14 March 2017
Key Observations from ECR 2017:
- Is Radiology Losing Grip on Imaging IT Decision-Making?
- Applying Deep Learning to Ultrasound – Is the Technology Ready to be Embedded?
- Canon + Toshiba + Vital + Olea = Serious Competitor, but do they have a missing link?
Is Radiology Losing Grip on Imaging IT Decision-Making?
Written by Steve Holloway
In stark contrast to the recent RSNA show in North America, imaging IT and vendor neutral archives (VNA) were far less evident at ECR. Of course, the traditional radiology PACS vendors were there alongside the well-known names in advanced visualisation, but you had to hunt hard for any independent VNA vendors. Even on the major vendors’ booths, imaging IT was far from prominent, hidden away behind the latest modality hardware systems.
Is this a result of limited space at the smaller exhibition, or a reflection of the state of imaging IT maturity in Europe?
For much of Europe today, radiology IT still means PACS and RIS, be it at the radiology department-level, or increasingly “super PACS” at the hospital-network or regional level (such as Spain, Ireland and Scotland). In some cases, VNA has been used to bring disparate PACS systems together between hospital clusters, but for the most part it has remained heavily driven by DICOM radiology and cardiology image management and archiving. If we consider Europe’s largest five markets (Germany, France, UK, Italy, Spain) only the UK and Spain are starting to show any real development towards integrating non-DICOM content into VNA. For the remainder, there are very few examples of collaboration outside of the core DICOM applications, with most limited to academic or university hospitals.
Perhaps more telling was the lack of attendance from two of the largest VNA vendors globally: IBM Merge and Lexmark Healthcare (recently acquired by Kofax). Both are predominantly active in the North American market, but also have customers in Europe. Their lack of exhibition attendance might suggest they don’t yet see enough enterprise VNA opportunity in Europe.
Alternatively, there could be another factor – that enterprise IT adoption (including multi-application VNA) will be decided not by imaging specialists (such as radiologists and cardiologists) but by Chief Information Officers (CIOs), as we’ve seen more recently in North America. While this could have a positive impact in driving enterprise IT strategy and connecting disparate parts of health organisations together, it could also have negative connotations for radiologists; less choice of radiology software, overarching clinical IT decision-making and Electronic Health Record (EHR) vendors with greater customer influence.
Today, radiology IT decision-making remains very much in the hands of radiology departments for most of Europe, but it might not stay there for too much longer.
Applying Deep Learning to Ultrasound – Is the Technology Ready to be Embedded?
Written by Simon Harris
Much like at last year’s RSNA conference, deep learning was one of the key themes at ECR 2017. Several speakers at the scientific sessions presented promising research results for the application of deep learning in specific use-cases. In one of the professional challenges sessions, Dr. Angel Alberich-Bayarri from QUIBIM suggested that convolutional neural networks (CNNs) may already be old news, with generative adversarial nets (GANs), a new architecture for unsupervised neural networks, showing promise for medical imaging applications. GANs may be a solution to one of the major challenges with developing deep learning algorithms – the need for large training data sets.
On the exhibition floor, there were fewer companies showing machine learning solutions than at RSNA (there were at least 20 at RSNA but less than 10 at ECR) and in our conversations with vendors it was evident that expectations were more measured, with less marketing hype. Several of the better known deep learning start-ups were notable by their absence, including Enlitic and Zebra Medical, as was IBM Watson Health.
Samsung chose ECR to make a big push for its S-Detect™ deep learning feature, which is currently available as an option on its RS80A premium ultrasound system. S-Detect™ for Breast makes recommendations about whether a breast abnormality is benign or cancerous. It is commercially available in parts of Europe, the Middle East and Korea and is pending FDA approval in the US. S-Detect™ for Thyroid uses deep learning algorithms to detect and classify suspicious thyroid lesions semi-automatically based on Thyroid Image Reporting and Data System (TI-RADS) scores. With both applications, S-Detect™ produces a report to show the characteristics of the lesion, including composition, echogenecity, orientation, shape, etc., along with the risk of malignancy, e.g. “high suspicion”.
ContextVision, the leading independent vendor of ultrasound image enhancement software, showcased its latest research in artificial intelligence at ECR. Its prototype VEPiO (Virtual Expert Personal Image Optimizer), which is built on the company’s Virtual Expert artificial intelligence platform, can automatically optimize ultrasound images for individual patients. VEPiO aims to improve diagnostic accuracy and reduce scan times, particularly for more challenging patients, by making automated setting adjustments to obtain the optimal image quality. The company is also exploring the use of deep learning to optimise image quality, for organ-specific segmentation and for decision-support functionalities.
Ultrasound OEMs must decide whether deep learning technology is ready to be embedded into their systems, or to take a “wait and see” approach. Although many research papers have found that deep learning can produce good results in specific medical imaging applications, often at or near the performance of experienced radiologists, these are usually based on relatively small datasets and/or small reader studies. It remains to be seen if deep learning will perform as expected in routine clinical use. Although Samsung has taken an early lead and is the first of the major ultrasound vendors to embed deep learning, it carefully positions S-Detect™ for Breast as a decision support tool for “the beginner or non-breast radiologist”.
OEMs must also decide whether to establish an in-house deep learning capability or to partner with a specialist. Deep learning engineers are a scarce and expensive resource and most mid-tier ultrasound OEMs will struggle to attract and retain talent. Instead we expect they will partner with independent software vendors, such as ContextVision. For the major OEMs, we expect to see a combination of build, buy and partner strategies. Most of the major modality OEMs have, to varying extents, established in-house R&D efforts for machine learning and with over 50 start-ups developing artificial intelligence solutions for medical imaging, there’s certainly no shortage of options for acquisitions and partnerships.
Another limiting factor is the additional processing power, typically GPUs, required for embedded deep learning algorithms. Ultrasound is a fiercely contended and price sensitive market and OEMs will be reluctant to add additional hardware cost. Initially we expect deep learning to be an optional feature on premium systems only, such as with the Samsung example, but as is often the case in ultrasound, features that start out on premium systems typically cascade to less expensive high-end and mid-range systems over time.
With deep learning technology progressing at a rapid pace, and ultrasound OEMs constantly on the look-out for the next “big thing” to differentiate their products, it seems inevitable that deep learning will increasingly be embedded in ultrasound systems, both as workflow tools to help with productivity and decision support tools to improve clinical outcomes. It’s no longer a question of will it happen, but when will it happen, and the OEMs that wait too long will get left behind in the AI race.
Canon + Toshiba + Vital + Olea = Serious Competitor, but do they have a missing link?
Written by Steve Holloway
Following the announcement of the completed acquisition of Toshiba Medical Systems by Canon, co-branding for the new firm was proudly displayed at the exhibition. For other exhibitors at the show, it was an ominous sign. Here’s a few reasons why:
Canon DR fills a hole in the Toshiba X-ray portfolio: Toshiba Medical Europe has a solid presence in the European X-ray market, but only in the interventional and fluoroscopy X-ray segments, two saturated and mature markets. Most growth in the European X-ray market in the last five years has come from Flat Panel Detector (FPD) digital radiography for both fixed and mobile systems. This is a market where Canon has a strong reputation for FPD panel technology and smaller equipment sales through their acquisition of Delft DI. Combining the two offerings with Toshiba’s strong CT, MRI and ultrasound offerings will allow the combined entity to target large imaging equipment bundle deals with a full complement of systems.
Strong focus on R&D and innovation: Both Canon and Toshiba Medical are well-known and respected for technically strong products, especially in their core application sectors. While it will take some time for the two firms to integrate R&D and manufacturing capability, the combined brand will no doubt continue to be viewed as a leading vendor for technical capability and image quality, putting them in a good position to challenge the “big three” (Philips Healthcare, Siemens Healthineers and GE Healthcare) for top spot in European imaging hardware.
Back on the acquisition trail: Perhaps the biggest challenge for the combined entity will not be imaging hardware-related, but software-related. While the Vital and Olea Medical products are highly-regarded for advanced imaging and visualisation, the combined offering will be missing a central software platform for managing imaging content and workflow.
While not yet essential in Europe, the importance of clinical content management and enterprise imaging is increasing. What’s more, all major competitors have established imaging IT platforms (Philips Healthcare Intellispace platform, Siemens syngo and Digital Ecosystem, GE Healthcare Centricity platform and Healthcloud). Even mid-size vendors such as AGFA Healthcare (Orbis) and Carestream (Vue) have a significant installed base in Europe.
Vital Images (a subsidiary of Toshiba Medical Systems) more recently started to expand its capability to include workflow tools and VNA in their Vitrea product line, but do not yet have the scale of installed base or feature-set to match other major competitors. Without such a platform, the new Canon-Toshiba venture may still find it hard to compete in large hospital networks and regional tenders requiring both hardware and software capability.
So, while the new vendor will increasingly be able to compete in Europe, it will need to make more acquisitions to boost its clinical software offerings to challenge for top-spot in Europe in the long-term.
New Service from Signify Research: Clinical Content Management IT – 2017
This and other issues will be explored in full in Signify Research’s upcoming intelligence service ‘Clinical Content Management IT – World’, with first delieverable due in February 2017. For further details please click here or contact firstname.lastname@example.org