Synopsis

Keywords: Image acquisition: Image processing, Image acquisition: Sequences, Image acquisition: Motion Correction

Intrauterine growth restriction (IUGR) is the fetal failure to reach its biological growth potential and placental dysfunction is the main cause IUGR is distinguished into Small for Gestational Age (SGA) and Fetal Growth Restriction (FGR). Ultrasound represents the technique of choice for the diagnosis and the Functional Doppler parameters improve this distinction as established by the Expert Consensus Statement. The differentiation is important: many SGAs are constitutionally small but healthy without risk of poor outcomes unlike FGRs.
Magnetic Resonance Imaging (MRI) and Diffusion weighted imaging (DWI) can evaluate placenta and fetal growth supporting the diagnosis of developmental pathologies

Advanced Morphofunctional Sequences & AI in the Evaluation of Fetal Growth Anomalies

Fetal magnetic resonance imaging (MRI) represents as a second level tool of investigation performed after a standard anatomical ultrasonography (US) to provide valuable insights into fetal and placental evaluation. Its advantages, including safety, enhanced tissue contrast resolution, and the ability to overcome ultrasound limitations, make it an invaluable tool. Fetal MRI indications encompass both central nervous system and fetal body evaluations.Predominant central nervous system indications comprise ventriculomegaly, midline malformations, posterior fossa abnormalities, supratentorial parenchymal issues, ischemic hemorrhagic lesions, infections, facial malformations (such as cleft lip/palate), and neural tube defects. Recent technological advancements have introduced new sequences and software capable of interpreting qualitative and quantitative data for diverse pathological conditions and fetal malformations. Additionally, studies confirm the safety of exposing fetuses to high magnetic fields, such as those from 3T scanners.Recent enhancements in fetal MRI sequences, notably diffusion-weighted imaging (DWI) techniques like intravoxel incoherent motion (IVIM), along with the transition to 3-T magnets and the integration of artificial intelligence (AI) and radiomics, are enhancing our ability to study pathological changes in fetuses, particularly those affected by intrauterine growth restriction.Fetal MRI, thanks to DWI techniques such as apparent diffusion coefficient (ADC) maps and IVIM protocols, facilitates quantitative assessments of fetal microstructure and perfusion, complementing US examinations. IVIM allows for precise quantification of perfusion fraction and diffusion coefficient D, eliminating biases associated with low b-values in DWI signals. These advancements are particularly beneficial for studying the highly perfused fetal brain.IVIM allows distinct quantification of perfusion and diffusion in tissues, using a suitable number of b-values: low (b<200 s/mm2) and medium (200<b<1200 s/mm2) and a bi-exponential fit of the signal decay. In particular, the diffusion coefficient D has the advantage of not being affected by the bias due to perfusion as is the ADC coefficient in conventional DWI.In our experience, according to the literature, IVIM-DWI does not result in an increase in SAR which indicates that IVIM examination is feasible in terms of imaging safetyOur work revealed significant differences in diffusion and perfusion properties between IUGR and healthy fetuses.The results demonstrate the role of perfusion fraction (f) parameter in the diagnosis. Actually f can help to discriminate between intrauterine growth restricted and healthy fetuses in the placenta and lung parenchyma.Additionally the diffusion coefficient (D) behavior would reflect the physiological maturational processes of the fetal brain and may detect a delay in brain myelination in intrauterine growth-restricted fetuses. Variants of the IVIM model, like the joint flow-compensated and non-compensated IVIM model, offer improved detection of placental flow alterations in fetuses with fetal growth restriction (FGR). Furthermore, 3D reconstructions of the fetal diaphragm aid in diagnosing congenital diaphragmatic hernias, optimizing surgical planning, and customizing patch designs.The integration of AI software in MRI preprocessing and postprocessing techniques promises automated image acquisition, improved quality, and time-efficient examinations. AI algorithms can predict fetal positions and generate high-resolution 3D images of fetal brain, thorax, and vascular structures, surpassing traditional 2D MRI data in diagnostic quality and visualization.The emerging field of radiomics, involving the extraction of a multitude of data from images using statistical operations, offers insights into tissue characteristics and predictive outcomes. Studies demonstrate the potential of MRI-based placental radiomics in accurately predicting FGR, especially when combined with ultrasound indicators.In conclusion, AI-driven reconstruction algorithms and radiomics represent promising avenues in fetal MRI, offering enhanced imaging capabilities, accurate diagnostic predictions, and improved understanding of fetal pathologies. These advancements hold significant potential in advancing prenatal care and fetal health assessment.

Acknowledgements

My acknowledgement To Silvia Capuani CNR ISC, Physics Department, “Sapienza” University of Rome, Italy for her support in this research