cell lung cancer (NSCLC) will respond to immunotherapy, potentially allowing more precise and effective treatment decisions.

Immunotherapy can significantly improve outcomes for patients with advanced NSCLC, and it has already “rapidly altered” the treatment landscape. 

• However, only ~25% of advanced NSCLC patients respond to immunotherapy, and current biomarkers used to predict response have proved to be “only modestly helpful.”  

The researchers collected baseline diagnostic data from 247 patients with advanced NSCLC, including CTs, histopathology slides, and genomic sequencing. 

• They then had domain experts curate and annotate this data, and leveraged a computational workflow to extract patient-level features (e.g. CT radiomics), before using their DyAM model to integrate the data and predict therapy response.

Using diagnostic data from the same 247 patients, the multimodal DyAM system predicted immunotherapy response with an 0.80 AUC. 

• That’s far higher than the current FDA-cleared predictive biomarkers – tumor mutational burden and PD-L1 immunohistochemistry score (AUCs: 0.61 & 0.73) – and all imaging approaches examined in the study (AUCs: 0.62 to 0.64).

The Takeaway

Although MSK’s multimodal immunotherapy response research is still in its very early stages and would be difficult to clinically implement, this study “represents a proof of principle” that integrating diagnostic data that is already being captured could improve treatment predictions – and treatment outcomes.

This study also adds to the recent momentum we’re seeing with multi-modal diagnostics and treatment guidance, driven by efforts from academia and highly-funded AI startups like SOPHiA GENETICS and Owkin.

annalise.ai’s AI Confidence Bar

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.

Putting the Hyperfine Swoop to Use

Hyperfine’s Swoop Portable MR Imaging System is redefining MR accessibility, deploying MR-enabled decision-making across clinical settings within minutes. But do you know how the Swoop is actually being used? Check out this clinical case study detailing the settings and patient scenarios that the Swoop is supporting today.

The Wire

• Referrer Feedback Works: A new JAMA study showed that giving referring physicians individualized feedback about their high imaging volumes reduces their future imaging orders. The researchers randomized 3,660 Australian general practitioners (all among the top-20% of referrers for 11 different MSK imaging exams) into either an interventional group or a control group (n = 2,933 & 727). The interventional group’s imaging orders were significantly lower 12 months after they received the data-based feedback (27.7 vs. 30.4 per 1k patient consultations), resulting in an estimated 47,318 fewer exams.

• Scan.com’s Series A: UK-based “diagnostics-as-a-service” startup, Scan.com, completed a £2.2M Series A round (total funding now £4.2M) that it will use to fund its expansion to the US and Germany, and to support the development of new B2B products. Scan.com provides an API-connected portal that clinicians and patients can use to schedule imaging exams and view results. It previously targeted the UK’s private healthcare providers and patients, while serving as an NHS alternative.

• Google’s TB AI: A Google Health-led team developed a deep learning system that detected tuberculosis in chest X-rays as accurately as radiologists, and could help address TB-stricken countries’ radiologist access and cost barriers. The researchers trained the AI model using 165k CXRs from 22k people across 10 countries, and tested it against CXRs from 1,236 patients from four countries (17% with active TB). Compared to radiologists, the DL system detected TB with a higher AUC (0.89), greater sensitivity (88% vs. 75%), and noninferior specificity (79% vs. 84%). They estimated that it could reduce the cost of TB detection by 40% to 80% per patient.

• Osteoporosis DL Progress: University of Wisconsin researchers developed a DL system that accurately detected osteoporosis in abdominal CT scans, showing solid progress from a previous feature-based bone mineral density (BMD) algorithm. The researchers used the DL and BMD algorithms to analyze 11k CTs (automated level selection & L1 trabecular ROIs), using manual measurements as the reference standard. The DL model achieved a far higher success rate compared to the older BMD algorithm (99.3% vs. 89.4%), while allowing optimization for either specificity or sensitivity depending on ROI slice selection (single-slice = 39.4%/98.3%; seven-slice = 71.3%/94.6%). 

• Samsung’s V7 Ultrasound: Samsung’s US-based ultrasound and radiography subsidiary, Boston Imaging, announced the FDA clearance of its new V7 general ultrasound, highlighting its clinical versatility, image quality, clinical apps, and ease-of-operation. The V7 system is positioned just below Samsung’s V8 ultrasound that launched last fall, and shares many of the same core features.

• PSMA PET/CT’s Management Impact: A new study in JNM showed that 18F-DCFPyL PET/CT (Lantheus’ Pylarify PSMA tracer) has a significant impact on the management of prostate cancer patients who are being considered for salvage radiotherapy after radical prostatectomy and PSA recurrence. Among 98 participants, PSMA PET/CT detected disease in far more patients than diagnostic CT (46.9% vs. 15.5%), and prompted more “major” and “moderate” changes in treatment recommendations (12.5% vs. 3.2% & 31.3% vs. 13.7%). This comes a few months after another study found that Pylarify similarly improves prostate cancer treatment staging.

• Unnecessary X-Ray Sentencing: California orthopedic surgeon, Dr. Gary Wisner, was sentenced to 7 years in prison for performing unnecessary X-rays on his patients and fraudulently billing for those exams. Government investigators reviewed a random sample of Dr. Wisner’s patients, finding that at least 10 patients received “hundreds of unnecessary X-rays,” including scans of multiple body parts that weren’t associated with the patients’ condition.

• EchoNous’s AS Evidence: A paper published in JASE found that EchoNous’s Kosmos handheld ultrasound with CW Doppler (CWD) capabilities can reliably detect and grade aortic stenosis (AS). Of 118 patients with known or suspected AS, Kosmos with CWD achieved “excellent” agreement with a high-end cart-based echo system (intraclass correlation: 0.97). The handheld device detected at least moderate AS with 93% sensitivity, 98% specificity, and 96% total diagnostic accuracy.

• xWave’s Seed Round: Ireland-base imaging referral software startup, xWave Technologies, completed a €1.3M Seed round to fund its commercial expansion across Ireland, the UK, and Northern Europe. xWave’s xRefer platform uses evidence guidelines to help clinicians make appropriate referrals, reportedly reducing the time to create/send a referral and have a radiologist review it by 99.6%. xWave estimates that it would reduce the Irish healthcare system’s duplicate and unnecessary radiology referrals by more than 60% if adopted nationwide.

• Dismal U.S. Life Expectancy: Although this probably won’t come as a shock to many readers, a new report from the Commonwealth Fund found that the U.S.’ 78.8-year average life expectancy trails far behind other developed nations such as Japan (84.4 years), Spain (84), and Switzerland (84). Deaths from avoidable causes (including some imaging-screened diseases) and a weak commitment to community-based primary care were singled out as reasons for the poor performance, which looks even worse when considering that healthcare accounts for 20% of U.S. GDP, the highest among any country.

• Fixed C-Arm Demand: A IMV Medical survey revealed that the broad use of fixed C-arms is driving procedure and unit demand. Fixed C-arm units were spread across hospitals’ cardiology (42%), radiology (35%), and surgery departments (23%) during 2022, especially in larger >200 bed hospitals (73%-85% have C-arms in multiple departments). IMV forecasts that fixed C-arm procedure volumes will increase by 16% year-over-year, prompting 62% of hospitals to consider buying new fixed C-arm systems through 2025.

Canon Across America

Check out this Imaging Wire Show featuring Canon Medical Systems’ David Hashimoto, exploring how Canon is connecting with imaging teams and working to address their needs.

Data Governance Frustrations

How much time are you spending on interruptions? Based on this Enlitic report, it could be quite a lot, and data governance can eliminate many of them.

The Resource Wire

• See how healthcare leader Aster DM Healthcare leveraged the CARPL platform to connect its doctors, data scientists, and imaging workflows, and support its AI projects and development infrastructure.

• Precision medicine startup BAMF Health recently installed United Imaging’s uEXPLORER scanner, making it the first total-body PET/CT used for theranostics in the US. See how this combination will allow BAMF Health to deliver more effective and efficient theranostics treatments.

• How much time are you spending on interruptions? Based on this Enlitic report, it could be quite a lot, and data governance can eliminate many of them.

• With ongoing radiologist shortages and higher rates of burnout, there’s a great need for fast, effective, efficient medical imaging technologies – and those factors are driving 2022’s major medical imaging trends detailed in this Arterys report.

• Faced with growing subspecialty volumes, Diversified Radiology of Colorado adopted Intelerad’s InteleOne XE enterprise workflow solution to ensure all appropriate cases are read by subspecialists, while reducing its average TAT to under 17 minutes.

• Working out your AI business case? Check out this helpful Blackford Analysis report on how to calculate AI’s clinical, IT, administrative, and financial value to your organization.

• Over 9 out of 10 people who should be screened for lung cancer aren’t, and nearly 50% of lung cancer cases are caught in the advanced stages. We know from prostate and breast cancer screening that clear guidelines and increased screening saves lives. But lung cancer screening has been challenging. Riverain strives to make everything about the lungs clearer, so they assembled this resource page for anyone interested in starting or improving their lung screening program.

• Ready to make MRI more accessible to your patients? See how Siemens Healthineers’ MAGNETOM Free.Max expands MR imaging to more patients, sites, and providers.

• With radiation dose management now largely considered best practice, this Bayer white paper details the top five benefits of adopting contrast dose management.

• Check out this Imaging Wire Show featuring Novarad founder and CEO, Wendell Gibby, MD, exploring Novarad’s technological evolution and how its recent augmented reality expansion is bringing medical imaging into completely new clinical settings.
  
• Deciding between public and private cloud? Watch this brief Change Healthcare animation detailing the operational benefits of public cloud hosting for imaging.

• “When will I be back?” is athletes’ first question following a sports-related injury, and New York’s Hospital for Special Surgery increasingly relies on GE Healthcare MR technology for its answers. See how HSS is leveraging GE’s MR scanners, coils, and solutions to achieve more accurate assessments and better patient experiences.