- Image-Guided Bronchoscopy, Better with Experience: Using cone-beam CT and augmented fluoroscopy to guide bronchoscopy navigation could improve lung cancer diagnosis and reduce complication risks, especially as clinical teams gain more experience. Dutch researchers analyzed 238 consecutive patients who underwent CBCT/AF-guided navigation bronchoscopy, finding that diagnostic accuracy increased (72% to 90%) and radiation exposure declined (47.5 Gy·cm2 to 25.4 Gy·cm2) over the course of the study period.
- DIY X-Ray: A California-based YouTuber became the focus of the DIY tech community after he built his own fully-functional home X-ray system. After receiving a $69k medical bill (most of it was covered), the former engineer sourced the required parts over several months (a 20yr-old dental X-ray head, power source, DSLR camera, intensifying screen) and then built the X-ray system in one day.
- PSMA-PET for HCC: German researchers found that 68Ga-PSMA-11 PET/CT is more effective than CT for detecting and managing hepatocellular carcinoma (HCC, the most common liver cancer), and could deserve a role in HCC staging. Three blinded readers interpreted 68Ga-PSMA-11 PET/CT scans from 40 HCC patients, finding that PSMA-PET re-staged 10 patients (8 upstaged, 2 downstaged) and led to 19 patient management changes compared to CT.
- Medtech Chips: The Wall Street Journal detailed how the global microchip shortage is impacting the medical device supply chain (including imaging), and how medtech makers are taking advantage of a unique leverage point to get the chips they need. Because other industries order far more microchips (and have far more leverage), medical device manufacturers are increasingly forced to emphasize that their products “save lives” when asking microchip makers for priority access to shipments.
- AI Jitters: A new Clinical Imaging study revealed that one-sixth of US medical students who would have chosen radiology as their first choice did not do so because of AI (n = 463, 32 schools, 2020 survey), while half of students who listed radiology among their top-three choices are concerned about AI. Like previous studies, students who don’t fully understand AI are most concerned about its impact.
- MGH Says No to Aduhelm: MGH joined the growing list of large/influential health systems to decide not to provide/prescribe Biogen’s Aduhelm Alzheimer’s treatment, following similar moves from Cleveland Clinic, Mount Sinai, and the VA. Aduhelm’s FDA clearance in June briefly created optimism for Alzheimer’s advocates and imaging vendors (it requires brain PET for diagnosis and ongoing MRIs), but the new drug has been surrounded by consistent criticism about its effectiveness and costliness since then. MGH will reevaluate its Aduhelm policy when more data becomes available.
- Samsung V8: Samsung Medison unveiled its first new ultrasound system in nearly two years, highlighting its new V8 ultrasound’s clinical versatility. Samsung’s new flagship ultrasound comes with a suite of diagnostic assistant solutions for obstetrics (blood flow), MSK (nerve tracking), and radiology (Shearwave Elastography & MRI alignment).
- RadiLens & AUHS: Georgia-based AI startup RadiLens will collaborate with the Augusta University Health System to develop an NLP-driven system intended to improve follow-up management and overall radiology workflow efficiency. RadiLens’ technology analyzes radiology orders and reports for follow-up indications, unscheduled exams, and closed-loop communications, and it appears that this new collaboration will build upon its technology.
- One More Surprise: The ACR and other provider groups came out strongly against a late change to the upcoming surprise medical billing rule that could favor insurers when cases go to arbitration. The No Surprises Act’s interim final rule includes a change to the ‘independent dispute resolution’ process that makes health plans’ calculated ‘qualifying payment amount’ the primary factor considered when surprise billing cases go to arbitration (these are often lower than real-world payment rates).
- BMRI Triage AI: A team of Dutch researchers developed a breast MRI DL triage tool that was able to eliminate 40% of negative screening exams without missing any cancers. The researchers tested the DL triage tool with 4,581 MRI exams from women with extremely dense breasts (9,162 breasts, 838 w/ BI-RADS 2-5 lesions, 77 malignant). At 100% sensitivity for malignant lesions, the DL tool dismissed 9.3% of the MRI exams, removing 39.7% of normal exams from radiologist workflows and assigning all “nonnormal” exams to the radiologists.
- NC MRI Loophole: Major North Carolina imaging practice Wake Radiology lost its legal case to block an orthopedic clinic from operating a whole-body MRI system. Wake Radiology alleged that Bone and Joint Surgery Clinic took advantage of an advanced imaging Certificate-of-Need (CON) loophole by upgrading its extremity-only MRI CON to a whole-body MRI CON when it replaced its scanner. The NC Appeals Court rejected the case because it’s not responsible for defining/correcting loopholes (that’s up to the legislature). Wake isn’t the only imaging practice frustrated with NC’s CON laws.
- CT VR: Surgical templating startup Kinomatic unveiled its new VR platform that converts CT scans into Oculus Quest-viewable 3D models used for orthopedic surgery simulations, starting with knee and hip arthroplasty.
