Big Three’s Big Plans | AI for Consciousness | Brazil Scandal’s Next Phase

“Imagine that in the future, we have a patient with all their organ functions, all their cellular functions, and we are able to simulate this complexity.”

Heidelberg University Hospital cardiologist, Benjamin Meder, on Siemens Healthineers’ Digital Heart twin simulation software and AI’s potential to support holistic, personalized diagnosis.

 

 


The Imaging Wire

Expects Big Things from Imaging’s Big Three
A recent article from Evaluate’s Vantage publication explored the theory that an increased medical focus from Philips Healthcare, Siemens Healthineers, and GE Healthcare will bring an influx of M&A activity from the three industry heavyweights and make them more reactive to industry changes. The article used Philips as evidence of this theory, as the one-time conglomerate was the first of the three to narrow its focus on health and wellness, and has since been arguably the industry’s most active acquirer. The article pointed out that Siemens Healthineers is expected to put its IPO-generated cash to use soon enough, although Siemens has been patient so far, focusing recent activity more on partnerships than acquisitions. Vantage was understandably the least specific about GE Healthcare’s short-term plans, suggesting that GE Healthcare will look to partnerships to grow its capabilities, with a focus on AI. Although some of these ideas have been well documented by the press/analysts and from the manufacturers themselves (Philips, Siemens, GE), the article adds weight to a growing perception that medical imaging’s big three players will make even more disruptive healthcare moves in the coming years.

 

Chinese AI Algorithm Predicts if Patients will Regain Consciousness
Chinese researchers developed an AI model that uses resting state functional MRI (fMRI) to help determine whether patients with severe brain damage might regain consciousness. The researchers trained the AI model on tens of thousands of brain images from 63 patients in a “disorder of consciousness” (vegetative) state at least one month after their brain injury. The algorithm was able to diagnose which patients would and would not recover consciousness with 88% accuracy. Seems niche, but anyone who has ever faced the decisions associated with a loved one (or patient) in a vegetative state would find this solution valuable.

 

Brazilian Corruption Scandal, Phase Two
The Brazilian prosecutors launched the second phase of their investigation into the corruption scandal that brought down GE Latin America CEO, Daurio Speranzini Jr. The new phase is highlighted by the re-arrest of Rio’s former Secretary of Health, Sergio Cortes, after being released in April, as well as the arrest and/or re-arrest of twenty other accused participants. Prosecutors also revealed new details on how public and private sector conspirators worked together and revealed the first official estimate of damages to the state (74m reais / $17m – we thought it would be higher). With a Governor, a CEO, and a Secretary of Health all arrested, this scandal is certainly bringing down a star-studded cast during its first two phases. It will be interesting to see who else falls by the time its complete.

 

 


The Wire

  • Check-Cap revealed interim results for Version 3 of its C-Scan capsule-based X-ray colorectal cancer screening system, revealing promising polyp detection results. A 31-patient study resulted in 76% sensitivity and 80% specificity for polyp detection, attributed to the C-Scan V3’s better scan imaging density, 2D/3D imaging, and a new motility analysis algorithm. Through the end of 2018, C-Scan plans to establish a manufacturing line with GE Healthcare and begin its US regulatory path including a pilot trial in the fourth quarter.
  • Researchers in Iran found that by combining lung nodule features with patient-specific characteristics they were able to improve their cancer screening model’s predictive ability, potentially improving physicians’ ability to identify high-risk nodules and reduce false-positive screening results. Using multilevel logistic regression, the researchers studied nodules nested within rounds and rounds nested within individuals, using nodule-level features to achieve a 7.94% predictive value (twice a National Lung Screening Trial’s predictive value).
  • Body Vision Medical announced that it received $8.5 million in funding that will be used to accelerate the commercialization of its LungVision system in the US and extend its product line. The FDA-cleared LungVision system is a real-time imaging bronchoscopy tool that specializes in diagnosing small peripheral lesions and integrates with standard imaging equipment including CT, ultrasound, and fluoroscopy systems.
  • A study in Peru found that a new rapid cardiac magnetic resonance (CMR) method using a contrast agent could allow faster, lower-cost, and easier-to-deliver performance, making it ideal for developing countries. The new protocol, which utilizes contrast dye to measure cardiac structure, function, and scarring can be performed in 18 minutes for $150, compared to 45 minutes and 3-5x higher costs for advanced cardiac imaging in developed countries.
  • Research from Thomas Jefferson University revealed that radiologists’ use of coronary CT angiography (CCTA) has hit an all-time high since Medicare began covering the test in 2006, recently surpassing cardiologists’ CCTA usage by 20% (largely in hospital outpatient departments). Overall CCTA utilization peaked in 2007, when cardiologists performed nearly 2-times more CCTA procedures than radiologists, however by 2016 cardiologist usage of CCTA fell by 40% and radiologist utilization increased by 13%.
  • Here is one more reason for cardiologists to use CCTA… Danish researchers found that fractional flow reserve computed tomography (FFRCT) could be sufficient for evaluating coronary artery disease, sparing patients from invasive FFR procedures. The study applied FFRCT to patients with moderate stenosis to assess ischaemia by calculating FRR from standard coronary CTA scans and an algorithm for simulating blood flow (possibly Heartflow), demonstrating the ability to measure stenosis and determine treatment.
  • Sectra released its earnings for its fiscal Q1, highlighted by a 15.5% increase in net sales to SEK 306.1 million ($337m), a 25.9% jump in order bookings to SEK 286.5 million ($31.6m), and 4.5% higher operating profit at SEK 46.3 million ($5.1m). However, without positive currency fluctuations, operating profit would have decreased by 1.6%. The company highlighted an increased investment in the US healthcare market, driving higher revenue and costs.
  • Researchers from the University Health Network in Toronto developed a new technique that uses real-time imaging of 3D CT models to guide chest wall tumor resection, reporting that it is more accurate and easier than conventional thoracoscopy. The technique combines conebeam CT and optical tracking with traditional video-based thoracoscopy to perform augmented real-time image guidance, significantly reducing tumor localization errors in the chest’s medial and superior regions and improving surgical margin accuracy, compared with standard thoracoscopy.
  • Wayne State University scientists are working to develop a rigorous computing system intended to make it easier to understand and analyze the internal workings of the human body. The project will try to make it possible to represent and process 4D Riemannian volumetric shapes captured by various imaging devices, helping doctors visualize, track, and diagnose patient conditions.

 

 


The Resource Wire

– Future home of sponsored content.  Maybe yours.

 

 

Leave a Reply

Basic HTML is allowed. Your email address will not be published.

Subscribe to this comment feed via RSS

Get every issue of The Imaging Wire

Delivered right to your inbox. We will never sell or share your information.