“This clearly needs to change.”
King’s College London professor of cancer imaging, Vicky Goh, on the UK’s widespread use of X-ray to diagnose myeloma, even though MRI is far more effective.
Imaging Wire Sponsors
- Arterys – Reinventing imaging so you can practice better and faster.
- 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.
- Zebra Medical Vision – Transforming patient care with the power of AI.
The Imaging Wire
Patient-Aided Interpretations
A new MGH study revealed that info from patient pain questionnaires might help improve lumbar MRI interpretation accuracy, and potentially treatment effectiveness. Here are some details:
- Background – Lumbar spine MRI interpretations have high error rates, which could be improved with symptom information that’s often available to treating physicians, but not always in the reading room.
- The Study – The researchers had 120 patients who were scheduled for lumbar spine MRIs and pain management injections fill out electronic symptom questionnaires. They then had six radiologists diagnose the patients’ pain generators (type, level, side), both with and without the questionnaire symptom information.
- Reference Standard Results – When the six radiologists interpreted MRI exams using the patient symptom information, they achieved “almost perfect” alignment (κ=0.82-0.90) with the “reference standard” diagnoses from the radiologists who actually performed the injections (w/ patient histories and symptom info from patient interviews). Meanwhile, interpretations performed without the questionnaire info achieved only “fair to moderate” alignment (κ=0.28-.51).
- Interreading Results – As you might expect, the six radiologists’ interreading agreements were also “almost perfect” when they all leveraged the questionnaire symptom information (κ=0.82-0.90), but only “moderate” without symptom information (κ=0.42-0.56).
- The Takeaway – The idea that patient information helps support diagnosis is pretty straightforward, but these comparisons with and without patient information is notable. This is also the latest in a long list of studies confirming that clinical and patient data improves image interpretation.
The Wire
- Feds Probing Healthcare M&A: The U.S. Federal Trade Commission just launched a study into the impact of physician group and healthcare facility M&A. The FTC will leverage 2015-2020 claims data from six top health insurers to understand how consolidation affects competition and “the proper functioning of healthcare markets,” using the study results to support policy decisions. As we learned from the surprise billing fight, policy change can take a while, but there’s a solid chance that the probe reveals that consolidation stifles competition.
- EMs & Rads Align: Radiologists and emergency medicine physicians are generally on the same page when it comes to how they view key radiology report phrases. That’s at least the case at Tacoma’s Madigan Army Medical Center, where a survey of 42 EM residents/attendings and 32 radiology residents/attendings revealed a high level of agreement on the diagnostic probability of 17 out of 18 common phrases. The only phrase they weren’t aligned on was “compatible with,” which the radiologists associated with far higher diagnostic probability than the EM physicians.
- Viz RECRUIT: Viz.ai just expanded into the clinical trial recruitment business with the launch of its new Viz RECRUIT software, which detects a range of suspected diseases (e.g. brain hemorrhages, tumors, cardiac structural abnormalities) and automatically notifies research teams about potentially eligible trial patients. Viz RECRUIT also features tools to support trial enrollment and related communications.
- WBMRI’s Myeloma Advantages: Whole body MRI outperforms 18F-FDG PET/CT for multiple myeloma diagnosis, potentially allowing earlier treatment. That’s from a King’s College London study (n = 46 patients) that found WBMRI identified bone lesions with far higher sensitivity than 18F-FDG PET/CT (91.3% vs. 69.6%), while the combination of clinical data with WBMRI identified more patients for treatment than combining clinical data with 18F-FDG PET/CT (93.5% vs. 87%).
- Siemens’ DR Approvals: Siemens Healthineers announced the FDA approval of its MULTIX Impact C ceiling-mounted DR system and MULTIX Impact VA20 floor-mounted DR system. The new systems share many of the same features (wireless detectors, motorized tube heads, floating flat table, in-room touch UI) and are clearly highlighted by Siemens’ new MyExam Companion intelligent UI concept (guides/automates rad tech operations).
- Sleep & Accuracy: New research out of Australia (n = 133 rads) reveals that less experienced radiologists’ diagnostic accuracy is significantly affected by how long they sleep and how long it’s been since they’ve slept, while experienced rads aren’t affected by these factors. In a test set of 60 digital mammograms, the less experienced rads identified lesions with much lower ROC values when they slept less than 6 hours the night before compared to more than 6 hours of sleep (0.72 vs. 0.77). The less experienced rads also achieved significantly lower lesion sensitivity early in their day (awake for <2 hours, or 4-6 hours) than later in their day (awake for >8 hours).
- GE & EXACT’s ACT Probe: GE Healthcare and EXACT Therapeutics are working together to develop an ultrasound probe for Acoustic Cluster Therapy (ACT), an emerging ultrasound-activated therapeutics delivery method. The probe will initially support a range of ACT clinical trials.
- CCA’s Practice Impact: Healthcare Administrative Partners just shared a detailed review of how the “Consolidated Appropriations Act, 2021” (CAA) will impact radiology practices. Here are some of the main takeaways: 1) Medicare reimbursements for radiology services will fall by 4% (originally 10%); 2) The 2% Medicare fee-for-service withholding will remain suspended through March 31; 3) Surprise billing payment limits will start in 2022; 4) There’s a new PPP program with new rules; 5) The CARES act family leave and employee retention credits are extended.
- Validating Knee MRI AI: A UCSF team developed a deep learning system that can effectively detect and assess lesion severity in knee MRI exams (cartilage, bone marrow, meniscus, ACL), suggesting that AI tools might be able to support these tasks in the real world. Using a set of 1,435 knee MRIs, the model was able to identify lesions with 70%–88% sensitivity and 85%–89% specificity, while grading lesion severity with 0.83–0.93 AUCs. Using a separate external dataset, the model improved four physicians’ severity grading agreements (2 attendings, 2 trainees) in 10 out of 16 comparisons.
- Canon Aquilion Exceed LB CT’s FDA: Canon Medical announced the FDA approval of its Aquilion Exceed LB CT radiation therapy planning system, highlighting its AiCE reconstruction technology and industry-leading bore opening (90cm), field of view (90cm), and detector coverage (4cm).
- Defining BC Risk: A new NEJM study identified the exact genetic variants associated with higher breast cancer risks, potentially representing a major step in risk-based breast cancer screening and management. The study identified germline pathogenic variants in 28 cancer-predisposition genes among 32,247 women with breast cancer and 32,544 unaffected women (controls), finding that 12 variants were associated with a higher risk of breast cancer (led by BRCA1 & BRCA2’s 7.62 odds ratio), while 16 variants had no association with higher breast cancer risks.
- Cleveland Clinic Billing Suit, Not Dismissed: An Ohio State Court denied Cleveland Clinic’s motion to dismiss a lawsuit filed by a patient who underwent a medical imaging exam, but never received an upfront cost estimate or receipts for her ongoing payments. The woman eventually stopped paying, prompting Cleveland Clinic to hire a debt collector, which led to her lawsuit alleging that the clinic violated the Ohio Consumer Sales Practices Act (requires estimates and receipts) and filing for class action status. The court denied Cleveland Clinic’s four different motions to dismiss the lawsuit as well as its motion to block the class action, opening the lawsuit to other patients with similar billing experiences.
The Resource Wire
– This is sponsored content.
- Watch this recorded webinar from Healthcare Administrative Partners where they examine how the 2021 Medicare Physician Fee Schedule (MPFS) and Quality Payment Program (QPP) final rules will impact radiologists.
- Check out how Zebra Med’s Chest Solution prioritizes chest X-ray worklists, while providing specific insights into each study.
- These two independent Phase III trials provided new evidence supporting the accuracy and effectiveness of using Bayer’s Gadavist contrast agent for cardiac MRI.
- Check out this interview with Arterys’ Maya Khalife and Stephane Boyer, discussing the future of medical imaging AI.
- This Hitachi blog details how a complete CVIS solution drives efficiencies and increases productivity.
- Learn how Hemet Global Medical Center improved efficiencies, safety, and security when it moved to the Nuance PowerShare Network and #ditchedthedisk.
- Siemens Healthineers’ FDA-approved AI-Rad Companion Organs RT AI-based software automates the contouring process for organs at risk during radiation therapy planning.
- 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%.
- Check out GE Healthcare’s upcoming webinar where a panel of breast health leaders discuss how AI is changing the game in breast imaging across modalities.
