Three-dimensional prenatal ultrasound scans of a baby's facial features have become increasingly popular among parents in both private and clinical settings. Ultrasound practitioners often draw on their experience to identify factors that influence image quality when discussing scan outcomes with parents. This study aims to identify the maternal, fetal and technical factors that may affect the quality of three-dimensional souvenir face images during ultrasound scan. A retrospective quality review study was performed with data from a single-centre research study, including ultrasound videos of the fetal growth scans, three-dimensional facial ultrasound acquisitions and post-processing steps. A total of 342 three-dimensional surface-rendered images were attempted from 41 singleton pregnancy subjects, average gestational age 26.69 weeks, range: 21-30. The retrospective image quality for all images was assessed by two observers. Univariable ordinal regression test was used to investigate the associations between demographic/technical factors and the best image quality acheiveable. Of the 41 pregnancies, three-dimensional acquisition time was an average of 03:07 (mm:ss), (range: 01:22-5:31). In total, 49% of women had at least one good or moderate quality image, and 51% women had a poor quality or failed three-dimensional scan as the best quality possible. Image quality was associated with placenta site, explaining 18% of the variation (p < 0.05). We found a maternal-fetal factor which has a high impact on three-dimensional image quality of the prenatal face but nonetheless, sonographer skill, training and other technical factors may be employed to minimise the impact of detrimental factors.

ABSTRACT Objective To use artificial intelligence (AI) to automatically extract video clips of the fetal heart from a stream of ultrasound video, and to assess the performance of these when used for remote second review. Methods Using a dataset from a previous clinical trial of AI to assist in fetal ultrasound scanning, AI was used to automatically extract video clips of the fetal heart from ultrasound scans of 48 fetuses in which the diagnosis was known: 24 normal and 24 with congenital heart disease (CHD). These, and manually still saved images, were shown in a random order to expert clinicians, who were asked to detect cardiac abnormalities. Results The initial manual scan had a sensitivity of 0.792 and specificity of 0.917 for detecting CHD in this cohort. The addition of second review improved the sensitivity to 0.975 using video clips, which was significantly higher than using still images (0.892, p = 0.002). There was a significant drop in specificity to 0.767 and 0.833 ( p < 0.001) for the video and still method, respectively, which were statistically similar to each other ( p = 0.117). The median review time was 1.0 min (IQR 0.71) for the still images, and 3.75 min (IQR 3.12) for the AI‐generated video clips. Conclusion AI can be used to automatically extract fetal cardiac video clips, and these can be used for remote second review to improve detection rates. Video clips are superior to still images, but both methods result in a significant drop in specificity.

Accurate analysis of prenatal ultrasound (US) is essential for early detection of developmental anomalies. However, operator dependency and technical limitations (e.g. intrinsic artefacts and effects, setting errors) can complicate image interpretation and the assessment of diagnostic uncertainty. We present L-FUSION (Laplacian Fetal US Segmentation with Integrated FoundatiON models), a framework that integrates uncertainty quantification through unsupervised, normative learning and large-scale foundation models for robust segmentation of fetal structures in normal and pathological scans. We propose to utilise the aleatoric logit distributions of Stochastic Segmentation Networks and Laplace approximations with fast Hessian estimations to estimate epistemic uncertainty only from the segmentation head. This enables us to achieve reliable abnormality quantification for instant diagnostic feedback. Combined with an integrated Dropout component, L-FUSION enables reliable differentiation of lesions from normal fetal anatomy with enhanced uncertainty maps and segmentation counterfactuals in US imaging. It improves epistemic and aleatoric uncertainty interpretation and removes the need for manual disease-labelling. Evaluations across multiple datasets show that L-FUSION achieves superior segmentation accuracy and consistent uncertainty quantification, supporting on-site decision-making and offering a scalable solution for advancing fetal ultrasound analysis in clinical settings. Code is available at https://github.com/ividja/L-FUSION .

The current approach to fetal anomaly screening is based on biometric measurements derived from individually selected ultrasound images. In this paper, we introduce a paradigm shift that attains human-level performance in biometric measurement by aggregating automatically extracted biometrics from every frame across an entire scan, with no need for operator intervention. We use a neural network to classify each frame of an ultrasound video recording. We then measure fetal biometrics in every frame where appropriate anatomy is visible. We use a Bayesian method to estimate the true value of each biometric from a large number of measurements and probabilistically reject outliers. We performed a retrospective experiment on 1457 recordings (comprising 48 million frames) of 20-week ultrasound scans, estimated fetal biometrics in those scans and compared our estimates to real-time manual measurements. Our method achieves human-level performance in estimating fetal biometrics and estimates well-calibrated credible intervals for the true biometric value.

Advances in fetal three-dimensional (3D) ultrasound (US) and magnetic resonance imaging (MRI) have revolutionized the study of fetal brain development, enabling detailed analysis of brain structures and growth. Despite their complementary capabilities, these modalities capture fundamentally different physical signals, potentially leading to systematic differences in image-derived phenotypes (IDPs). Here, we evaluate the agreement of IDPs between US and MRI by comparing the volumes of eight brain structures from 90 subjects derived using deep-learning algorithms from majority same-day imaging (days between scans: mean = 1.2, mode = 0 and max = 4). Excellent agreement (intra-class correlation coefficient, ICC > 0.75 $$ ICC>0.75 $$ ) was observed for the cerebellum, cavum septum pellucidum, thalamus, white matter and deep grey matter volumes, with significant correlations p < 0.001 $$ \left(p<0.001\right) $$ for most structures, except the ventricular system. Bland-Altman analysis revealed some systematic biases: intracranial and cortical plate volumes were larger on US than MRI, by an average of 35 cm 3 $$ 35\ {\mathrm{cm}}^3 $$ and 4.1 cm 3 $$ 4.1\ {\mathrm{cm}}^3 $$ , respectively. Finally, we found the labels of the brainstem and ventricular system were not comparable between the modalities. These findings highlight the necessity of structure-specific adjustments when interpreting fetal brain IPDs across modalities and underscore the complementary roles of US and MRI in advancing fetal neuroimaging.

Background Artificial intelligence (AI) has shown potential in improving the performance of screening fetal anomaly ultrasound scans. We aimed to assess the effect of AI on fetal ultrasound scanning, in terms of diagnostic performance, biometry, scan duration, and sonographer cognitive load. Methods This was a randomised, single centre, open label trial in a large teaching hospital. Pregnant participants with fetal congenital heart disease (CHD) and with healthy fetuses were recruited and scanned with both methods. Screening sonographers were recruited from regional hospitals and were randomised to scan with the AI tool or in the standard fashion, blinded to the fetal CHD status. For the AI-assisted scans, the AI models identified and saved 13 standard image planes, and measured four biometrics. Findings 78 pregnant participants (26 with fetal CHD) and 58 sonographers were recruited. The sensitivity and specificity of the AI-assisted scan in detecting fetal malformation was 88.9% and 98.0% respectively, with the standard scan achieving 81.5% and 92.2% (not significant). AI-assisted scans were significantly shorter than standard scans (median 11.4 min vs 19.7 min, p <0.001). Sonographer cognitive load was significantly lower in the AI-assisted group (median NASA TLX score 35.2 vs 46.5, p <0.001). For all biometrics, the AI repeatability and reproducibility was superior to manual measurements. Interpretation AI assistance in the routine fetal anomaly ultrasound scan results in a significant time saving, along with a reduction in sonographer cognitive load, without a reduction in diagnostic performance. Funding The study was funded by an NIHR doctoral fellowship (NIHR301448) and was supported by grants from the Wellcome Trust (IEH Award, 102431), by core funding from the Wellcome Trust/EPSRC Centre for Medical Engineering (WT203148/Z/16/Z), and the London AI Centre for Value Based Healthcare via funding from the Office for Life Sciences.### Competing Interest StatementTD, JM, SB, LV, RW, AF, JH, BK, and RR are co-founders and shareholders of Fraiya Ltd, a University-NHS spinout company that is aiming to commercialise an AI tool for use in the screening obstetric ultrasound scan.### Clinical TrialISRCTN 65824874### Funding StatementThe study was funded by an NIHR doctoral fellowship (NIHR301448) and was supported by grants from the Wellcome Trust (IEH Award, 102431), by core funding from the Wellcome Trust/EPSRC Centre for Medical Engineering (WT203148/Z/16/Z), and the London AI Centre for Value Based Healthcare via funding from the Office for Life Sciences. BK received funding by the ERC - project MIA-NORMAL 101083647 ### Author DeclarationsI confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained.YesThe details of the IRB/oversight body that provided approval or exemption for the research described are given below:The London Dulwich Research Ethics Committee of The Health Research Authority gave ethical approval for this work (reference 22/LO/0163).I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals.YesI understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance).YesI have followed all appropriate research reporting guidelines, such as any relevant EQUATOR Network research reporting checklist(s) and other pertinent material, if applicable.YesPatient-level imaging data from the trial, and the imaging data used to train the AI models are not available for sharing due to ethical restrictions. The study protocol, patient information sheet, and example consent forms are available on request. The code used to train the AI models is available on request, but the model weights used in the trial are not available.

This study explores the potential of 3D Slice-to-Volume Registration (SVR) motion-corrected fetal MRI for craniofacial assessment, traditionally used only for fetal brain analysis. In addition, we present the first description of an automated pipeline based on 3D Attention UNet trained for 3D fetal MRI craniofacial segmentation, followed by surface refinement. Results of 3D printing of selected models are also presented. Qualitative analysis of multiplanar volumes, based on the SVR output and surface segmentations outputs, were assessed with computer and printed models, using standardised protocols that we developed for evaluating image quality and visibility of diagnostic craniofacial features. A test set of 25, postnatally confirmed, Trisomy 21 fetal cases (24–36 weeks gestational age), revealed that 3D reconstructed T2 SVR images provided 66–100% visibility of relevant craniofacial and head structures in the SVR output, and 20–100% and 60–90% anatomical visibility was seen for the baseline and refined 3D computer surface model outputs respectively. Furthermore, 12 of 25 cases, 48%, of refined surface models demonstrated good or excellent overall quality with a further 9 cases, 36%, demonstrating moderate quality to include facial, scalp and external ears. Additional 3D printing of 12 physical real-size models (20–36 weeks gestational age) revealed good/excellent overall quality in all cases and distinguishable features between healthy control cases and cases with confirmed anomalies, with only minor manual adjustments required before 3D printing. Despite varying image quality and data heterogeneity, 3D T2w SVR reconstructions and models provided sufficient resolution for the subjective characterisation of subtle craniofacial features. We also contributed a publicly accessible online 3D T2w MRI atlas of the fetal head, validated for accurate representation of normal fetal anatomy. Future research will focus on quantitative analysis, optimizing the pipeline, and exploring diagnostic, counselling, and educational applications in fetal craniofacial assessment.

OBJECTIVES:Evaluating craniofacial phenotype-genotype correlations prenatally is increasingly important; however, it is subjective and challenging with 3D ultrasound. We developed an automated label propagation pipeline using 3D motion- corrected, slice-to-volume reconstructed (SVR) fetal MRI for craniofacial measurements. METHODS:A literature review and expert consensus identified 31 craniofacial biometrics for fetal MRI. An MRI atlas with defined anatomical landmarks served as a template for subject registration, auto-labelling, and biometric calculation. We assessed 108 healthy controls and 24 fetuses with Down syndrome (T21) in the third trimester (29-36 weeks gestational age, GA) to identify meaningful biometrics in T21. Reliability and reproducibility were evaluated in 10 random datasets by four observers. RESULTS:Automated labels were produced for all 132 subjects with a 0.3% placement error rate. Seven measurements, including anterior base of skull length and maxillary length, showed significant differences with large effect sizes between T21 and control groups (ANOVA, p<0.001). Manual measurements took 25-35 minutes per case, while automated extraction took approximately 5 minutes. Bland-Altman plots showed agreement within manual observer ranges except for mandibular width, which had higher variability. Extended GA growth charts (19-39 weeks), based on 280 control fetuses, were produced for future research. CONCLUSION:This is the first automated atlas-based protocol using 3D SVR MRI for fetal craniofacial biometrics, accurately revealing morphological craniofacial differences in a T21 cohort. Future work should focus on improving measurement reliability, larger clinical cohorts, and technical advancements, to enhance prenatal care and phenotypic characterisation.

Objectives Prenatal characterisation of craniofacial development remains a challenge for ultrasound.¹ We sought to develop an MRI protocol for the automated extraction of craniofacial measurements using 3D motion-corrected, slice-to-volume reconstructed (SVR) fetal MRI² and atlas-based label propagation of anatomical landmarks. Methods 24 fetuses with genetically confirmed Down syndrome (DS) and 85 control fetuses were retrospectively selected if: scanned between 29–37 weeks GA; had maternal written informed consent via; fetal MRI [REC:07/H0707/105], dHCP [REC:14/LO/1169], PiP [REC:16/LO/1573], eBIDS [REC:19/LO/0667], or iFIND [REC:14/LO/1806]; had a 1.5T or 3T MRI protocol amenable to SVR from 2D acquisitions; and, if the reconstruction quality score was ‘good/excellent’. Using a control dataset, 4D spatiotemporal atlases were developed for 16 discrete time-points from 21–36 weeks GA range.³ A clinician reviewed the literature and 46 fetal MRI-reliable craniofacial landmarks for biometry were labelled in three atlases using research software (ITK-SNAP).4 The label propagation pipeline was followed by the calculation of the distances between selected landmark centre-points. The performance was tested on five datasets with DS, by comparing manual measurements to the automated distances. Lastly, we investigated the feasibility of this approach by comparing the automated DS biometry to the control groups with different acquisition protocols (fig 1). Results The automated craniofacial anatomical landmarks were visually assessed for accuracy. No landmarks in the control group required modification. However, in the DS group 4 out of 120 automated landmarks required minor manual adjustment. Automated biometry, compared to manual measurements, showed small mean paired relative errors of <10%, except for the foramen magnum measurements (figure 2). The differences were primarily caused by variability in multiplanar manual adjustment of images and suboptimal regional visibility of finer features. The process of verifying correct positioning of landmarks was significantly faster than extracting manual biometry (5 vs 25 minutes/case). There were no significant differences in measurements within the control cohorts and between different acquisition parameters (1.5T, 3T; TE=80ms, TE=180ms, TE=250ms – see figure 2). However, there were significant differences between DS and control cohorts in the OFD, ASBL and HPL distances (ANOVA, p<0.001). These differences are likely associated with shorter/wider skulls (brachycephaly) and smaller mid-facies (midface hypoplasia) in DS, which is consistent with ultrasound and neonatal findings.5 Conclusion We present the first automated atlas-based label propagation protocol using 3D motion-corrected MRI for 12 fetal craniofacial measurements across varied echo times and field strengths. The method shows differences in craniofacial growth between fetuses with Down syndrome and control subjects. References Mak ASL, Leung KY. Prenatal ultrasonography of craniofacial abnormalities. Ultrasonography 2019. doi:10.14366/usg.18031 Kuklisova-Murgasova M, Quaghebeur G, Rutherford MA, Hajnal J V., Schnabel JA. Reconstruction of fetal brain MRI with intensity matching and complete outlier removal. Med Image Anal 2012; 16: 1550. Uus A, Matthew J, Grigorescu I, Jupp S, Grande LC, Price A, et al. Spatio-Temporal Atlas of Normal Fetal Craniofacial Feature Development and CNN-Based Ocular Biometry for Motion-Corrected Fetal MRI. Lect Notes Comput Sci (including Subser Lect Notes Artif Intell Lect Notes Bioinformatics) 2021; 12959 LNCS: 168–178. ITK-SNAP tool. http://www.itksnap.org/pmwiki/pmwiki.php Vicente A, Bravo-González LA, López-Romero A, Muñoz CS, Sánchez-Meca J. Craniofacial morphology in down syndrome: a systematic review and meta-analysis. Sci Reports 2020 101 2020;10:1–14.

We present PRETUS - a Plugin-based Real Time UltraSound software platform for live ultrasound image analysis and operator support. The software is lightweight; functionality is brought in via independent plug-ins that can be arranged in sequence. The software allows to capture the real-time stream of ultrasound images from virtually any ultrasound machine, applies computational methods and visualizes the results on-the-fly. Plug-ins can run concurrently without blocking each other. They can be implemented in C++ and Python. A graphical user interface can be implemented for each plug-in, and presented to the user in a compact way. The software is free and open source, and allows for rapid prototyping and testing of real-time ultrasound imaging methods in a manufacturer-agnostic fashion. The software is provided with input, output and processing plug-ins, as well as with tutorials to illustrate how to develop new plug-ins for PRETUS.

Automatic segmentation of the placenta in fetal ultrasound (US) is challenging due to the (i) high diversity of placenta appearance, (ii) the restricted quality in US resulting in highly variable reference annotations, and (iii) the limited field-of-view of US prohibiting whole placenta assessment at late gestation. In this work, we address these three challenges with a multi-task learning approach that combines the classification of placental location (e.g., anterior, posterior) and semantic placenta segmentation in a single convolutional neural network. Through the classification task the model can learn from larger and more diverse datasets while improving the accuracy of the segmentation task in particular in limited training set conditions. With this approach we investigate the variability in annotations from multiple raters and show that our automatic segmentations (Dice of 0.86 for anterior and 0.83 for posterior placentas) achieve human-level performance as compared to intra- and inter-observer variability. Lastly, our approach can deliver whole placenta segmentation using a multi-view US acquisition pipeline consisting of three stages: multi-probe image acquisition, image fusion and image segmentation. This results in high quality segmentation of larger structures such as the placenta in US with reduced image artifacts which are beyond the field-of-view of single probes.

The Developing Human Connectome Project has created a large open science resource which provides researchers with data for investigating typical and atypical brain development across the perinatal period. It has collected 1228 multimodal magnetic resonance imaging (MRI) brain datasets from 1173 fetal and/or neonatal participants, together with collateral demographic, clinical, family, neurocognitive and genomic data from 1173 participants, together with collateral demographic, clinical, family, neurocognitive and genomic data. All subjects were studied in utero and/or soon after birth on a single MRI scanner using specially developed scanning sequences which included novel motion-tolerant imaging methods. Imaging data are complemented by rich demographic, clinical, neurodevelopmental, and genomic information. The project is now releasing a large set of neonatal data; fetal data will be described and released separately. This release includes scans from 783 infants of whom: 583 were healthy infants born at term; as well as preterm infants; and infants at high risk of atypical neurocognitive development. Many infants were imaged more than once to provide longitudinal data, and the total number of datasets being released is 887. We now describe the dHCP image acquisition and processing protocols, summarize the available imaging and collateral data, and provide information on how the data can be accessed.

OBJECTIVE:Advances in artificial intelligence (AI) have demonstrated potential to improve medical diagnosis. We piloted the end-to-end automation of the mid-trimester screening ultrasound scan using AI-enabled tools. METHODS:A prospective method comparison study was conducted. Participants had both standard and AI-assisted US scans performed. The AI tools automated image acquisition, biometric measurement, and report production. A feedback survey captured the sonographers' perceptions of scanning. RESULTS:Twenty-three subjects were studied. The average time saving per scan was 7.62 min (34.7%) with the AI-assisted method (p < 0.0001). There was no difference in reporting time. There were no clinically significant differences in biometric measurements between the two methods. The AI tools saved a satisfactory view in 93% of the cases (four core views only), and 73% for the full 13 views, compared to 98% for both using the manual scan. Survey responses suggest that the AI tools helped sonographers to concentrate on image interpretation by removing disruptive tasks. CONCLUSION:Separating freehand scanning from image capture and measurement resulted in a faster scan and altered workflow. Removing repetitive tasks may allow more attention to be directed identifying fetal malformation. Further work is required to improve the image plane detection algorithm for use in real time.

Motion-corrected fetal magnetic resonance imaging (MRI) is widely employed in large-scale fetal brain studies. However, the current processing pipelines and spatio-temporal atlases tend to omit craniofacial structures, which are known to be linked to genetic syndromes. In this work, we present the first spatio-temporal atlas of the fetal head that includes craniofacial features and covers 21 to 36 weeks gestational age range. Additionally, we propose a fully automated pipeline for fetal ocular biometry based on a 3D convolutional neural network (CNN). The extracted biometric indices are used for the growth trajectory analysis of changes in ocular metrics for 253 normal fetal subjects from the developing human connectome project (dHCP).

Automatic segmentation of the placenta in fetal ultrasound (US) is challenging due to its high variations in shape, position and appearance. Convolutional neural networks (CNN) are the state-of-the-art in medical image segmentation and have already been applied successfully to extract the placenta in US. However, the performance of CNNs depends highly on the availability of large training sets which also need to be representative for new unseen data. In this work, we propose to inform the network about the variability in the data distribution via an auxiliary task to improve performances for under representative training sets. The auxiliary task has two objectives: (i) enlarging of the training set with easily obtainable labels, and (ii) including more information about the variability of the data in the training process. In particular, we use transfer learning and multi-task learning to incorporate the placental position in a U-Net architecture. We test different models for the segmentation of anterior and posterior placentas in fetal US. Our results suggest that these placenta types represent different distributions. By including the position of the placenta as an auxiliary task, the segmentation accuracy for both anterior and posterior placentas is improved when the specific type of placenta is not included in the training set.

Ultrasound (US) images suffer from artefacts which limit its diagnostic value, notably acoustic shadow. Shadows are dependent on probe orientation, with each view giving a distinct, partial view of the anatomy. In this work, we fuse the partially imaged fetal head anatomy, acquired from numerous views, into a single coherent compounding of the full anatomy. Firstly, a stream of freehand 3D US images is acquired, capturing as many different views as possible. The imaged anatomy at each time-point is then independently aligned to a canonical pose using an iterative spatial transformer network (iSTN), making our approach robust to fast fetal and probe motion. Secondly, images are fused by averaging only the best (most salient) features from all images, producing a more detailed compounding. Finally, the compounding is iteratively refined using a groupwise registration approach. We evaluate our compounding approach quantitatively and qualitatively, comparing it with average compounding and individual US frames. We also evaluate our alignment accuracy using two physically attached probes, that capture separate views simultaneously, providing ground-truth. Lastly, we demonstrate the potential clinical impact of our method for assessing cranial, facial and external ear abnormalities, with automated atlas-based masking and 3D volume rendering.

We propose an image reconstruction framework to combine a large number of overlapping image patches into a fused reconstruction of the object of interest, that is robust to inconsistencies between patches (e.g. motion artefacts) without explicitly modelling them. This is achieved through two mechanisms: first, manifold embedding, where patches are distributed on a manifold with similar patches (where similarity is defined only in the region where they overlap) closer to each other. As a result, inconsistent patches are set far apart in the manifold. Second, fusion, where a sample in the manifold is mapped back to image space, combining features from all patches in the region of the sample. For the manifold embedding mechanism, a new method based on a Convolutional Variational Autoencoder ( β -VAE) is proposed, and compared to classical manifold embedding techniques: linear (Multi Dimensional Scaling) and non-linear (Laplacian Eigenmaps). Experiments using synthetic data and on real fetal ultrasound images yield fused images of the whole fetus where, in average, β -VAE outperforms all the other methods in terms of preservation of patch information and overall image quality.

We propose a method to extract the human placenta at late gestation using multi-view 3D US images. This is the first step towards automatic quantification of placental volume and morphology from US images along the whole pregnancy beyond early stages (where the entire placenta can be captured with a single 3D US image). Our method uses 3D US images from different views acquired with a multi-probe system. A whole placenta segmentation is obtained from these images by using a novel technique based on 3D convolutional neural networks. We demonstrate the performance of our method on 3D US images of the placenta in the last trimester. We achieve a high Dice overlap of up to 0.8 with respect to manual annotations, and the derived placental volumes are comparable to corresponding volumes extracted from MR.

We propose a method for constructing a spatio-temporal cortical surface atlas of neonatal brains aged between 36 and 44 weeks of post-menstrual age (PMA) at the time of scan. The data were acquired as part of the Developing Human Connectome Project (dHCP), and the constructed surface atlases are publicly available. The method is based on a spherical registration approach: Multimodal Surface Matching (MSM), using cortical folding for driving the alignment. Templates have been generated for the anatomical cortical surface and for the cortical feature maps: sulcal depth, curvature, thickness, T1w/T2w myelin maps and cortical regions. To achieve this, cortical surfaces from 270 infants were first projected onto the sphere. Templates were then generated in two stages: first, a reference space was initialised via affine alignment to a group average adult template. Following this, templates were iteratively refined through repeated alignment of individuals to the template space until the variability of the average feature sets converged. Finally, bias towards the adult reference was removed by applying the inverse of the average affine transformations on the template and de-drifting the template. We used temporal adaptive kernel regression to produce age-dependant atlases for 9 weeks (36-44 weeks PMA). The generated templates capture expected patterns of cortical development including an increase in gyrification as well as an increase in thickness and T1w/T2w myelination with increasing age.