Three-dimensional (3D) facial shape analysis plays an increasingly important role in biomedical and clinical research. However, the high cost and limited accessibility of advanced 3D acquisition systems constrain their widespread use, fostering the development of low-cost alternatives for facial model acquisition and reconstruction. This study presents a comprehensive evaluation methodology that integrates geometric accuracy metrics with landmark-based Geometric Morphometrics, providing a statistically robust and anatomically meaningful framework for assessing facial morphology preservation. The proposed methodology was validated through a comparative study using high-resolution stereophotogrammetry (SPG) as the gold standard, a smartphone-based infrared structured light scanner, and state-of-the-art deep learning approaches for 3D reconstruction from 2D images. Quantitative and morphometric results demonstrated that smartphone-based scans achieved the highest geometric and morphological fidelity, with over 80

Individuals with Down syndrome (DS) have a high risk of obstructive sleep apnea (OSA), likely driven by craniofacial and upper airway abnormalities. However, three-dimensional upper airway morphology and its association with OSA severity remain poorly characterized. Here, we provide the first comprehensive three-dimensional characterization of upper airway volume and shape in adults with DS and euploid controls (EU). Three-dimensional upper airway models were reconstructed from T1-weighted MRI scans of 164 adults (63 EU, 101 DS). Upper airway, nasal cavity, and pharyngeal morphology were quantified using volumetry and geometric morphometrics (23 landmarks). OSA associations were assessed in a DS subsample (n = 91) using the apnea-hypopnea index and OSA severity categories. Adults with DS exhibited significantly reduced upper airway, nasal cavity, and pharyngeal volumes and distinct shape alterations relative to EU controls, with sex-specific patterns. In DS, increasing OSA severity was associated with reduced upper airway and nasal cavity volumes, particularly in males. OSA-related shape differences were confined to the nasal cavity of males with severe OSA, characterized by bilateral constriction and anteroposterior shortening. These findings identify upper airway morphology, particularly nasal cavity volume and shape, as imaging-derived anatomical biomarkers associated with OSA severity in adults with DS.

Generative models have emerged as powerful tools in medical imaging, enabling tasks such as segmentation, anomaly detection, and high-quality synthetic data generation. These models typically rely on learning meaningful latent representations, which are particularly valuable given the high-dimensional nature of 3D medical images like brain magnetic resonance imaging (MRI) scans. Despite their potential, latent representations remain underexplored in terms of their structure, information content, and applicability to downstream clinical tasks. Investigating these representations is crucial for advancing the use of generative models in neuroimaging research and clinical decision-making. In this work, we develop multiple variational autoencoders (VAEs) to encode 3D brain MRI scans into compact latent space representations for generative and predictive applications. We systematically evaluate the effectiveness of the learned representations through three key analyses: (i) a quantitative and qualitative assessment of MRI reconstruction quality, (ii) a visualisation of the latent space structure using Principal Component Analysis, and (iii) downstream classification tasks on a proprietary dataset of euploid and Down syndrome individuals brain MRI scans. Our results demonstrate that the VAE successfully captures essential brain features while maintaining high reconstruction fidelity. The latent space exhibits clear clustering patterns, particularly in distinguishing individuals with Down syndrome from euploid controls.

Three-dimensional (3D) facial shape analysis has gained interest due to its potential clinical applications. However, the high cost of advanced 3D facial acquisition systems limits their widespread use, driving the development of low-cost acquisition and reconstruction methods. This study introduces a novel evaluation methodology that goes beyond traditional geometry-based benchmarks by integrating morphometric shape analysis techniques, providing a statistical framework for assessing facial morphology preservation. As a case study, we compare smartphone-based 3D scans with state-of-the-art deep learning reconstruction methods from 2D images, using high-end stereophotogrammetry models as ground truth. This methodology enables a quantitative assessment of global and local shape differences, offering a biologically meaningful validation approach for low-cost 3D facial acquisition and reconstruction techniques.

BACKGROUND AND OBJECTIVES:Facial dysmorphologies have emerged as potential critical indicators in the diagnosis and prognosis of genetic, psychotic, and rare disorders. While some conditions present with severe dysmorphologies, others exhibit subtler traits that may not be perceivable to the human eye, requiring the use of precise quantitative tools for accurate identification. Manual annotation remains time-consuming and prone to inter- and intra-observer variability. Existing tools provide partial solutions, but no end-to-end automated pipeline integrates the full process of 3D facial biomarker extraction from magnetic resonance imaging. METHODS AND RESULTS:We introduce BioFace3D, an open-source pipeline designed to automate the discovery of potential 3D facial biomarkers from magnetic resonance imaging. BioFace3D consists of three automated modules: (i) 3D facial model extraction from magnetic resonance images, (ii) deep learning-based registration of homologous anatomical landmarks, and (iii) computation of geometric morphometric biomarkers from landmark coordinates. CONCLUSIONS:The evaluation of BioFace3D is performed both at a global level and within each individual module, through a series of exhaustive experiments using proprietary and public datasets, demonstrating the robustness and reliability of the results obtained by the tool. Source code, along with trained models, can be found at https://bitbucket.org/cv_her_lasalle/bioface3d.

Three-dimensional (3D) facial shape analysis has gained interest due to its potential clinical applications. However, the high cost of advanced 3D facial acquisition systems limits their widespread use, driving the development of low-cost acquisition and reconstruction methods. This study introduces a novel evaluation methodology that goes beyond traditional geometry-based benchmarks by integrating morphometric shape analysis techniques, providing a statistical framework for assessing facial morphology preservation. As a case study, we compare smartphone-based 3D scans with state-of-the-art deep learning reconstruction methods from 2D images, using high-end stereophotogrammetry models as ground truth. This methodology enables a quantitative assessment of global and local shape differences, offering a biologically meaningful validation approach for low-cost 3D facial acquisition and reconstruction techniques.

Generative models have emerged as powerful tools in medical imaging, enabling tasks such as segmentation, anomaly detection, and high-quality synthetic data generation. These models typically rely on learning meaningful latent representations, which are particularly valuable given the high-dimensional nature of 3D medical images like brain magnetic resonance imaging (MRI) scans. Despite their potential, latent representations remain underexplored in terms of their structure, information content, and applicability to downstream clinical tasks. Investigating these representations is crucial for advancing the use of generative models in neuroimaging research and clinical decision-making. In this work, we develop a variational autoencoder (VAE) to encode 3D brain MRI scans into a compact latent space for generative and predictive applications. We systematically evaluate the effectiveness of the learned representations through three key analyses: (i) a qualitative assessment of MRI reconstruction quality, (ii) a visualization of the latent space structure using Principal Component Analysis, and (iii) different down-stream classification tasks on a proprietary dataset of euploid and Down syndrome individuals brain MRI scans. Our results demonstrate that the VAE successfully captures essential brain features while maintaining high reconstruction fidelity. The latent space exhibits clear clustering patterns, particularly in distinguishing euploid subjects from persons with Down syndrome. Furthermore, classification experiments on this latent space reveal the potential of generative models for encoding biologically relevant brain anatomical features, facilitating research on disorders with associated neuroanatomical alterations.

Brain imaging has allowed neuroscientists to analyze brain morphology in genetic and neurodevelopmental disorders, such as Down syndrome, pinpointing regions of interest to unravel the neuroanatomical underpinnings of cognitive impairment and memory deficits. However, the connections between brain anatomy, cognitive performance and comorbidities like Alzheimer’s disease are still poorly understood in the Down syndrome population. The latest advances in artificial intelligence constitute an opportunity for developing automatic tools to analyze large volumes of brain magnetic resonance imaging scans, overcoming the bottleneck of manual analysis. In this study, we propose the use of generative models for detecting brain alterations in people with Down syndrome affected by various degrees of neurodegeneration caused by Alzheimer’s disease. To that end, we evaluate state-of-the-art brain anomaly detection models based on Variational Autoencoders and Diffusion Models, leveraging a proprietary dataset of brain magnetic resonance imaging scans. Following a comprehensive evaluation process, our study includes several key analyses. First, we conducted a qualitative evaluation by expert neuroradiologists. Second, we performed both quantitative and qualitative reconstruction fidelity studies for the generative models. Third, we carried out an ablation study to examine how the incorporation of histogram post-processing can enhance model performance. Finally, we executed a quantitative volumetric analysis of subcortical structures. Our findings indicate that some models effectively detect the primary alterations characterizing Down syndrome’s brain anatomy, including a smaller cerebellum, enlarged ventricles, and cerebral cortex reduction, as well as the parietal lobe alterations caused by Alzheimer’s disease. These results provide preliminary evidence supporting the automatic, data-driven discovery of brain biomarkers for Down syndrome and its associated comorbidities.

Up to 30-40% of genetic and rare disorders present with unique facial patterns. Clinical geneticists have traditionally assessed facial traits to suggest a first diagnosis and direct confirmatory genetic testing. However, to enhance the accuracy of early diagnosis using facial biomarkers, it is crucial to analyze the complexity of facial dysmorphologies using 3D technologies, further understand how facial ontogeny is altered within each condition, and broaden the analysis to include diverse human populations. In this study, we analyzed the 3D facial phenotypes associated to Down (DS), Morquio (MS), Noonan (NS), and Neurofibromatosis type 1 (NF1) syndromes, in an admixed Latin American population from Colombia, including 47 patients and 49 age matched controls. For each subject, we generated a three-dimensional facial model from 2D images captured by a multi-camera photogrammetric system, and recorded the 3D coordinates of 21 anatomical landmarks. Using geometric morphometrics, we characterized the 3D facial dysmorphologies associated with each syndrome and assessed the range of variation across syndromes as compared to controls. Finally, we tested whether these syndromes alter the ontogenetic trajectory of facial growth. Our results confirmed statistically significant facial shape differences associated with these genetic conditions, except for NF1. Consistent with our previous 2D analyses, we identified population-specific facial features in the Colombian patients that were not reported in individuals of European descent. Additionally, the pooled 3D analyses revealed a continuous spectrum of facial dysmorphology, with MS exhibiting the most severe dysmorphologies compared to controls. The ontogenic analyses further demonstrated that craniofacial development is altered in DS and MS, but not in RASopathies such as NS and NF1. Overall, our findings indicate that interpopulation and ontogenic differences in facial phenotypes should be considered to optimize and universalize the use of facial biomarkers. This approach could further help reduce the diagnostic odyssey associated with syndromic and rare conditions. ### Competing Interest Statement The authors have declared no competing interest. ### Funding Statement This study was funded by Proyecto COL0012168-1097 Interfacultades-ICESI, Beca predoctoral Fundacion Alvaro Entrecanales-Lejeune to LME-Q, Joan Oro grant (2024 FI-3 00160) from the Recerca i Universitats Departament (DRU) of the Generalitat de Catalunya and the European Social Fund to AH-L, Wenner Gren Foundation for Anthropological Research (Gr. 10657), Agencia de Gestio Ajuts Universitaris i de Recerca (AGAUR) of the Generalitat de Catalunya (2021 SGR00706 and 2021 SGR0139), and Biological Anthropological Master UB-UAB. ### Author Declarations I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained. Yes The details of the IRB/oversight body that provided approval or exemption for the research described are given below: The study was approved by the Ethics Committee for Human Research at Universidad Icesi with approval record no. 309, and complied with the national guidelines and protocols. 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. Yes I 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). Yes I have followed all appropriate research reporting guidelines, such as any relevant EQUATOR Network research reporting checklist(s) and other pertinent material, if applicable. Yes Raw phenotype data from the Colombian population cannot be made available due to restrictions imposed by the ethics approval.

Down syndrome (DS), caused by trisomy 21, is associated with an increased risk of Alzheimer's disease (AD) and obstructive sleep apnea (OSA). Traditional diagnostic methods for AD and OSA, like cerebrospinal fluid analysis and polysomnography, are invasive and challenging for people with DS. In this study, we assessed whether facial morphology could be used as a potential noninvasive biomarker for these conditions in both DS and the general population. We performed a comprehensive 3D analysis of facial shape variation by registering the 3D coordinates of 21 landmarks on facial models extracted from magnetic resonance images of 131 individuals with DS and 216 euploid (EU) adult controls, including AD and OSA cases. Procrustes ANOVA and MANOVA quantified shape variation by sex, age, and facial size, while geometric morphometrics assessed diagnostic group differences. Significant facial shape differences were observed between the DS and EU groups, indicating sex-dependent differences and altered age-related changes in DS, particularly in females. Facial shape correlated with the amyloid beta ratio (Aβ1-42/Aβ1-40), a key AD biomarker. In DS, facial shape differences by AD diagnosis were not significant after adjusting for age and facial size, but significant shape differences were detected in the EU population. For OSA, facial shape correlated with the apnea-hypopnea index (AHI), and DS individuals with severe OSA showed distinct facial morphology compared with those without OSA, suggesting an association between facial shape and sleep respiratory disturbances. These results highlight the potential of facial morphology as a noninvasive biomarker for AD and OSA detection and management.

Facial dysmorphologies have emerged as potential critical indicators in the diagnosis and prognosis of genetic, psychotic and rare disorders. While in certain conditions these dysmorphologies are severe, in other cases may be subtle and not perceivable to the human eye, requiring precise quantitative tools for their identification. Manual coding of facial dysmorphologies is a burdensome task and is subject to inter- and intra-observer variability. To overcome this gap, we present BioFace3D as a fully automatic tool for the calculation of facial biomarkers using facial models reconstructed from magnetic resonance images. The tool is divided into three automatic modules for the extraction of 3D facial models from magnetic resonance images, the registration of homologous 3D landmarks encoding facial morphology, and the calculation of facial biomarkers from anatomical landmarks coordinates using geometric morphometrics techniques.

As shape alterations in three-dimensional biological structures are associated to numerous pathological processes, quantitative shape analysis for obtaining phenotypic biomarkers of diagnostic potential has become a prominent research area. In this context, the automatic detection of landmarks on 3D anatomical structures is crucial for developing high-throughput phenotyping tools. This study evaluates the performance of multi-view consensus convolutional networks - originally developed for facial landmarking- in automatically detecting landmarks on three different 3D anatomical structures: the face, the upper respiratory airways and the brain hippocampi. Leveraging magnetic resonance imaging datasets, we trained multiple models and assessed their accuracy against manual annotations, while analyzing the impact of different network hyperparameters on the results.

Shape alterations in body organs are common pathological hallmarks of multiple disorders, making quantitative shape analysis key for obtaining diagnostic and prognostic biomarkers. In this context, Geometric Morphometrics (GM) is a powerful approach to capture subtle yet significant dysmorphologies. Since GM relies on registering landmarks on 3D anatomical structures, developing generic, automatic and accurate 3D landmarking methods is key for building high-throughput morphometric tools. This study compares state-of-the-art deep learning and template-based 3D landmarking methods using MRI datasets of faces, upper airways, and hippocampi. We evaluated these methods in terms of landmarking error and morphometric variables relative to manual annotations. Our results show that architecture-reused deep learning methods are more accurate and faster in inference than template-based techniques, particularly for anatomical structures with high shape variability, even with fewer training examples.

Given the shared ectodermal origin and integrated development of the face and the brain, facial biomarkers emerge as potential candidates to assess vulnerability for disorders in which neurodevelopment is compromised, such as schizophrenia (SZ) and bipolar disorder (BD). The sample comprised 188 individuals (67 SZ patients, 46 BD patients and 75 healthy controls (HC)). Using a landmark-based approach on 3D facial reconstructions, we quantified global and local facial shape differences between SZ/BD patients and HC using geometric morphometrics. We also assessed correlations between facial and brain cortical measures. All analyses were performed separately by sex. Diagnosis explained 4.1 % - 5.9 % of global facial shape variance in males and females with SZ, and 4.5 % - 4.1 % in BD. Regarding local facial shape, we detected 43.2 % of significantly different distances in males and 47.4 % in females with SZ as compared to HC, whereas in BD the percentages decreased to 35.8 % and 26.8 %, respectively. We detected that brain area and volume significantly explained 2.2 % and 2 % of facial shape variance in the male SZ - HC sample. Our results support facial shape as a neurodevelopmental marker for SZ and BD and reveal sex-specific pathophysiological mechanisms modulating the interplay between the brain and the face.

Brain imaging techniques, particularly magnetic resonance imaging (MRI), play a crucial role in understanding the neurocognitive phenotype and associated challenges of many neurological disorders, providing detailed insights into the structural alterations in the brain. Despite advancements, the links between cognitive performance and brain anatomy remain unclear. The complexity of analyzing brain MRI scans requires expertise and time, prompting the exploration of artificial intelligence for automated assistance. In this context, unsupervised deep learning techniques, particularly Transformers and Autoencoders, offer a solution by learning the distribution of healthy brain anatomy and detecting alterations in unseen scans. In this work, we evaluate several unsupervised models to reconstruct healthy brain scans and detect synthetic anomalies.

This work is part of the research project “Sons al Balcó” conducted by La Salle - Universitat Ramon Llull, which examines the impacts of noise pollution on human perception and mental health, specifically focusing on the perception of noise in Catalonia during the lockdown in 2020 and the return to normalcy in 2021. The purpose of this research is to identify patterns between the soundscape and the visual landscape of participants’ environments. To achieve this, we have developed a pipeline to automatically analyse the visual landscape of participants’ environments by semantically segmenting the keyframes of their videos using deep neural networks. Specifically, we use the SegFormer model, a Transformer-based framework for semantic segmentation that integrates Transformers with lightweight MLP decoders. This pipeline facilitates the efficient and accurate identification of different objects, to understand the complex relationships among the acoustic environment, visual landscape, and human perception. We expect that our findings will offer insights into the design of urban and suburban areas that promote well-being and quality of life.

Recent studies in neuropsychiatry have highlighted the correlation between facial and brain dysmorphologies. One way of simultaneously analysing the brain and the face of a subject is by reconstructing a whole-head 3D model from structural magnetic resonance imaging (sMRI). However, the use of different reconstruction protocols generates undesired orthogonal rotations of the 3D models. This is a likely situation in multicentric studies that hampers the combination of data from different centers. Although the original sMRI files contain the subject orientation, it is not always possible to access this data. To solve this issue, in this work we propose a novel method to estimate the orientation of 3D heads with rotations of 90 ^∘ or multiples thereof around any of the three Cartesian axes as a required step for generating a normalised dataset in terms of orientation. Our proposal creates 2D images from orthogonal projections of the 3D object, transforming orientation estimation into an image classification problem. Experimental results show that our method, using three orthographic views of the 3D head to create the projection image and ResNet50 for classification, achieves an accuracy of 99.7

The deployment of an expert system running over a wireless acoustic sensors network made up of bioacoustic monitoring devices that recognize bird species from their sounds would enable the automation of many tasks of ecological value, including the analysis of bird population composition or the detection of endangered species in areas of environmental interest. Endowing these devices with accurate audio classification capabilities is possible thanks to the latest advances in artificial intelligence, among which deep learning techniques stand out. To train such algorithms, data from the sources to be classified is required. For this reason, this paper presents the Western Mediterranean Wetland Birds (WMWB) dataset, consisting of 201.6 min and 5795 annotated audio excerpts of 20 endemic bird species of the Aiguamolls de l'Empord`a Natural Park. The main objective of this work is to describe and analyze this new dataset. Moreover, this work presents the results of bird species clas-sification experiments using four well-known deep neural networks fine-tuned on our dataset, whose models are also made public along with the dataset. These results are aimed to serve as a performance baseline reference for the community when using the WMWB dataset for their experiments.