arXiv:2508.13544v5 Announce Type: replace 
Abstract: Implicit Neural Representations (INRs) leverage neural networks to map coordinates to corresponding signals, enabling continuous and compact representations. This paradigm has driven significant advances in various vision tasks. However, existing INRs lack frequency selectivity and spatial localization, leading to an over-reliance on redundant signal components. Consequently, they exhibit spectral bias, tending to learn low-frequency components early while struggling to capture fine high-frequency details. To address these issues, we propose FLAIR (Frequency- and Locality-Aware Implicit Neural Representations), which incorporates two key innovations. The first is Band-Localized Activation (BLA), a novel activation designed for joint frequency selection and spatial localization under the constraints of the time-frequency uncertainty principle (TFUP). Through structured frequency control and spatially localized responses, BLA effectively mitigates spectral bias and enhances training stability. The second is Wavelet-Energy-Guided Encoding (WEGE), which leverages the discrete wavelet transform to compute energy scores and explicitly guide frequency information to the network, enabling precise frequency selection and adaptive band control. Our method consistently outperforms existing INRs in 2D image representation, as well as 3D shape reconstruction and novel view synthesis.

FLAIR، أو التمثيلات العصبية الضمنية الحساسة للتردد والموقع، يقدم ابتكارات لتحسين التمثيلات العصبية الضمنية (INRs) من خلال معالجة قيودها في الانتقائية الترددية والتحديد المكاني. تستخدم هذه الطريقة الجديدة تنشيطًا محليًا حسب النطاق (BLA) لتحسين ربط الإحداثيات بالإشارات، مما يسمح بتمثيل أفضل للتفاصيل عالية التردد في مهام الرؤية المختلفة.

FLAIR, o Representaciones Neuronales Implícitas Sensibles a la Frecuencia y Localidad, introduce innovaciones para mejorar las Representaciones Neuronales Implícitas (INRs) al abordar sus limitaciones en selectividad de frecuencia y localización espacial. Este nuevo enfoque utiliza la Activación Localizada por Banda (BLA) para mejorar la asignación de coordenadas a señales, permitiendo una mejor representación de detalles de alta frecuencia en diversas tareas de visión.

FLAIR, ou Représentations Neuronales Implicites Sensibles à la Fréquence et à la Localité, introduit des innovations pour améliorer les Représentations Neuronales Implicites (INRs) en abordant leurs limitations en matière de sélectivité de fréquence et de localisation spatiale. Cette nouvelle approche utilise l'Activation Localisée par Bande (BLA) pour améliorer la cartographie des coordonnées aux signaux, permettant une meilleure représentation des détails haute fréquence dans diverses tâches de vision.

FLAIR, or Frequency- and Locality-Aware Implicit Neural Representations, introduces innovations to enhance Implicit Neural Representations (INRs) by addressing their limitations in frequency selectivity and spatial localization. This new approach utilizes Band-Localized Activation (BLA) to improve the mapping of coordinates to signals, enabling better representation of high-frequency details in various vision tasks.

FLAIR: Frequency- and Locality-Aware Implicit Neural Representations

arXiv:2601.08701v1 Announce Type: cross 
Abstract: Accurate and generalisable segmentation of stroke lesions from magnetic resonance imaging (MRI) is essential for advancing clinical research, prognostic modelling, and personalised interventions. Although deep learning has improved automated lesion delineation, many existing models are optimised for narrow imaging contexts and generalise poorly to independent datasets, modalities, and stroke stages. Here, we systematically evaluated stroke lesion segmentation using the nnU-Net framework across multiple heterogeneous, publicly available MRI datasets spanning acute and chronic stroke. Models were trained and tested on diffusion-weighted imaging (DWI), fluid-attenuated inversion recovery (FLAIR), and T1-weighted MRI, and evaluated on independent datasets. Across stroke stages, models showed robust generalisation, with segmentation accuracy approaching reported inter-rater reliability. Performance varied with imaging modality and training data characteristics. In acute stroke, DWI-trained models consistently outperformed FLAIR-based models, with only modest gains from multimodal combinations. In chronic stroke, increasing training set size improved performance, with diminishing returns beyond several hundred cases. Lesion volume was a key determinant of accuracy: smaller lesions were harder to segment, and models trained on restricted volume ranges generalised poorly. MRI image quality further constrained generalisability: models trained on lower-quality scans transferred poorly, whereas those trained on higher-quality data generalised well to noisier images. Discrepancies between predictions and reference masks were often attributable to limitations in manual annotations. Together, these findings show that automated lesion segmentation can approach human-level performance while identifying key factors governing generalisability and informing the development of lesion segmentation tools.

تم إجراء تقييم خارجي شامل لإطار nnU-Net لعملية تقسيم الآفات تلقائيًا في تصوير الرنين المغناطيسي للسكتة الدماغية، مما أظهر تعميمًا قويًا عبر مجموعات بيانات متعددة وطرق تصوير، بما في ذلك DWI وFLAIR وMRI بتقنية T1. تسلط هذه الدراسة الضوء على أهمية تحديد الآفات بدقة لأغراض البحث السريري والتدخلات الشخصية في إدارة السكتات الدماغية.

Se ha llevado a cabo una validación externa integral del marco nnU-Net para la segmentación automatizada de lesiones en la IRM de accidentes cerebrovasculares, demostrando una robusta generalización a través de múltiples conjuntos de datos y modalidades de imagen, incluyendo DWI, FLAIR y IRM ponderada en T1. Este estudio resalta la importancia de la delimitación precisa de lesiones para la investigación clínica y las intervenciones personalizadas en la gestión de accidentes cerebrovasculares.

Une validation externe complète du cadre nnU-Net pour la segmentation automatisée des lésions dans l'IRM des AVC a été réalisée, démontrant une généralisation robuste à travers divers ensembles de données et modalités d'imagerie, y compris DWI, FLAIR et IRM T1. Cette étude souligne l'importance d'une délimitation précise des lésions pour la recherche clinique et les interventions personnalisées dans la gestion des AVC.

A comprehensive external validation of the nnU-Net framework for automated lesion segmentation in stroke MRI has been conducted, demonstrating robust generalization across various datasets and imaging modalities, including DWI, FLAIR, and T1-weighted MRI. This study highlights the importance of accurate lesion delineation for clinical research and personalized interventions in stroke management.

FLAIR: Frequency- and Locality-Aware Implicit Neural Representations

Was this article worth reading? Share it

LucidQuery AI

AINF Membership

Attentive AI

The Visualizer

FLUX AI ART

VibeFrame

Ready to build your own newsroom?