|
PharmaAI | Radiologist Swarm
October 25, 2021
|
|
|
|
|
Together with
|
|
|
|
“As digitalization of health increases, so does the potential of leveraging AI for improving care for many diseases.”
|
|
UCB executive Emmanuel Caeymaex on imaging AI’s potential to improve early detection — and to increase demand for treatments made by companies like UCB.
|
|
|
|
Global pharmaceutical company UCB recently licensed its osteoporosis AI technology to MSK AI startup ImageBiopsy Lab, representing an interesting milestone for several emerging AI business models.
The UCB & ImageBiopsy Lab Alliance – ImageBiopsy Lab will use UCB’s BoneBot AI technology to develop and commercialize a tool that screens CT scans for “silent” spinal fractures to identify patients who should be receiving osteoporosis treatments. The new tool will launch by 2023 as part of ImageBiopsy Lab’s ZOO MSK platform.
UCB’s AI Angle – UCB produces an osteoporosis drug that would be prescribed far more often if detection rates improve (over 2/3 of vertebral fractures are currently undiagnosed). That’s why UCB developed and launched BoneBot AI in 2019 and it’s why the pharma giant is now working with ImageBiopsy Lab to bring it into clinical use.
The PharmaAI Trend – We’re seeing a growing trend of drug and device companies working with AI developers to help increase treatment demand. The list is getting pretty long, including quite a few PharmaAI alliances targeting lung cancer treatment (Aidence & AstraZeneca, Qure.ai & AstraZeneca, Huma & Bayer, Optellum & J&J) and a diverse set of AI alliances with medical device companies (Imbio & Olympus for emphysema, Aidoc & Inari for PE, Viz.ai & Medtronic for stroke).
The Population Health AI Trend – ImageBiopsy Lab’s BoneBot AI licensing is also a sign of AI’s growing momentum in population health, following increased interest from academia and major commercial efforts from Cleerly (cardiac screening) and Zebra Medical Vision (cardiac and osteoporosis screening… so far). This alliance also introduces a new type of population health AI beneficiary (pharma companies), in addition to risk holders and patients.
The Takeaway – ImageBiopsy Lab and UCB’s new alliance didn’t get a lot of media attention last week, but it tells an interesting story about how AI business models are evolving beyond triage, and how those changes are bringing some of healthcare’s biggest names into the imaging AI arena.
|
|
|
CVIS’ Cloud Advantages
This Diagnostic and Interventional Cardiology article details the unique advantages of cloud-based CVIS systems (off-property access, team collaboration), with insights from one Mississippi-based cardiologist on the benefits of Fujifilm Healthcare’s VidiStar CVIS.
|
|
- Swarm AI Advantage: A new study from UCSF and “swarm AI” startup Unanimous AI showed how swarm intelligence (combining AI w/ human team analysis) can significantly improve radiologists’ consensus diagnoses. Two blinded cohorts (3 rads & 5 radres) used Unanimous AI’s Swarm platform to grade meniscal lesions on knee MR exams, achieving significantly higher inter-reader reliability (+23% & +30%) and specificity (up to +50%) with their swarm-based consensus compared to their majority votes. The swarm consensus votes also outperformed individual votes and predictions by a standalone AI tool.
- OIA Finds a Forever Home: Outpatient Imaging Affiliates (OIA) has a new long-term parent company, after selling to “permanent capital” firm The Cranemere Group. This represents a solid exit for OIA’s former parent company, ICV partners ($400m, ~12.5x to 16x EBITA), which acquired OIA in 2018 and quickly grew it over the next three years (36 to 56 imaging centers, doubled earnings). It will also provide OIA with far more stability, given The Cranemere Group’s long-term approach to the companies that it invests in.
- NHS’ Insufficient Imaging: The Guardian reported that about a third of NHS England trusts have at least one CT scanner and/or one MRI system older than 10 years (27% & 34.5% of trusts), while some trusts are using X-ray systems from the 1970s. Even if the NHS replaced these “technically obsolete” scanners, a separate RCR report revealed that 10% of UK-based radiologist roles are currently unfilled and the country could have a 6k radiologist shortage by 2030 unless changes are made.
- Radiology’s Commercial Premium: A new Urban Institute study found that commercial plan radiology reimbursements are 180% higher than Medicare rates, giving radiology one of the highest premiums across the major specialties (below neurosurgery 220%, emergency 250%, and anesthesia 330%). These premiums have helped radiologists, but also make radiology more exposed to future policies looking to right-size private coverage rates.
- Knee MRI AI: AI startup Owkin developed a “weakly supervised” deep learning method (uses image patterns, not labels) that can predict knee osteoarthritis progression using MRI exams and clinical variables (e.g. BMI). The team developed the model using data from 3,268 patients, predicting joint space narrowing with a 65% ROC AUC, notably above trained radiologists’ 59% ROC AUC.
- Walgreens Doubles Down on Primary Care: Walgreens announced a $5.2b investment in VillageMD, giving it a 63% stake in the primary care company (up from ~30) and supporting its plans to open least 1k “Village Medical at Walgreens” locations by 2027 (up from 57). Many Walgreen Clinics will reportedly have on-site X-ray and ultrasound systems, although the recent shift we’re seeing in how/where health services are delivered could have a much greater imaging impact.
- Two Sides to Mammography AI: A new study review published in Radiology Journal revealed that mammography AI tools might be more accurate than other recent studies suggest. The meta-analysis of 15 studies (185k total cases, 3 countries) found that AI tools exceeded radiologists’ pooled sensitivity (75.4% vs. 73%), specificity (90.6% vs. 88.6%), and AUCs (0.89 vs. 0.85). However, the authors warned that this analysis might overestimate AI’s performance (small sample, included sub-standard readers, only used top-performing algorithms) and encouraged future mammography AI researchers to use practices that better assess real world performance (randomized, prospective, independent, etc.).
- AI Visualize’s Patent Lawsuit: AI Visualize filed a patent infringement lawsuit against Mach7 and Nuance, alleging that Mach7’s eUnity diagnostic viewer and Nuance’s PowerScribe platform (which integrates with eUnity) infringes on four AI Visualize patents related to “fast access to advanced visualization of medical scans using a dedicated web portal.”
- Breast Cancer Screening Equity: A new Annals of Internal Medicine study found that if Black women begin biannual mammography screening at 40 years old it would reduce their breast cancer mortality disparity versus White women by 57% (… only if White women continue to begin screening at 50yrs). However, a responding editorial countered that race-based screening would have unintended consequences (more radiation & false positives) and proposed using more personalized risk assessments.
- Multimodal LC Risk: Vanderbilt scientists developed a colearning predictive model that uses chest CTs and clinical data elements (CDEs; e.g. smoking & cancer history) to automatically estimate patient cancer risk. The team developed and assessed the colearning model using data from 23.5k patients in the NLST database and tested it against data from 147 Vanderbilt patients, estimating cancer risk with a 0.91 AUC, above a pair of CDE-only and CT-only models (0.59 & 0.88 AUCs).
- IT Priorities: A Frost & Sullivan survey of 349 healthcare IT decision-makers found that over 50% of hospitals are accelerating IT investments to address pandemic-related challenges, largely prioritizing patient experience and operational efficiencies. Those priorities will specifically lead to increased investments in remote patient monitoring, telehealth, AI for enterprise / patient communications, and data visualization. The report didn’t specifically mention imaging, but imaging teams and vendors will definitely have to adapt to many of these trends (e.g. more remote / tele patients & greater patient experience / communication expectations).
|
|
UCSD’s Case for Arterys Lung AI
Check out this UCSD lung nodule detection study detailing how Arterys Lung AI drove a “clinically meaningful and statistically significant increase in sensitivity,” without changing reading time.
|
|
- See how United Imaging’s new uCT ATLAS combines advancements in image quality, patient comfort, and operator efficiency to bring you one step closer to your masterpiece.
- This Riverain Technologies case study details how Einstein Medical Center adopted ClearRead CT enterprise-wide (all 13 CT scanners) and how the solution allowed Einstein radiologists to identify small nodules faster and more reliably.
- Despite significant interest, there’s still confusion about the value of imaging AI. This Blackford Analysis white paper explores the key cost considerations and ROI factors that radiology groups can use to figure out how to make AI valuable for them.
- See how Einstein Healthcare Network reduced its syringe expenses, enhanced its syringe loading, and improved its contrast documentation when it upgraded to Bayer Radiology’s MEDRAD Stellant FLEX CT Injection System.
- After more than 15 years of development, the world’s first photon-counting system is here to redefine CT. Register now and join Siemens Healthineers at the launch event on November 18 to be part of this quantum leap forward in technology.
|
|
|
|
|
