#454 – The Wire

  • Multi-Institutional Data Harvesting: Nature detailed a multi-institutional data harvesting (MIDH) method that was able to analyze imaging utilization and reporting trends across multiple health systems. The researchers repurposed Aidoc’s PE AI and NLP solutions to identify and evaluate CTPA exams performed at 13 US intuitions before and during COVID’s 2020 peak. The analysis revealed COVID-driven declines in CTPA exams (9,806 vs. 12,106) and a significant increase in PE positivity rates (11.6% vs. 9.9%), while showing how AI-based MIDH could help improve our understanding of how imaging is changing.
  • Change’s Stratus Launch: Change Healthcare officially launched its cloud-native and SaaS-based Stratus Imaging PACS to the radiology practice market. This a key milestone for the cloud-centric PACS vendor, coming roughly one year after launching to beta users and two years after acquiring Nucleus.io. It could also prove to be a key milestone for radiology practices’ cloud adoption, noting how well their needs align with the cloud value proposition (security, IT labor, cost structure, speed), and the expanding field of practice-targeted cloud solutions that are now becoming available to them.
  • Cardiac Nuclear-CT Guidelines: The ASNC and three other major nuclear imaging and cardiac CT societies released a new set of guidelines for using hybrid nuclear-CT cardiac imaging. The 45-page guideline goes step-by-step on how to prepare a lab for cardiac SPECT/CT and PET/CT imaging, including new recommended training requirements, models for cross-specialty study interpretation, and suggested solutions for reimbursement challenges.
  • Incepto Adds €27M: French AI platform company Incepto raised €27M (total funding now €32.6M) to fund its expansion across Europe, including the creation of four new subsidiaries in Germany, Spain, Italy, and Portugal. Incepto will also use the funds to support its three strategic objectives: (1) creating a robust panel of AI partner solutions, (2) collaborating with physicians to develop applications that support unmet needs (e.g. knee MRI diagnostic assistance), and (3) supporting imaging organizations’ AI transformations.
  • Incepto’s Partner Expansion: Speaking of Incepto’s goal to expand its partner list, the European AI platform company announced the addition of five new AI partners (IB Lab, Milvue, PAIRE, SmartSoft, Thirona) that expand its portfolio into key clinical areas (emergency care, oncology, orthopedics, rheumatology, and pneumology). These partnerships nearly double Incepto’s current list of seven AI partners, which includes a number of leading AI players (e.g. Subtle, ScreenPoint, Qure.ai, Aidence, Gleamer).
  • Predicting PTC Recurrence with ETE: New research out of China found that clinicians can predict tumor recurrence in patients with papillary thyroid carcinoma (PTC) based on the presence of extrathyroidal extension (ETE) in ultrasound exams. Among 863 patients with PTC (59 w/ recurrence), preoperative lateral cervical lymph node metastasis was found more often in patients with ETE and extensive ETE detected, while microscopic ETE did not predict recurrence. 
  • New York’s Unused X-Rays: The state of New York spent $250M on medical technology during the COVID pandemic, including $86.4M on 1,179 X-ray machines, and all of these devices have gone unused since they were purchased in March 2020. That’s from a Politico report, also revealing that the National Guard is managing New York’s medical technology stockpile in warehouses, and there are no current plans to distribute them or put them to immediate use.  
  • DBT-Guided Biopsy Advantages: Johns Hopkins researchers showed that DBT-guided biopsies of suspicious breast calcifications produce similar clinical results as digital mammography-guided biopsies, but reduce patient discomfort and radiation exposure. Analysis of 1,310 patients’ biopsies (1,006 w/ DBT; 348 w/ DM) found no significant difference in malignancy rates (22% vs. 23%), while the DBT-guided biopsy had much lower procedure times (14.9 vs. 24.7) and total exposures (4.1 vs. 9.1).
  • CT Nomogram Lung Cancer Detection: A new AJR study detailed a nomogram combining clinical data and CT-based radiomics that was able to accurately differentiate pneumonia from pneumonia-type invasive mucinous adenocarcinoma (IMA; a lung cancer). Using data from 314 patients for training, validation, and external testing (208 w/ pneumonia; 106 w/ pneumonia-type IMA), the nomogram differentiated the pneumonia types more accurate than a clinical model and two cardiothoracic radiologists (AUCs: 0.85 vs. 0.71 & 0.70 & 0.67).
  • Konica Minolta’s DDR CSA Clearance: Konica Minolta announced the FDA clearance of its Chiropractic Straight Arm System, featuring the company’s unique dynamic digital radiography (DDR) technology, which allows clinicians to view anatomical movement from rapidly-acquired X-ray images. This might be the first chiropractic product launch we’ve covered, but DDR’s ability to show how bone and tissue interact over time seems to make it a good fit for this space (and the clinical MSK space too).
  • Kaiser Scores Beat BI-RADS: A new study in EJR highlighted how Kaiser Scores (KS) can help clinicians accurately distinguish between malignant and benign breast lesions on contrast-enhanced mammography, avoiding up to 47.9% of unnecessary biopsies. Among 359 patients with breast lesions (231 malignant and 144 benign), the study’s two readers distinguished lesions with higher AUCs using KS compared to BI-RADS (Reader 1: 0.915 vs. 0.821; Reader 2:0.876 vs. 0.842).

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!