“When everybody is in charge, nobody is in charge.”
An anonymous Chief Information Security Officer on what happens when the ownership of medical device security isn’t clear.
The Imaging Wire is Brought to You By
- 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
Keep these companies in mind each time you enjoy The Imaging Wire. Check them out and see how they’re driving our industry forward, and shoot them a note if you’re interested in learning more. They’re all great companies run by solid people.
The Imaging Wire
The 96/18 Rule
Eighteen percent of healthcare IT executives say their organization’s medical devices have been the target of a malware or ransomware attack… in the last 18 months! This is from a report by the College of Healthcare Information Management Executives (CHIME) and Klas Research (n=148), which also revealed that only 39% of the IT execs were “confident” or “very confident” that their cybersecurity strategy protects patient safety and prevents disruptions in care. An overwhelming 96% suggested that “manufacturer-related factors” were the root cause of their device vulnerability (dated operating systems, unavailable patches, etc.) compared to 68% blaming internal “organizational causes.” This isn’t surprising given the ongoing debate over responsibility for legacy devices. However, the need to protect against malware or ransomware attacks is far less debatable (it’s important).
Google DeepMind Looks Internationally to Fight Density Bias
Google DeepMind is collecting breast images from different countries to train a new breast cancer screening algorithm that is less prone to regional breast density biases. DeepMind just gained access to 30,000 mammograms from Jikei University Hospital in Tokyo and was promised future access to 30,000 breast ultrasounds and 3,000 breast MRIs from the hospital, helping to train its dataset with images from a country known for dense breasts. These images will be combined with DeepMind’s 7,500 existing mammograms from Cancer Research UK Imperial Centre at Imperial College London and other future image databases from other regions.
Radiopharmaceuticals’ Fight for Drug Status
To the applause of MITA, congress introduced the Medicare Diagnostic Radiopharmaceutical Payment Equity Act of 2018, seeking to change the CMS status of radiopharmaceuticals to a drug (vs. a supply) that will therefore be paid separately in Medicare outpatient settings (vs. bundled with the cost of the scan). With the current bundling method, the “cost of the diagnostic radiopharmaceutical (sometimes) exceeds the entire packaged amount of reimbursement, creating the potential for hospitals to lose money every time they perform the procedure.” This of course makes providers hesitate before performing certain imaging procedures and drags down overall radiopharmaceutical and imaging demand.
AI Still An Aid
Boston area researchers tested Qure.ai’s deep learning (DL) algorithm on chest x-rays, finding that it can effectively help radiologists, but is “unlikely to replace radiologists.” The researchers tested 874 de-identified chest radiographs (for pulmonary opacities, pleural effusions, hilar prominence and enlarged cardiac silhouette), pitting the Qure.ai solution against two trained radiologists to establish a standard of reference (SOR) and again against four test radiologists. The results showed no statistical difference between DL and SOR for the above abnormalities, but the DL algorithm performed worse than the test radiologists when assessing changes in pulmonary opacities and when chest wall implanted devices were present in tests for pulmonary and hilar abnormalities.
NTT Data’s AI Investment
Japanese IT service provider NTT Data made an unspecified investment for about a 10% stake in US/Indian medical imaging AI startup DeepTek, with plans to add DeepTek’s AI solutions to NTT’s United Clinical Archive platform and expand the solution globally in 2019. DeepTek’s AI-based radiology decision support system reportedly cuts MRI and CT diagnosis time in half and in a recent proof of concept test, the solution detected emphysema with over three-times greater accuracy than a non-AI method. NTT Data’s United Clinical Archive has a strong presence and the company is certainly capable of expanding the DeepTek solution across its global territories.
The Wire
- IMRIS, Deerfield Imaging Inc. and Siemens Healthineers are making their relationship exclusive. IMRIS agreed to only use Siemens MRI, CT, and angiography systems in its IMRIS Surgical Theater neurosurgery suite (it will still work with other companies for other devices) and Siemens will distribute IMRIS’ solutions through its direct sales force and to its customer base.
- Siemens Healthineers announced the FDA clearance of the MAGNETOM Sola 1.5T MRI Scanner, releasing the new system to the US healthcare system seven months after it was first announced and about six weeks after it was cleared by Health Canada. The 70-cm MAGNETOM Sola 1.5T is highlighted by its role as the first 1.5T MRI with Siemens’ BioMatrix tech (reduces setup time, monitors and corrects patient motion, improves head/neck imaging via distortion-mitigation tech), its new Multi-Slice TSE software that reduce MSK scan times by as much as 46%, and its Compressed Sensing for faster cardiac and liver exams.
- A team of Canadian researchers developed a new artificial intelligence algorithm that can spot signs of cognitive decline by studying brain MRIs, genetics, and clinical data, helping doctors predict whether a patient is likely to deteriorate towards Alzheimer’s in the next five years. The algorithm was trained on over 800 people with different neurological status levels (normal to Alzheimer’s) and is currently being refined with new data with a goal of making predictions beyond 5 years.
- Kheiron Medical Technologies announced the European CE clearance of its new AI-based breast screening software (its FDA clearance process is underway). The software serves as a second mammogram reader, feeding results into radiologists’ existing workflows “within seconds” to support triage and prioritization, and streamlining overall mammogram screening workflow.
- Canon Medical Systems announced its Healthy Sonographer Program, focused on (you guessed it) preventing sonographer injuries through education on positioning best practices – including training events with CME credits, and its new HealthySonographer.com resource site. Canon has good reason to launch this program, given that 90% of clinical sonographers have experienced work-related musculoskeletal disorders at the expense of their comfort and providers’ staffing costs.
- CureMetrix agreed to provide its cmAssist AI-based computer aided detection (AI-CAD) solution to the 300-radiologist teleradiology company, DocPanel, for use with second opinion mammograms.
- Oregon’s Asante Ashland Community Hospital purchased five Carestream imaging systems, including the DRX-Evolution, DRX-Excel Plus (fluoroscopy + general table), DRX-Ascend System, and DRX-Revolution Mobile X-Ray System, while upgrading an existing portable CR system to DR with a CARESTREAM DRX-1. Asante Ashland was already a Carestream client, with a number of its X-ray systems in use.
- Research from Taiwan found that synthetic mammogram combined with digital breast tomosynthesis (DBT) produce similar sensitivity and specificity as full-field digital mammography (FFDM) in identifying microcalcifications. The retrospective study performed FFDM and DBT on 72 mammograms that were recalled for microcalcifications (54 benign scans, 18 with malignant microcalcifications, and 20 controls), and as noted above, DBT combined with synthetic mammography had similar sensitivity and specificity as FFDM.
- Glassbeam announced the integration of Monnit’s external environmental sensors (measures: temp, humidity, equipment water flow, compressor power) into its Clinical Engineering Analytics (CLEAN) Blueprint software, which is used to remotely monitor medical imaging systems. Before now, CLEAN exclusively monitored machine logs.
- Siemens Healthineers rolled out a new version of its free online service portal, LifeNet, which providers use to manage the productivity of their Siemens equipment and service contracts (planning maintenance, monitoring equipment status, accessing service docs, etc.¬). The improvements offered in the new version are currently unclear.
The Resource Wire
– This is sponsored content.
- A University of Chicago survey found that 35% of patients have been surprised by a medical imaging bill. The Medmo healthcare marketplace eliminates billing surprises, allowing uninsured and underinsured patients to enter the price they can afford and schedule their procedure with an imaging provider willing to accept that same exact price.
- What can Focused Ultrasound treat? Good question. This article from the Focused Ultrasound Foundation details the diseases and conditions the technology is approved to treat and the over 100 treatments currently in development.
- In this Carestream video, an orthopedic surgeon opens up about why he decided to add the OnSight 3D Extremity System and how his practice benefits from the weight bearing CT.
- The POCUS Systems founding team has over 80 years of combined experience in the ultrasound industry.
