Market and Technology Trends
Artificial Intelligence for Medical Imaging 2020
By Yole Intelligence —
With the emergence of AI in imaging, the medical industry and the radiology profession have begun to dramatically change.
Table of contents 3
Report objectives 5
Report scope 6
Methodologies and definitions 8
About the authors 9
Companies cited in this report 10
Who should be interested by this report 11
Yole Group related reports 12
Executive summary 16
- AI overview
- Medical applications
- Data in healthcare
- What about future?
Market forecasts 76
- Assumptions on penetration rates and ASP
- Total AI revenue
- Evolution by application
- Evolution by modality
- Evolution of AI revenues per segment
Market trends 92
- AI for medical imaging: a software business
- Insurance benefits
- The regulations and constraints
- What’s next
- Market shares
- Mergers and acquisitions
- Recent product launch
- Interesting start-ups
Technology trends 212
- Why using AI
- Modality specifics
- Software development
- IT infrastructure
Annex – Company profiles 244
Annex – Algorithms review 263
Yole Group of Companies 279
ARTIFICIAL INTELLIGENCE ALGORITHMS: THREATS AND OPPORTUNITIES
Why is artificial intelligence (AI) a term we find regularly in the medical industry? In particular, why is this technology at the center of debates in the medical imaging market? Because this technology has the potential to change all of our diagnostics and treatment procedures to enable more personalized and effective medicine.
Artificial intelligence is based on the training of algorithms. Deep learning is a type of AI technology based on artificial neural networks which can detect more precise details in the data. This technology has initially been implemented for recognition models and is specialized for the study of images.
Radiology is mutating with the adoption of deep learning models for the recognition of lesions in the body, to prioritize cases for the direct treatment of patients at risk, to predict the evolution of pathologies. Furthermore, AI affects all the imaging modalities in particular Magnetic Resonance Imaging (MRI), Computed Tomography (CT) scanning, X-rays and ultrasound imaging. These are the ones at the center of our study.
Not every type of modality requires the same types of algorithms. In fact, modalities can be organized into two types of procedures: quality procedures, which include MRI and CT scans, and fast imaging procedures, which include ultrasound and X-rays.
The professionals’ needs depend highly on the imaging modality used.
On the one hand, MRI and CT scans are intensive procedures able to acquire high quality images. With the addition of annotations on the images, the model can reach very high accuracy to classify pathologies or to segment objects. Furthermore, the execution speed of the model does not need to be very fast, as the imaging procedure is usually long. On the other hand, models trained on ultrasound images are in need of very fast execution to be able to process real-time images. Those models are then used to detect abnormalities faster and prioritize cases, implying an important productivity gain.
The application of the models empowered by AI can be classified as:
- Screening models: in charge of the detection of abnormalities
- Diagnostic models: in charge of the evaluation of the disease
- Treatment planning models: able to predict the most pertinent treatment according to the pathology and the physical condition of the patient.
The value generated by the use of such models in hospitals depends on their applications.
The Artificial Intelligence for Medical Imaging 2020 report evaluates the impact of deep learning on the medical imaging field.
AI FOR MEDICAL IMAGING: A FAST GROWING MARKET. WORTH $2.9B IN 2025, ITS VALUE WILL MULTIPLY BY 15-FOLD IN FIVE YEARS
The medical imaging field is very restrictive for the development and deployment of AI algorithms. The constraints are either related to medical data or to regulations.
Medical data and more specifically medical imaging depend on the parametrization of the imaging machine and on its vendor. The development of the interoperability of the models is highly challenging for software companies and hospitals.
To predict the evolution of a pathology, researchers must do longitudinal studies, requiring access to medical images of a patient over time, years in general. Inconsistencies in most of the medical data make the development of accurate and complex studies challenging.
Furthermore, regulations regarding data privacy make data access more complex for every player in the ecosystem. Annotated images are the fuel for AI, and access to them is often restricted, as well as other interesting data concerning the patients’ age, gender, size and so on.
Although the medical field is highly supervised, the AI market in this field will grow rapidly, first with the introduction of the AI products from original equipment manufacturer (OEMs), in particular Siemens and Philips, and secondly with the introduction of diagnostic algorithms into the market from 2021 onwards. The large productivity gain linked to the use of such models is valuable, enabling hospitals to more effectively diagnose and treat patients.
Yole Développement’s analysts estimate that the total market in 2025 for software generated revenues through the sale of AI tools will reach $2.9B with a Compound Annual Growth Rate for 2019 to 2025 (CAGR2019-2025) of 36%, shared between the main applications: improved image capture, noise reduction, image reconstruction, screening, diagnostic and treatment planning.
AN EVER-MOVING ECOSYSTEM WITH DIVERSE MARKET POSITIONS AND STRUCTURES
The artificial intelligence for medical imaging market is made up of a myriad of start-ups, many of which being spin-offs from universities. It highlights the fact that this market is driven by medical research, as well as the value it can generate for hospitals with regards to all the hardware, and cloud companies as well as OEMs that revolve around it.
Each software company is highly specific, providing AI algorithms for dedicated applications and modalities. They are able to sell their products directly to hospitals or through two different types of players: marketplaces such as TeraRecon, which are integrating those algorithms into the radiologist’s workflow, or through OEMs such as Siemens, which are optimizing those models on their medical imaging machines.
In fact, OEM machine manufacturers and marketplaces, which are software companies, are offering the approximately the same product. Which type of player will be in the stronger position in the coming years?
Yole Développement’s report provides some answers to these key questions as well as a detailed analysis of the different strategies of these players.
4quant, 16bit, Advantis, ai analysis inc, Aidence, Aidoc, Amazon, Aquila Medical Innovation, Arterys, Ascension, Avalon AI, Azmed, Balzano, Behold.ai, Blackford Analysis, Brainminer, BrainScan, Butterfly Network, Canon, Caption Health, Carestream, Cercare Medical, Circle Cardiovascular Imaging, Contextflow, Corindus Vascular Robotics, Curacloud, Curemetrix, Deepcare, DeepMind, Deepnoid, Deepradiology, Deepwise, Densitas, Deski, Dia Imaging Analysis, Dr CADx, eko.ai, Enlitic, Fujifilm, GE Healthcare, Gleamer, Google, Healthmyne, Hearthflow, IBM, Icometrix, Idx, Image Biopsy Lab, Imbio, Incepto, Infervision, Infinitt Healthcare, Innovationdx, Intel, Johnson & Johnson, Kheiron Medical Technologies, Koios, LPixel, Lunit, maxQ, Mazor Robotics, MD.ai, Medtronic, Mellanox Technologies, Methinks, Microsoft, Neuropsycad, Nuance, NVidia, Optellum, Oracle, Oxipit, Perceiv.ai, Peredoc, Perspectum, Philips, Pixyl, Predible, Qmenta, Quantib, Quantitative Insights, Quibim, Qure.ai, Samsung, Screenpoint, Siemens Healthineers, Terarecon, Therapanacea, Therapixel, Ultromics, Verb Surgical, Vida, Viz.ai, Volpara Solutions, Voxelcloud, Vuno, Yitu, Zebra Medical Vision and more.
Key features of the report
- Artificial intelligence (AI) technologies used in medical imaging applications
- AI software companies’ strategies
- Evolution of business models and positioning
- Cloud and on-premises computing for AI
- Regulations and constraints inherent to the medical imaging field
- Ecosystems, market forecasts and major trends
Objectives of the report
- Provide an overview of the market of AI in the field of medical imaging in terms of number of algorithms deployed and the value they generate for every player involved, at medical device level, AI platform level, and algorithm development level along with the understanding of the ecosystem, technologies used, strategic positioning, and how these will evolve in the coming years
- Present the current market data and forecasts depending on the modality studied (MRI, CT scans, X-rays and Ultrasound) and its application in the patient diagnostic process (noise reduction, screening or diagnostics), in dollar value and volume of images analyzed
- Identify where the opportunities lie for each type of player along with a detailed description of the regulations and constraints of this field. In addition to the current overview, an introduction of the evolution of those regulations in the coming years is presented