AI in medical imaging is a fast-growing area, but one that also poses multiple challenges around applications and regulatory clearance. Artificial Intelligence-based Machine Learning (ML) techniques allow for routine and more advanced workflows, such as detecting abnormalities or segmenting regions of anatomical interest, to be applied to 2D and 3D image data such as from Computed Tomography (CT) and Magnetic Resonance Imaging (MRI). One of the ongoing issues for medical device companies and clinicians is to ensure that new technology is available for healthcare whilst maintaining accuracy and safety for patients, particularly when dealing with algorithms that change or adapt over time.

What are the Benefits of AI in Medical Imaging?

The main benefits of using in AI in medical imaging include the ability to use algorithms to carry out large-scale projects involving multiple patient datasets, therefore reducing planning time and potentially improving clinical decision-making. By using AI, clinicians can obtain rapid insights or information that could not be easily found using traditional imaging review methods, for example in the heart. These inputs can then help with diagnosis, but also in terms of real-time monitoring and recommendations for patient’s health, managing hospital workflows, and helping with medical device design testing and simulation by speeding up common steps needed to produce 3D models from imaging.

How does AI in Medical Imaging Work?

According to the U.S. Food & Drug Administration (FDA), there are several types of AI used in healthcare, including CADe to detect information such as abnormalities within an image, and CADx that may use algorithms that add additional diagnostic information (such as a tumor grade). In addition, there are Non-CADe algorithms that can be used to help segment anatomical structures. More broadly, AI algorithms are trained based on large amounts of data for a specific purpose, for example to identify risk factors, and then developed and validated against new data to show that accurate results can be repeated.

Most AI or ML-based algorithms are ‘locked’, so that they will not adapt or learn beyond the specific training and performance testing, and are generally easier to control and reduce the risk of errors for. By comparison, ‘adaptive’ or ‘continuously learning’ algorithms will update their outputs based on real-world experience or new data; being able to regulate for these types of algorithm is more difficult, as it introduces a significant degree of uncertainty to whether the algorithms are still providing suitable outputs.

A further category is generative AI, for example ChatGPT, which uses very large amounts of data to generate new material, for example text. Adaptive AI is expected to continuously learn and update based on new data, and is more suited to dynamic data-driven situations, compared to the use of generative AI for creating synthetic creative content. Generative AI may have some applications to medical diagnosis, such as for creating realistic and anonymized patient data for medical research and training to test out new clinical inputs.

AI in Medical Imaging and Synopsys

Where and when does AI in Medical Imaging fit in the product portfolio?

AI-powered tools are a key part of Synopsys’s focus on Pervasive Intelligence across product lifecycles. In terms of medical imaging, the Synopsys Simpleware Product Group have developed off-the-shelf locked AI-enabled ML algorithms for segmenting and landmarking regions of interest for different anatomies, for example in orthopedics or cardiology.  For users that need more specialized algorithms focused on their particular workflows, customized algorithms built from the ground-up can also be developed and deployed. These algorithms are intended to reduce the time-intensive segmentation process by automating routine steps for segmenting regions and adding landmark measurements, while still allowing for manual edits and verification from the user. By doing so, these tools free up valuable engineering time for building specific workflows for solving challenges with medical imaging, and help automate multiple parts of a workflow to scale up medical image analysis and device development.


Going beyond AI in Medical Imaging

The use of AI in medical imaging is changing rapidly, most notably in terms of regulatory clearance, and the scaling of algorithms to suit different types of challenges. For example, consultancy projects at Synopsys are available for using AI/ML techniques to fully automate specific workflows. Simpleware Custom Modeler shows how consultancy can be used to develop user-specific and patient-specific processes that can bring in additional image processing steps such as statistics and meshing for export to related fields such as point of care 3D printing.


Putting AI in Medical Imaging into Practice

One example of AI in medical imaging using automated segmentation for a larger workflow involved a recent collaboration between Synopsys and nTopology for patient-specific design. In this case, Simpleware software was used to segment knee CT data, followed by customized design of surgical guide in nTopology, with scripting used to automate and scale up the steps for generating the guides.

Automated tibia segmentation using Simpleware AS Ortho/CMF.

Automated tibia segmentation using Simpleware AS Ortho/CMF.


  1. 50 patient-specific tibia models were imported to Simpleware software, with AI-based segmentation and landmarking used to automate extracting regions of interest and measurements from the image data.
  2. Data was then exported to nTopology to design the interaction between a cutting guide and the patient-specific models.
  3. The workflow was scripted to avoid repetitive manual work, resulting in a set of models suitable for further design work and 3D printing.

Other examples of AI being used in clinical workflows include the development of a patient-specific total hip arthroplasty tool with medical device company Corin: a customized process was built that uses Simpleware automated tools to help plan procedures, and has been used in over 20,000 cases. In addition, the Cardiovascular Surgery Advanced Projects Laboratory (APL) at Nicklaus Children’s Hospital used Simpleware AI-enabled tools to rapidly generate 3D print-ready models from DICOM data of the heart for planning complex surgeries.

Continue Reading