|
Better Prostate MRI with AI | No Republican Residents?
August 8, 2024
|
|
|
|
|
Together with
|
|
|
|
“Though wrong, these policies will be moot in short order as radiology is certain to become a domain of AI usurping the need for trained physicians. These docs can then enjoy discussing the best brand of blue hair dye with their GP patients as they administer their twenty second covid booster.”
|
|
A commenter to an article in National Review on how extracurricular activities can affect radiology residency applications.
|
|
|
|
A homegrown AI algorithm was able to detect clinically significant prostate cancer on MRI scans with the same accuracy as experienced radiologists. In a new study in Radiology, researchers say the algorithm could improve radiologists’ ability to detect prostate cancer on MRI, with fewer false positives.
In past issues of The Imaging Wire, we’ve discussed the need to improve on existing tools like PSA tests to make prostate cancer screening more precise with fewer false positives and less need for patient work-up.
- Adding MRI to prostate screening protocols is a step forward, but MRI is an expensive technology that requires experienced radiologists to interpret.
Could AI help? In the new study, researchers tested a deep learning algorithm developed at the Mayo Clinic to detect clinically significant prostate cancer on multiparametric (mpMRI) scans.
- In an interesting wrinkle, the Mayo algorithm does not indicate tumor location, so a second algorithm – called Grad-CAM – was employed to localize tumors.
The Mayo algorithm was trained on a population of 5k patients with a cancer prevalence similar to a screening population, then tested in an external test set of 204 patients, finding …
- No statistically significant difference in performance between the Mayo algorithm and radiologists based on AUC (0.86 vs. 0.84, p=0.68)
- The highest AUC was with the combination of AI and radiologists (0.89, p<0.001)
- The Grad-CAM algorithm was accurate in localizing 56 of 58 true-positive exams
An editorial noted that the study employed the Mayo algorithm on multiparametric MRI exams.
- Prostate cancer imaging is moving from mpMRI toward biparametric MRI (bpMRI) due to its faster scan times and lack of contrast, and if validated on bpMRI, AI’s impact could be even more dramatic.
The Takeaway
The current study illustrates the exciting developments underway to make prostate imaging more accurate and easier to perform. They also support the technology evolution that could one day make prostate cancer screening a more widely accepted test.
|
|
|
How a New PACS Helped Regional Medical Imaging
Regional Medical Imaging in Michigan was looking for a partner to help them streamline PACS workflows for breast imaging. Learn how Merge by Merative helped them break up the logjam created by different PACS silos to create a workflow tailored to their needs.
|
|
Turning Medical Imaging into Great Medicine
TeraRecon is continually innovating to bring the latest in advanced visualization capabilities to radiology. Intuition 4.7 includes a variety of new features, from structural heart workflow to the ability to trigger AI on demand. Learn more about them on this page.
|
|
- No Republican Residents? A study in JACR on radiology residency admissions is drawing fire from conservative publication National Review. Researchers surveyed 244 participants running admissions at 30 U.S. radiology programs, finding evaluators with liberal views preferred medical student applicants who participated in left-leaning extracurricular activities like LGBTQ pride groups while downgrading applicants from conservative groups like Young Republicans. A similar bias was not found with conservative evaluators (perhaps due to a smaller sample size). The good news? National Review called JACR “a major medical journal.”
- FDA Finalizes TCET Guidance: The FDA this week published the final guidance on its new Transitional Coverage for Emerging Technologies (TCET) regulatory pathway. TCET is voluntary and is designed to streamline the process for getting breakthrough medical devices to market, but a major potential downside is that rather than automatically allowing Medicare coverage for authorized devices it requires vendors to go through the cumbersome National Coverage Determination and Coverage with Evidence Development processes. This may limit the number of vendors choosing the TCET pathway.
- CT Screening for Diabetes: Opportunistic screening is in the news again with a study in Radiology in which South Korean researchers analyzed PET/CT images of 32.2k people who got health screenings as part of employee health evaluations. They used Medical IP’s FDA-cleared DeepCatch AI algorithm to analyze CT images for measurements of skeletal muscle, subcutaneous fat, and visceral fat, finding type 2 diabetes mellitus prevalence of 6% on baseline scans and 9% over seven-year follow-up. AI worked better than traditional diabetes prediction methods.
- 4DMedical Lands Vanderbilt: Respiratory imaging company 4DMedical has installed its XV scanner for measuring lung function at Vanderbilt University Institute of Imaging Science (VUIIS). XV is a fluoroscopy-based system that produces color-coded maps of air flow in the lungs. VUIIS will use XV initially for research, but clinicians believe it could play a role in diagnosing patients with hard-to-spot respiratory conditions such as those found in military veterans with exposure to burn pits.
- Costs of Cancer Screening: Screening for five common types of cancer in the U.S. costs $43B a year in the U.S., with colon cancer screening making up 64% of this expense. In all, private insurance paid for 88% of screening, Medicare 8.5%, and Medicaid 3.2%. An accompanying editorial suggested that instead of screening the general population, resources could instead be allocated to preventive care for high-risk racial and ethnic groups – an approach they acknowledged that could have ethical and public policy challenges, however.
- Bayer Partners with Alara: Bayer and radiation measurement firm Alara Imaging are collaborating to help imaging providers monitor and track radiation dose during CT scans. Bayer already offers the Radimetrics dose-management software, and Alara’s solutions for tracking electronic clinical quality measures (eCQMs) will be available through Bayer’s Calantic Digital Solutions platform, enabling providers to comply with CMS quality measures for radiation dose calculation and reporting. Recent studies indicate that radiation dose can vary widely, even between CT scanners from the same manufacturer.
- No Lung Screening for Never-Smokers? People who have never smoked probably shouldn’t get CT lung cancer screening. That’s the advice of an article in Journal of Thoracic Oncology that reviews recent lung screening research, particularly a study in Taiwan that screened people who never smoked or were only light smokers – but had a family history of lung cancer. That study saw cancer detection rates double those of high-risk people meeting USPSTF criteria, but JTO authors don’t believe it’s worth the harms of overdiagnosis and false positives.
- Patient-Friendly AI Echo Reporting: A NYU team used Chat GPT4 to automatically – and relatively accurately – produce patient-friendly echo report explanations. The researchers produced 100 GPT4-generated echo report explanations and had them reviewed by five echocardiographers, who largely found that the explanations were “all true” (84%) and included “all of the important information” (76%), while they found none of the missing information to be “potentially dangerous.”
- Rating AI for TB: Lunit’s AI algorithm for tuberculosis detection topped 11 other commercially available AI applications in a new study in Lancet Digital Health. The algorithms were used to analyze chest X-rays from 774 people suspected of having TB in South Africa, targeting 90% sensitivity and 70% specificity. The top-performing algorithms by AUC were Lunit’s Insight CXR (0.902), Nexus’ Nexus CXR (0.897), and Qure.ai’s qXR (0.878). The authors, however, cautioned against head-to-head performance comparisons due to the study’s small sample size.
- AI Speeds Stroke Diagnosis: Meanwhile, researchers in PLOS Global Public Health used Qure.ai’s qER AI algorithm to analyze non-contrast CT brain scans for stroke at a hospital in India. In 174 patients who were scanned, AI reduced median time to intervention with thrombolysis by 27% (58.5 vs. 80 minutes) and improved intervention rates within 30 minutes after imaging (26% vs. 4%). AI’s sensitivity and specificity for detecting intracranial hemorrhage was 0.89 and 0.99, respectively. The findings show AI’s impact in resource-challenged settings.
- RSNA Gets $2M for Global Imaging: RSNA has received a five-year $2M grant to improve access to radiology in low- and middle-income countries around the world. The funds come from the U.S. DOE’s National Nuclear Security Administration (NNSA) and are intended to improve patient care and peaceful uses of nuclear energy. This is RSNA’s second NNSA grant – a $1M funding paid for the deployment of an RSNA Global Learning Center in Tanzania. RSNA hopes to open a second center with the new funds.
- Breast Ultrasound Better in High-Risk Women: Researchers found ultrasound to be a good tool for supplemental breast cancer screening of women with heterogeneously or extremely dense tissue – particularly those at high risk. In a study of 26.5k women in Radiology, breast ultrasound’s overall cancer detection rate was 2 cancers per 1k exams, but the CDR was far higher in high- versus low- or average-risk women (5.5 vs. 1.3). The results should help guide recommendations using ultrasound for supplemental breast screening.
- Rad Therapy’s Breast Cancer Benefits: Administering radiation therapy to women after breast cancer can prevent recurrence for up to 10 years. In a new paper in The Lancet Oncology, researchers studied 585 Scottish women with early-stage breast cancer; after 10 years, those who got radiation therapy had lower cancer recurrence in the same location (16% vs. 36%). The radiation therapy group had fewer breast cancer deaths (37% vs. 46%), but deaths from other cancers were higher (20% vs. 11%) and overall survival was similar (19.2 vs. 18.7 years).
|
|
Smart Bytes on Cardiology Workflows
Check out Smart Bytes, a new series of webcasts giving you bite-sized insights for smarter imaging from AGFA HealthCare. At 12:30 ET on August 15 learn about improved cardiology workflows using cutting-edge AI technology.
|
|
AI-Empowered CT Workflow
CT systems from United Imaging are designed for high image quality and low dose, and their AI-empowered workflow enables fast and reproducible positioning, helping you image patients with confidence. Learn more on this page.
|
|
Best in KLAS for Image Exchange
With over 15k connected facilities and a dedicated outreach team building new connections every day, it’s no wonder Nuance PowerShare Image Sharing was named #1 Best in KLAS 2024 for image exchange. Learn more about this award-winning solution.
|
|
- Improving Patient Outcomes in Lung Cancer: Early detection is key to improving treatment pathways and patient outcomes in lung cancer. Learn more facts about lung cancer screening on this resource page from Riverain Technologies.
- Data Quality Issues in Healthcare: Data quality issues in healthcare are common challenges that can impact patient care, research, and overall healthcare management. Learn how data standardization solutions like Enlitic’s ENDEX can help by ensuring complete, accurate information.
- AI for Incidental Osteoporosis Findings: Looking to learn more about how AI is used for incidental osteoporosis findings? Watch this on-demand webinar from Blackford in which clinical leaders discuss the benefits of preventive care and share their real-life experiences.
- In Their Own Words: What are radiology professionals saying about the Visage 7 enterprise imaging platform? Visage Imaging has curated a select group of video interviews with Visage 7 users so you can hear what they think in their own words.
- Radiology Academy – Medical Education on Demand: Visit Calantic Radiology Academy by Bayer, where you’ll find the latest keynotes and symposium sessions on the use of artificial intelligence in radiology, ranging from challenges facing AI to bias in machine learning.
- Keep Patients Engaged with Your Healthcare System: After using PocketHealth, 94% of patients are more confident about their healthcare experience. Learn how to increase follow-up adherence and improve patient experience with PocketHealth MyCare Navigator.
- AI for Detecting Post-Traumatic Bone Fractures: Check out this new research paper to learn how Gleamer’s BoneView AI solution enhanced radiologists’ ability to detect post-traumatic bone fractures by identifying cases without abnormalities, translating into improved patient care.
- Reimagine What’s Next with Intelerad: Every journey in healthcare is about seeking and providing answers. Intelerad is sharing a refined brand identity that encapsulates its commitment to simplifying the search for answers for both clinicians and patients. Learn more about what’s next today.
- Gain Clarity at Speed in MRI: Acquire images faster per MRI machine per day with STAGE from SpinTech MRI. Learn how STAGE reduces the time for MRI brain protocols by 30% for your most common exams.
- Creating a Theranostics Center of Excellence: See how Siemens Healthineers can work with you to create a radiopharmaceutical therapy center of excellence that’s built around your needs by providing consulting and design services that put you on the path to success.
- Stop Shipping Discs! By pivoting to a 100% digital fulfillment model for patient images and records, you can improve their experience while significantly reducing labor and shipping costs. Find out how on this page from Clearpath.
- Preparing for the Future of Enterprise Imaging: Check out this white paper from Optum to learn what you need to know when moving your enterprise imaging to the cloud. Learn how to assess the various approaches and develop a strategy that works for you.
- Get to Know DeepHealth: What’s the latest from DeepHealth? In this episode of the Imaging Wire Show, we talked to COO/CTO Sham Sokka about the company’s recent launch and their take on the value AI provides to radiology, especially for screening.
- Advances in Fully Automated Echocardiography: Echocardiography is a pillar of cardiac imaging, but it is operator-dependent and time-consuming to perform. Learn how new echo AI-based software from Us2.ai can automate image acquisition and calculate dozens of measurements that previously had to be performed manually.
|
|
|
|
|
