Siddharth Shah

The dust is now beginning to settle on the idea of AI in imaging-an idea that once took the industry by storm. The questions being asked by radiologists have now changed from “will it replace me?” to “how can it help me?”-and rightly so. AI continues to make significant progress in the field of diagnostic imaging, as can be gauged using the recently concluded Radiological Society of North America’s (RSNA) Annual Meeting in Chicago as a barometer.

Last year, there were 49 exhibitors at RSNA tagged as machine learning companies, and 22 of those were first time exhibitors. This year, the same number more than doubled to 104, 25 of which were first-time exhibitors. More importantly, the incumbents of medical imaging equipment have also made notable AI efforts, with each one launching new or enhanced AI capabilities. Clearly, AI was one of the key themes at RSNA.

Based on recent analysis, Frost & Sullivan notes that of the 114 startups active in the AI for medical imaging space, a significant majority target the image analysis aspect of medical imaging. Identifying and analyzing specific features in an image form the crux of a radiologists’ job, and since they base their findings on this analysis, it forms the most important clinical step in the imaging workflow. The startup disruption rampant in overall healthcare has focused on image analysis in the case of medical imaging AI.

While computed tomography and magnetic resonance imaging are the top modalities of focus, these are followed closely by digital x-ray, mammography (including 3D tomosynthesis), fundus imaging (of the eye), ultrasound, and echocardiography. While most imaging equipment vendors have not touched upon this area, Siemens Healthineers debuted the AI-Rad Companion Chest CT at RSNA this year, leveraging datasets that add up to 325 million curated and annotated images available for training AI algorithms. The intelligent software assistant can take vendor-agnostic CT images and highlight various thoracic structures, as well as mark any potential abnormalities and include their measurements in structured reports-targeting the heart, aorta, lungs, and coronary arteries. GE Healthcare, on the other hand, launched the Edison platform that brings together all AI algorithms from its existing partnerships.

Beyond image analysis, there are several other steps in the imaging workflow that can also benefit from AI. Right from the ordering stage, all the way down to reporting, AI solutions either exist or can be developed.

Naturally, image analysis is the most advanced application, followed by assignments-AI applications for triaging, worklist assignment, and workflow orchestration are available from Aidoc, Zebra Medical Vision, vRad, and GE Healthcare, for example.

The next most advanced area is of decision support for helping decide the next steps in the clinical pathway-representative examples include Imaging Related Clinical Context from GE healthcare, IntelliSpace Oncology by Philips, and PowerScribe 360 Clinical Guidance from Nuance.

Three Reasons AI is ‘Wildly Different’ than Other Healthcare Technologies

Without going into the details of all other steps, the two workflow steps that could still benefit from the attention of AI solution developers are the ordering and scheduling steps. Companies providing decision support to physicians are best suited to take advantage of such solutions for the ordering stage, while scheduling solutions would be useful for optimizing utilization via scheduling improvements around cancellations and no-shows, for example. There is ample opportunity for growth in this area, using AI to make the workflow more efficient and support the radiologists in ways not previously possible.

Equipment manufacturers are gearing up to make their imaging machines intelligent by infusing AI in their scanners, or at the edge. Canon, for example, debuted the Advanced Intelligent Clear-IQ Engine (AICE) image reconstruction solution that uses AI to deliver high-quality images from lower radiation-dose, low-quality images. Siemens Healthineers already has the FAST 3D camera, which helps with patient isocentering, eliminating the need for a higher dose, improving image quality, and preventing repeat scans which may not be reimbursed.

Beyond these use cases, there are other areas such as self-awareness that allow equipment to “self-diagnose” health of their components, self-driving capabilities to make suggestions or perform additional steps when warranted based on patient condition and so on. These functionalities could be the future of AI with imaging equipment, making the machines truly intelligent.

While on the technology front we continue to make several strides, the value proposition of AI needs to be studied and well-understood for a solution to be adopted in the market. Different stakeholders in the radiology workflow-technicians, radiologists, radiology managers, hospital managers, and even payers-have different requirements that can be addressed by AI. For example, automation of redundant time-intensive tasks can help with increasing productivity, AI-supported image analysis can assist with improving accuracy, and predictability and personalization of patient care pathways can help improve institutional outcomes, and so on.

The development, validation, and deployment of these solutions on a commercial scale requires an ecosystem approach, and several kinds of partnerships and agreements make this a complex ecosystem. From the development of online marketplaces that provide hospitals access to several vendors’ AI solutions on a single platform, to equipment manufacturers partnering with vendors to integrate the solutions with their own are examples that are closest to the end-users. But there are other developments happening in the background as well-technology giants partnering with vendors for cloud-based approaches, computing hardware manufacturers providing solutions for improved computing prowess to equipment manufacturers, and academic institutions continuing to build new AI algorithms and spinning out startups based on those solutions, for example.

The investment horizon for the technology has been promising-Frost & Sullivan estimates that $3.7 billion has been invested globally in the development of this technology so far (including $1.9 billion in startup funding alone, as of September 2018), and we continue to see more AI startups being funded.

However, the promise of AI in the imaging field will have to deliver savings for the end-users: time savings, resource optimization, accuracy gains, and perception gains (bordering on precision health approaches). While the first two refer to the productivity aspect, the latter two cater to quality. It seems that the initial promised savings will be delivered on the productivity front, whereas the quality savings will probably take a long time to catch up-this is also attributed to the lack of appropriate metrics to measure outcomes and resultant quality savings, which is relatively easier to measure from a productivity standpoint.

AI in imaging is not just here to stay-it’s already helping build new equations in the industry. The next frontier will be improving upon the lives of patients and helping radiologists do this in a more efficient way-from image analysis, workflow applications, and later with intelligent medical imaging machines.

Robin Joffe and Siddharth Shah work in transformational health at Frost & Sullivan

Articles

No longer in its infancy, AI is beginning to make real waves in radiology. Here's how. 

How AI is Evolving in Diagnostic Imaging

Trust-or a lack thereof-has led to the rise of the blockchain wave in healthcare. As the industry increasingly adopts digital technologies and undergoes a digital transformation, issues such as fraud, inefficiencies, and data privacy have cropped up-

 is lost in global healthcare annually due to fraudulent means, discarded and mishandled data/records. With these issues being exposed, and more digital solutions becoming available to solve them, blockchain presents itself as the perfect underlying technology to address the trust deficit in healthcare.

 defines blockchain as “a new data structure that creates trusted, distributed digital ledgers for assets and other data. It is an immutable and time-stamped record of digital events shared peer to peer between different parties. It can only be updated by consensus of a majority of the participants in the system and, once entered, information is very hard to erase (immutable).”

This distributed digital ledger approach is already seeing adoption in hospitals for medical records (Vanderbilt University, Mayo Clinic, and Mount Sinai), for medical services platforms (Myongji Hospital, South Korea) and to improve data security (Columbia University).

Other organizations such as Humana, Optum, and United Health Insurance are piloting the technology for building physician directories and to reduce administrative burdens. Additionally, the Australian federal government is piloting the technology for providing researchers access to health records, while ZhongAn (insurance giant founded by Tencent and Alibaba) is piloting the tech for data collection and processing with 100 hospitals in China.

The market opportunity is expected to expand further, and according to industry research estimates nearly 1/3rd of healthcare organizations are working with blockchain technology. However, the current focus seems to around B2B enterprise use cases such as: health professional credentialing, medical billing management, revenue cycle management, contract adjudication, and track and trace. So what about radiology?

The size of a typical medical imaging scan, on the conservative side, is about 200 MB. As the demand for radiological studies continues to rise globally, the redundant storage requirements mean a significant need for storage space. While this can be addressed by cloud storage applications, it is not just about storage-it is also about accessing those images when necessary for making interpretations in the future as well and in a secure manner. With precision medicine approaches now gaining ground, a holistic view of the patient means also accessing scans from other sources, such as pathology and dermatology, for example.

Such access to medical information is demanded by patients, as well as their doctors from other facilities. It is well understood that repeat imaging scans in different facilities for the same patient, and with cost variance of imaging services, results in inefficient spending of healthcare dollars-not to mention the frustration caused to the patient in an era of healthcare consumerization.

Top 5 Healthcare Technologies of the Future

Moving large image files from one point to another poses some technical challenges, but adding to the complexity are stringent patient privacy laws, which induce bureaucracy in healthcare systems to provide access to data. And as a proverbial cherry on top, healthcare as an industry is a prime target for hackers to gain access to data-making cybersecurity a primary concern for custodians of such records.

Imaging centers that own vast tomes of medical imaging data are unable to efficiently leverage or monetize the data to support the development of artificial intelligence (AI) algorithms because of these challenges. Because of burgeoning cost pressures in a value-based environment, they would benefit immensely from the additional revenue source. In addition, a shortage of radiologists globally (especially after-hours, tier-II, rural areas, and for sub-specialties) means lack of access or longer wait times-resulting in elongated turnaround times even for physicians to make diagnoses. For this, teleradiology acts a buoyant solution, but accessibility towards patient imaging scans continues to incur challenges in this field.

The same challenges also hinder the development of AI algorithms in the medical imaging space, which must be trained on existing imaging scans for them to ‘learn.’ While AI is poised to address some of the challenges afflicting radiology today, deep learning and machine learning systems depend on easier access to data with a wide variety (available through larger numbers of studies) to improve their capabilities. In essence, without easy access, AI development will remain stunted, and issues such as transfer learning challenges will persist.

There are a few companies currently employing blockchain in the field of radiology. Two representative examples include MDW (Medical Diagnostic Web) and DeepRadiology, both of whom deal with AI, but in very different ways.

DeepRadiology is primarily an AI company, with solutions developed for head CT scans, and others under development. However, it plans to integrate blockchain technology for increasing its international presence, to support and incentivize ecosystem members who are part of developing and using DeepRadiology solutions, and to reduce healthcare costs. These second and third aims are also shared by the other player, MDW.

MDW is a blockchain-based decentralized platform that connects imaging facilities, radiologists, and AI developers. At the 2018 RSNA Annual Meeting, they unveiled their platform and also have existing clients on their system-individual radiologists, smaller imaging facilities and teleradiology groups. The goal with the platform is to provide crowd reading, second opinion, peer review, or quality control, in an open and transparent environment, as promised by the underlying technology itself.

MDW also specializes in sourcing medical imaging datasets for AI developers to access, allowing monetization of the anonymized, annotated data for imaging centers, while providing access to crucial training data for the developers. It has also launched a marketplace for AI algorithms to facilitate imaging facilities to make use of existing AI algorithms-this is expected to be similar to the cloud-based, on-demand platforms available from EnvoyAI, Nuance, and Blackford Analysis, but based on a blockchain-based payment platform. Next year, MDW expects to bring more functionality to the platform, while scaling up and acquiring more customers.

Other examples include MedNetwork and MediBloc, which both enable storing and sharing for medical information via blockchain networks. The former is focused solely on medical imaging data, while the latter store’s health records (including diagnostic results), providing access to providers when visited by the patient. MedNetwork also has a telemedicine platform providing AI diagnostics and second opinions, somewhat similar to the approaches above.

The benefits of the technology for smaller players, and for teleradiologists are clear. And yet, these two platforms are still addressing only parts of the radiology paradigm for blockchain. Another use case for the technology for providers involves equipment lifecycle management. A solution provided by Spiritus Partners addresses critical industry challenges around device safety, quality and lifecycle management as they apply to all medical devices, including imaging equipment. As is evident, use cases for the technology are varied and, as yet, emerging.

Blockchain will not disrupt existing workflows, nor will it replace existing health IT systems such as DICOM or PACS, nor will it affect the interoperability standards such as HL7 or FHIR. However, it will act as an additional layer of trust and security for seamless interactions by automating the traditional consensus mechanism which is mutually agreed upon by all participants. Moreover, the convergence of the technology with other emerging technologies such as the Internet of Things (IoT) and AI will boost trust around them, promoting new care delivery models and monetization options.

Frost & Sullivan predicts that blockchain in healthcare will move beyond its hype next year. By the end of 2019, an estimated 5-10% of healthcare-focused enterprise blockchain applications will move from pilot stage to partial or limited commercial availability.

Blockchain for healthcare is still relatively immature, but select B2B commercial implementations examples include those by Change Healthcare and Hashed Health. This year, Change Healthcare announced limited availability of their Intelligent Healthcare Network for the claim adjudication use case. It also collaborated with TIBCO to develop blockchain-powered Smart Contracts for automating healthcare transaction processes. Hashed Health recently announced the initial corporate partners for their professional credentialing application called Procredex, including WellCare Health Plans, Spectrum Health, National Government Services, HealthLink Dimensions, LLC, and Accenture, among others. It is also planning to demo its performance-based contract adjudication and management platform called Signal Stream soon.

For radiology we expect some more organizations to get on board blockchain-based platforms such as MDW.

However, it is important to realize some of the limitations of this technology-blockchain is just one technology falling under the larger distributed-ledger technology umbrella, and it may not necessarily be the best. Over time, another technology may overcome blockchain as the better choice. Secondly, being a network play, the governance mechanisms need to be agreed upon first, before deployment of the infrastructure can begin, which is also why enterprise blockchain is a better near-time opportunity over a public network such as those being considered for national health records.

Radiology-and healthcare as a whole-has a trust problem, but blockchain could be the answer. 

Siddharth Shah

Articles

How AI is Evolving in Diagnostic Imaging

How Blockchain Could Solve Radiology's Problems