Litcius/Paper detail

FetSAM: Advanced Segmentation Techniques for Fetal Head Biometrics in Ultrasound Imagery

Mahmood Alzubaidi, Uzair Shah, Marco Agus, Mowafa Househ

2024IEEE Open Journal of Engineering in Medicine and Biology13 citationsDOIOpen Access PDF

Abstract

<italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Goal:</i> FetSAM represents a cutting-edge deep learning model aimed at revolutionizing fetal head ultrasound segmentation, thereby elevating prenatal diagnostic precision. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Methods:</i> Utilizing a comprehensive dataset-the largest to date for fetal head metrics-FetSAM incorporates prompt-based learning. It distinguishes itself with a dual loss mechanism, combining Weighted DiceLoss and Weighted Lovasz Loss, optimized through AdamW and underscored by class weight adjustments for better segmentation balance. Performance benchmarks against prominent models such as U-net, DeepLabV3, and Segformer highlight its efficacy. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Results:</i> FetSAM delivers unparalleled segmentation accuracy, demonstrated by a DSC of 0.90117, HD of 1.86484, and ASD of 0.46645. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Conclusion:</i> FetSAM sets a new benchmark in AI-enhanced prenatal ultrasound analysis, providing a robust, precise tool for clinical applications and pushing the envelope of prenatal care with its groundbreaking dataset and segmentation capabilities.

Topics & Concepts

BiometricsFetal headSegmentationArtificial intelligenceHead (geology)Computer scienceComputer visionUltrasoundFetusMedicineRadiologyGeologyPregnancyBiologyGeomorphologyGeneticsFetal and Pediatric Neurological DisordersCleft Lip and Palate ResearchAdvanced Neural Network Applications