When the Artificial Intelligence (AI) advisory group was formed by the Society of Radiographers (SoR) in 2020, many in the health sector were uncertain about the impact of AI. At that time, much of AI was still in the research phase, and there was widespread concern that AI might eventually take over jobs.
However, this hasn't been the case. As technology has advanced, many ultrasound machines now come equipped with built-in AI and machine learning capabilities, significantly improving the accuracy and precision of diagnoses. Some providers have integrated AI into their ultrasound machines to offer automated measurements. Initially, many sonographers were apprehensive about AI entering the department, fearing job displacement. However, as speakers like Christina Malamateniou, Reader and Associate Professor in Radiography from City St George’s University have highlighted in various forums, AI is not here to replace jobs but to enhance our work. It is essential for us to learn how to use AI effectively to thrive in the evolving workforce.
As technology and research has improved intelligent ultrasound imaging, interdisciplinary communication between AI developers and sonographers has also strengthened. Many sonographers such as Jackie Matthews and Emily Skelton have led the way in assisting with research in obstetric ultrasound AI (Matthew et al, 2021).
The SoR AI advisory Group have created an AI guidance which is available on the SoR website.
Six key points
Below are some of the benefits of AI in obstetric ultrasound
- Reduce variability between different users
- Improve reproducibility and accuracy
- Reducing scan time and improving interpretation
- Improving automating imaging quality control: Technology has been developed that can provide an audit of image quality based on standard sections Recent ultrasound machines can realise intelligent measurements based on the standard sections (He et al, 2021) such as those required by the national fetal screening programmes
- Improving workflow and efficiency: For example, AI can be used to identify fetal anatomy and add annotations and measurements by taking images in real-time and using international guidance e.g. ISUOG practice guidelines, to recognise anatomical structures during the scan, capture images and record them as being complete. This can reduce the scan time for an examination allowing sonographers more time to communicate with expectant parents
- Research into the use of ultrasound for computer-aided diagnosis is ongoing but has the potential to reduce inconsistency between operators and imaging centres
Reflection prompts:
- Consider how AI is already used within your own ultrasound imaging practice
- What barriers are there to AI use in ultrasound imaging?
- How can the use of AI in ultrasound imaging benefit the sonographer?
- What further CPD do you need to undertake to keep up with the rapidly changing AI field in ultrasound imaging?
- How can sonographers be sure that the AL tools are safe and fit for purpose in their own clinical practice?
- What are the medico-legal implications for the sonographer when using AI tools as part of their routing practice?
Further Reading
He, F. Wang, Y. Xiu, Y. Zhang, Y. Chen, L. (2021). Artificial Intelligence in Prenatal Ultrasound Diagnosis. Frontiers in Medicine, 8, 729978. https://doi.org/10.3389/fmed.2021.729978
Matthew, J. Skelton, E. Day, T. G. Zimmer, V. A. Gomez, A. Wheeler, G. Toussaint, N. Liu, T. Budd, S. Lloyd, K. Wright, R. Deng, S. Ghavami, N. Sinclair, M. Meng, Q. Kainz, B. Schnabel, J. A. Rueckert, D. Razavi, R. ... Hajnal, J. (2022). Exploring a new paradigm for the fetal anomaly ultrasound scan: Artificial intelligence in real-time. Prenatal Diagnosis, 42(1), 49-59. https://doi.org/10.1002/pd.6059
Society of Radiographers (2021) Artificial intelligence: Guidance for clinical imaging and therapeutic radiography workforce professionals. https://www.sor.org/learning-advice/professional-body-guidance-and-publications/documents-and-publications/policy-guidance-document-library/artificial-intelligence-guidance-for-clinical-imag
Xiao, S. Zhang, J. Zhu, Y. Zhang, Z. Cao, H. Xie, M. Zhang, L. (2023). Application and Progress of Artificial Intelligence in Fetal Ultrasound. Journal of Clinical Medicine, 12(9), 3298. https://doi.org/10.3390/jcm12093298