“Well, first of all I had no idea who he is. I’ve never watched that show. But secondly, it was like, ‘This is really weird!’ Thirdly, it was just a shrug: ‘We’re not going to do that.’”
Epic CEO, Judy Faulkner, on her three stages of learning about Jim Cramer’s suggestion that Apple should buy Epic. We didn’t think so either.
Imaging Wire Sponsors
- Carestream – Focused on delivering innovation that is life changing – for patients, customers, employees, communities and other stakeholders.
- Focused Ultrasound Foundation – Accelerating the development and adoption of focused ultrasound.
- Medmo – Helping underinsured Americans save on medical scans by connecting them to imaging providers with unfilled schedule time.
- Pocus Systems – A new Point of Care Ultrasound startup, combining a team of POCUS veterans with next-generation technology to disrupt the industry.
The Imaging Wire
The 8 Stages to EI
HIMSS Analytics unveiled its eight-stage adoption model intended to help measure and guide healthcare providers’ progress towards full enterprise imaging adoption. The Digital Imaging Adoption Model for Enterprise Imaging (DIAM – EI) expands DIAM from its previous radiology focus (available since 2016) to support all imaging, with the goals of improving workflows, safety, and care. The development of DIAM EI should help the many healthcare organizations still working towards full enterprise imaging adoption, while also benefiting the enterprise imaging players who could see their target market grow in size and motivation as a result of a prescribed framework like this.
Medivis unveiled its SurgicalAR augmented reality surgical planning platform at HIMSS last week, which allows physicians to access and view 3D medical images/data through a Microsoft HoloLens-based system, achieving what they claim is “the holy grail” of patient rendering. SurgicalAR will initially be used for surgery planning, although Medivis plans to eventually support surgery navigation, with that next phase likely funded by the $2.3 million seed investment that it also announced at HIMSS. Medivis’ technology seems impressive, but it’s perhaps more notable for its contribution towards what could be a key technology shift in radiology, as the last few years brought an influx of studies and forecasts around AR in radiology, followed by an increase in AR-related launches over the last six months.
Hospital Price Growth
Research published in Health Affairs reveals that hospital-based prices grew much faster than physician prices between 2007 and 2014. Hospital price growth over the eight years significantly outpaced physician prices for both inpatient care (+42% vs. +18%) and hospital-based outpatient care (+25% vs. +6%), prompting the group to suggest that efforts to reduce healthcare spending should primarily focus on hospital prices “including antitrust enforcement, administered pricing, the use of reference pricing, and incentivizing . . . more cost-efficient referrals.” The article didn’t touch on the role of hospital consolidation on pricing trends, but industry commentary positioned these two trends as directly related, suggesting that the restricted competition and greater inefficiencies that came from consolidation had a direct hand in hospital pricing growth.
Imaging Financials Remain Positive
The second round of medical imaging company financials from the October-December 2018 period revealed solid performances from Fujifilm, Shimadzu, and Mednax, following a largely positive first round of financials from most other players earlier this month.
- Fujifilm – Fiscal Q3 revenue at Fujifilm increased by 0.9% to ¥627.1 billion ($5.67 billion), due to growth across all segments (except print), while achieving a record ¥74.4 billion ($673m) operating income. Fujifilm’s healthcare business continued its positive run, scoring a 10.1% revenue increase to ¥119.5 billion ($1.08b), while operating profit increased by nearly 300% to ¥15 billion ($135m).
- Shimadzu – Shimadzu didn’t reveal its Oct-December financials, but it did reveal a strong performance from the first nine months of its fiscal year that ends in March (this is common for Shimadzu), with revenue up 5.7% to ¥278 billion ($2.5b) and operating income up 8.7% to ¥27.5 billion ($248m). This growth was helped in part by Shimadzu’s medical business, which posted a 7.9% increase in revenue to ¥49.1 billion ($444m) and a 37.4% increase in operating income to ¥929 million ($8.4m) during the April-December period, due to strong sales in Europe, Japan, and Southeast Asia.
- Mednax – Mednax’s Q4 saw a 2.4% increase in revenue to $933 million, while its net income fell from $136.1 million to $60.2 million due to a 1-time tax benefit in Q4 2017.
Ovarian Cancer Predictor
Researchers at the Imperial College London and the University of Melbourne developed a machine learning algorithm that uses CT images to predict ovarian cancer patients’ survival rates and response to treatments more accurately than current solutions. The researchers used the machine learning tool, TEXLab 2.0, to identify the aggressiveness of tumors in CT scans and tissue samples from 364 women with ovarian cancer, examining each tumor for characteristics that have a major impact on survival (structure, shape, size and genetic makeup) to produce a Radiomic Prognostic Vector (RPV) score. The RPV scores were then compared with current methods for ovarian cancer evaluations/predictions and found that their approach was up to four times more accurate. The team believes that their RPV score may prove to be an important biomarker for predicting response to treatment (e.g. high RPV = chemo resistance, poor surgical outcomes), potentially allowing physicians to make better and more personalized treatment decisions.
- A team of Italian researchers found that contrast enhanced ultrasound (CEUS) performs as well as contrast enhanced CT (CECT) for post-treatment liver cancer monitoring, suggesting that CEUS may become the preferred modality among physicians concerned about radiation exposure from ongoing monitoring. The researchers monitored 24 patients who completed radiation therapy within the previous 1-to-2 months with both CEUS and CECT, finding 40 secondary liver lesions, with a high level of consistency between the two modalities (Cohen K rates: 0.767, 0.881, and 1.00).
- IBM Watson Health announced a 10-year, $50 million investment to fund joint research collaboration projects with Brigham and Women’s Hospital and Vanderbilt University Medical Center intended to advance AI in healthcare. The research will particularly focus on using EHR and claims data to address significant public health issues (e.g. patient safety, precision medicine, and health equity) and explore the physician and patient experience with AI.
- A team of 18 radiologists read, cleaned, and annotated 30,000 frontal chest radiographs from a 112k-image NIH dataset, making the cleaned/labeled studies publicly available to support the future development of pneumonia-detection machine learning algorithms. This is believed to be the first effort of its kind and it’s a great example of the type of one-to-many labeling efforts that could help add efficiency to a very “human” (read: inefficient) step in the machine learning process. That said, this doesn’t appear to be the last effort of its kind, as the RSNA-involved team is seeking volunteers to explore new localization methods using future data sets.
- Strategic Radiology added Asheville, North Carolina-based Asheville Radiology Associates (ARA) to its consortium of independent radiology practices, adding the 38-radiologist practice (plus, 5 vascular surgeons) to a group that now includes 26 private practices and over 1,100 radiologists.
- Zebra Medical Vision received three grants from the Israeli government to provide AI solutions at three of the country’s largest healthcare providers: Ichilov Hospital (supporting radiology worklist prioritization), Maccabi Healthcare Services (providing a breast cancer “second reader” solution to reduce misdiagnosis), and Clalit Health Services (providing a solution for early osteoporosis and heart disease detection).
- New research from Denmark highlights the high diagnostic and cost-saving potential of point-of-care ultrasound (POCUS), once again referring to POCUS as the “the future stethoscope” and outlining a wide range of applications that POCUS could support. The researchers cited some impressive accuracy results from existing POCUS studies, while emphasizing POCUS’ advantages around ease of training, ease of use, procedure efficiency, and ability to eliminate additional higher-cost testing. However, the team did call for POCUS decision-making guidelines and the need for studies on the long-term effects of point-of-care ultrasound exposure to determine POCUS’ role in primary care.
- A team of Finnish researchers found that gadolinium is retained in gliomas, adjacent normal brain tissue, and necrosis at relatively high rates, while finding greater gadolinium retention in patients who received liner GBCAs (vs. macrocyclic GBCAs). The study of 69 patients with primary glioma who underwent contrast-enhanced MRIs (7 with linear GBCAs, 62 with macrocyclic GBCAs) found gadolinium in 57% of glioma tumor samples, 62% of normal brain samples, and 86% of necrotic samples.
- A team of Mount Sinai and USC researchers developed a machine learning framework that can distinguish between low- and high-risk prostate cancer in multiparametric MRI scans (mpMRI) with greater accuracy than any existing approach (specifically vs. PI-RADS v2), potentially making it a valuable decision support tool. The new approach combines ML with radiomics and uses a predictive framework based on 110 radiomic features. They tested it against a set of 54 prostate cancer patients, finding that PI-RADS v2 had a slightly higher AUC than the machine learning classifier (0.73 vs. 0.71), but the new classifier performed better when there were more low-risk cases than high-risk cases (typical mix).
- Polarean Imaging signed a deal to provide its polariser system to the University of British Columbia for use in UBC’s pulmonary imaging research based on hyperpolarised ‘129Xe’ technology.
- Research from China (n = 7,184 patients) found that between 2005 and 2014, CT was the country’s leading modality for lung cancer diagnosis, increasing from 65.8% to 81.4% of all diagnoses, while chest X-ray fell from 50.15% to 30.9% of all scans. Although still rare, chest MRI more than doubled its use for lung cancer diagnosis from 0.73% to 1.96%.
The Resource Wire
–This is sponsored content.
- Focused ultrasound researchers are making unprecedented progress towards developing life-extending treatments for patients with deadly glioblastoma multiforme (GBM) brain tumors, with key blood-brain barrier (BBB)-related breakthroughs at a range of global companies, academic institutions, and research centers.
- Carestream will showcase its entire portfolio of medical imaging and healthcare IT systems at the European Congress of Radiology (ECR) 2019 later this month.
- The POCUS Systems founding team has over 80 years of combined experience in the ultrasound industry.
- Did you know that imaging patients are most likely to no-show for their procedures on Mondays and Saturdays? By partnering with Medmo, imaging centers can keep their schedules full, despite the inevitable Monday no-shows.