Artificial intelligence (AI)-based solutions can easily be found in many different aspects of healthcare, from administrative support to innovative healthcare technology and clinical applications. In fact, industry experts anticipate even more growth in the AI healthcare industry in the coming years.[1]
Because of radiation oncology’s heavy reliance on digital data processing for accurate radiation therapy planning, AI is a natural fit. AI applications have the potential to provide a quantitative assessment of clinical conditions and improve the accuracy, precision, efficiency, and overall quality of radiation therapy for patients with cancer.[2]
Realizing MRI-only radiation oncology treatment planning
Approximately 60 percent of cancer patients need radiation therapy—a process that requires high-quality medical imaging. Using medical imaging to guide radiation therapy improves the efficiency of treatments with more accurate tumor response tracking so treatments may be adjusted if needed.
Today, clinicians rely on computed tomography (CT) as the standard imaging modality for radiotherapy planning; CT provides a reliable three-dimensional view of the tumor as well as data regarding tissue density, which is used for dose calculations in radiotherapy planning.[3] However, the development of AI-based tools makes effective radiation treatment planning and monitoring possible without the use of CT, but of other imaging modalities, such as magnetic resonance imaging (MRI). Industry leaders, such as GE Healthcare are offering new AI technology to enable MRI-only radiation therapy planning for clinicians to utilize in their daily clinical practices.
Additionally, leveraging partnerships, such as the use of Spectronic Medical’s MRI Planner, this AI-based software uses a single MR imaging sequence to generate high-quality synthetic CT images for adults, in the area of the brain, head-neck, and general pelvis. When combined with GE Healthcare’s MRI imaging technology, Spectronic Medical’s MRI Planner enables MRI-only based radiotherapy planning for better soft tissue differentiation than traditional CT. This is especially important for the most common types of radiation oncology therapies in the brain, head-neck, and pelvis.
Figure 1 Synthetic CT image for radiotherapy treatment planning, using MR and the MRI Planner from Spectronic Medical
Benefits of MRI-only radiation therapy planning
AI tools and deep learning applications are helping radiation oncologists integrate and co-register patient data, as well as automate many therapy-planning tasks, such as in prostate cancer. Using these tools in MRI, radiation oncologists can generate synthetic CT images from MRI data, supporting the emerging trend of MRI-only radiation therapy. When high-quality MR images are converted to synthetic CT images, clinicians can utilize the details in the MRI for tumor volume and organ-at-risk delineation, thereby eliminating the need for a conventional CT examination as well as the work and uncertainties involved in multimodal image registration. This innovation has the potential to streamline the clinical workflow and supports precise treatment delivery.
“In an MRI-only workflow,” explained Carl Siversson, CEO and founder of Spectronic Medical in Helsingborg, Sweden, “we eliminate the need for the CT and only rely on MRI throughout the treatment-planning session. From the MRI, we generate both the synthetic CT image, which is matched to the MRI, and for prostate patients we also generate a set of automated organs at risk contours.”
Additionally, with this type of technology, an organization may convert from a dual-modality radiation planning workflow to MRI-only due to reducing imaging registration uncertainties, as well as saving time by eliminating the patients’ CT exam. The patient also has the added benefit of reduced ionizing radiation as the CT is eliminated, explained Lars E. Olsson, Professor of Medical Radiation at Lund University, Skåne University Hospital.
Examining MRI-only radiation therapy planning in prostate cancer
In radiation therapy planning for prostate cancer, fiducial marker visualization is critical.
“The gold fiducial markers [used in prostate cancer treatment] are easily seen on conventional CT images,” Olsson said. “You see the strike artifacts. If you have calcification, you can use this artifact to distinguish between the markers and the calcification. In MRI images, however, you can use multi-echo images to perform a similar distinguishment.”
The MRI Planner software is designed to help visualize such fiducial markers on synthetic CT.
“The algorithm that we have developed,” added Siversson, “has quite a few advantages…particularly, it’s able to generate a very high level of detail in the synthetic CT.”
Because of the visualization capabilities, MRI-only radiation therapy planning is seeking to help healthcare professionals improve prostate cancer treatment.
“To date,” explained Sacha af Wetterstedt, Medical Physics MSc with several years of experience in the field of external radiotherapy at Skåne University Hospital, “around 200 patients have been planned and started treatment within the MRI-only regime over the past year at our radiotherapy department. We use it for prostate cancer patient treatments where we use the gold fiducial markers.”
Continuing to innovate in MRI-only radiation therapy planning
Leveraging industry partnerships and expertise, such as combining the technologies of Spectronic Medical’s MRI planner and GE Healthcare’s powerful MRI imaging technology, can provide clinicians with the CT images they have come to expect, along with the MR soft tissue details that are needed to accurately target lesions and help improve patients’ radiation therapy regimens.
More applications of MRI-only therapy planning are available in the brain, head-neck, and pelvic radiotherapy planning,[4] enabling high-precision treatment planning with the potential to improve clinical outcomes for countless radiation therapy patients.
RELATED CONTENT:
- Read more on Spectronic Medical and GE Healthcare’s radiation therapy planning solution: Enabling Radiation Oncology Treatment Planning with Advanced Imaging Techniques and AI-based Tools
- Article: Global Trends in Radiation Oncology
- Article: Top 5 Challenges in Radiation Oncology
- Discover our Radiation Oncology Solutions here
DISCLAIMER
Not all products or features are available in all geographies. Check with your local GE Healthcare representative for availability in your country.
REFERENCES
[1] https://www.prnewswire.com/news-releases/global-ai-in-healthcare-market-2021-to-2027---by-sections-diagnosis-end-user-and-geography-301465668.html
[2] Huynh, E., Hosny, A., Guthier, C. et al. Artificial intelligence in radiation oncology. Nat Rev Clin Oncol 17, 771–781 (2020). https://doi.org/10.1038/s41571-020-0417-8.
[3] Bhide, S. A., Newbold, K. L., Harrington, K. J. & Nutting, C. M. Clinical evaluation of
intensity-modulated radiotherapy for head and neck cancers. Br. J. Radiol. 85,
487–494 (2012).
[4] Spectronic Medical software is CE marked and FDA 510(k) cleared for use in adult patients in brain, head-neck, and pelvic radiotherapy planning in both Europe and the US.