5 Minute Healthtech Jargon Buster: Functional Imaging

by Lillian Hall, Research and Communications Associate

Functional imaging is a powerful non-invasive technique that provides an insight into the mechanisms of brain activity, achieved by tracking alterations in blood flow and oxygenation. Cortical areas which have higher blood flow are detected and used to infer brain activity (Stefano et al.).   This helps researchers and medical professionals, such as psychologists and neurologists, to understand how the brain functions, how diseases affect it, and how it recovers from injury.

Several steps must be undertaken prior to the data being used for research or medical purposes. The first step, called preprocessing, transforms raw scanner data into clear brain images. Since the scan captures different slices of the brain at slightly different times, slice-timing correction adjusts them, so they appear synchronised (Smith).

Figure 1 Segmentation of a foetal MRI scan using AI.  (Lo et al.)

Next, motion correction ensures that all brain images align properly, even if the subject moved slightly during the scan. To reduce noise and improve clarity, researchers apply spatial smoothing, which slightly blurs the images without losing important details. Finally, intensity normalization adjusts brightness levels across all images so that the data is consistent. (Smith) 

As Artificial intelligence (AI) continues to advance, it is playing an increasingly important role in automating and improving these processes, making functional imaging more efficient and insightful than ever before. The uses and challenges of its applications are detailed below.

Where to find out more

Romilly Life Sciences can offer several decades experience leading the validation, regulatory approval and implementation of novel technologies including the pivotal development of diagnostic tools and biomarkers based on functional imaging data.

To find out how you can reach patients faster, backed by compelling evidence, contact us.

References

Bassett, Mike. “RSNA’s QIBA: A Quantitative Success.” Rsna.org, 2025, www.rsna.org/news/2021/april/qiba-a-quantitative-success. Accessed 19 Mar. 2025.

Chen, Zhe Sage, et al. “Modern Views of Machine Learning for Precision Psychiatry.” Patterns, vol. 3, no. 11, 11 Nov. 2022, p. 100602, www.sciencedirect.com/science/article/pii/S2666389922002276#abs0010, https://doi.org/10.1016/j.patter.2022.100602. Accessed 26 Feb. 2023.

Larson, David B., et al. “Regulatory Frameworks for Development and Evaluation of Artificial Intelligence–Based Diagnostic Imaging Algorithms: Summary and Recommendations.” Journal of the American College of Radiology, vol. 0, no. 0, 20 Oct. 2020, www.jacr.org/article/S1546-1440(20)31020-6/fulltext, https://doi.org/10.1016/j.jacr.2020.09.060. Accessed 24 Jan. 2021.

Lo, Justin, et al. “Cross Attention Squeeze Excitation Network (CASE-Net) for Whole Body Fetal MRI Segmentation.” Sensors, vol. 21, no. 13, 30 June 2021, p. 4490, https://doi.org/10.3390/s21134490. Accessed 19 Mar. 2025.

Meshaka, Riwa, et al. “Artificial Intelligence Applied to Fetal MRI: A Scoping Review of Current Research.” The British Journal of Radiology, 18 Mar. 2022, https://doi.org/10.1259/bjr.20211205. Accessed 20 May 2022.

Ronneberger, Olaf, et al. “U-Net: Convolutional Networks for Biomedical Image Segmentation.” ArXiv.org, 18 May 2015, arxiv.org/abs/1505.04597.

Ryan, et al. “US FDA Approval of Pediatric Artificial Intelligence and Machine Learning–Enabled Medical Devices.” JAMA Pediatrics, 16 Dec. 2024, https://doi.org/10.1001/jamapediatrics.2024.5437. Accessed 31 Dec. 2024.

Smith, S M. “Overview of FMRI Analysis.” The British Journal of Radiology, vol. 77, no. suppl_2, Dec. 2004, pp. S167–S175, cfn.upenn.edu/stslab/wiki/lib/exe/fetch.php/fmri_club:preprocess1:smith_2004_brjrad.pdf, https://doi.org/10.1259/bjr/33553595.

Stefano, Valeria Di, et al. “Decoding Schizophrenia: How AI-Enhanced FMRI Unlocks New Pathways for Precision Psychiatry.” Brain Sciences, vol. 14, no. 12, 27 Nov. 2024, pp. 1196–1196, www.mdpi.com/2076-3425/14/12/1196, https://doi.org/10.3390/brainsci14121196. Accessed 6 Dec. 2024.

Vahedifard, Farzan, et al. “Artificial Intelligence in Fetal Resting-State Functional MRI Brain Segmentation: A Comparative Analysis of 3D UNet, VNet, and HighRes-Net Models.” ArXiv.org, 2023, arxiv.org/abs/2311.10844. Accessed 19 Mar. 2025.