Automatic diagnosis of early pregnancy fetal nasal bone development based on complex mid-sagittal section ultrasound imaging

Neurocomputing, Journal Year: 2025, Volume and Issue: unknown, P. 129773 - 129773

Published: Feb. 1, 2025

Language: Английский

---

Automated Left Ventricle Segmentation in Echocardiography Using YOLO: A Deep Learning Approach for Enhanced Cardiac Function Assessment

Electronics, Journal Year: 2024, Volume and Issue: 13(13), P. 2587 - 2587

Published: July 1, 2024

Accurate segmentation of the left ventricle (LV) using echocardiogram (Echo) images is essential for cardiovascular analysis. Conventional techniques are labor-intensive and exhibit inter-observer variability. Deep learning has emerged as a powerful tool automated medical image segmentation, offering advantages in speed potentially superior accuracy. This study explores efficacy employing YOLO (You Only Look Once) model LV Echo images. YOLO, cutting-edge object detection model, achieves exceptional speed–accuracy balance through its well-designed architecture. It utilizes efficient dilated convolutional layers bottleneck blocks feature extraction while incorporating innovations like path aggregation spatial attention mechanisms. These attributes make compelling candidate adaptation to We posit that by fine-tuning pre-trained YOLO-based on well-annotated dataset, we can leverage model’s strengths real-time processing precise localization achieve robust segmentation. The proposed approach entails rigorously labeled dataset. Model performance been evaluated established metrics such mean Average Precision (mAP) at an Intersection over Union (IoU) threshold 50% (mAP50) with 98.31% across range IoU thresholds from 95% (mAP50:95) 75.27%. Successful implementation potential significantly expedite standardize advancement could translate improved clinical decision-making enhanced patient care.

Language: Английский

---

Automated Left Ventricle Segmentation in Echocardiography using YOLO: A Deep Learning Approach for Enhanced Cardiac Function Assessment

Published: May 21, 2024

Accurate segmentation of the Left Ventricle (LV) in Echocardiogram (Echo) images is essential for cardiovascular analysis. Conventional techniques are labor-intensive and exhibit inter-observer variability. Deep learning has emerged as a powerful tool automated medical image segmentation, offering advantages speed potentially superior accuracy. This study explores efficacy employing YOLO (You Only Look Once) model LV Echo images. YOLO, cutting-edge object detection model, achieves exceptional speed-accuracy balance through its well-designed architecture. It utilizes efficient dilated convolutional layers bottleneck blocks feature extraction, while incorporating innovations like path aggregation spatial attention mechanisms. These attributes make compelling candidate adaptation to We posit that by fine-tuning pre-trained based on well-annotated dataset, we can leverage model's strengths real-time processing precise localization achieve robust segmentation. The proposed approach entails rigorously labeled dataset. Model performance been evaluated using established metrics such mean Average Precision (mAP) at an Intersection over Union (IoU) threshold 50% (mAP50) with 98.31%and across range IoU thresholds from 95% (mAP50:95) 75.27%. Successful implementation potential significantly expedite standardize advancement could translate improved clinical decision-making enhanced patient care.

Language: Английский

---

A rule-guided interpretable lightweight framework for fetal standard ultrasound plane capture and biometric measurement

Neurocomputing, Journal Year: 2024, Volume and Issue: unknown, P. 129290 - 129290

Published: Dec. 1, 2024

Language: Английский

---

Optimized Pyramidal Convolution Shuffle attention neural network based fetal cardiac cycle detection from echocardiograms

Biomedical Signal Processing and Control, Journal Year: 2024, Volume and Issue: 98, P. 106701 - 106701

Published: Aug. 30, 2024

Language: Английский

---

Fetal Cardiac Structure Detection Using Multi-task Learning

Lecture notes in computer science, Journal Year: 2024, Volume and Issue: unknown, P. 405 - 419

Published: Jan. 1, 2024

Language: Английский

---

Automatic Diagnosis of Early Pregnancy Fetal Nasal Bone Development Based on Complex Mid-Sagittal Section Ultrasound Imaging

Published: Jan. 1, 2024

Language: Английский

---

Anatomical structures detection using topological constraint knowledge in fetal ultrasound

Neurocomputing, Journal Year: 2024, Volume and Issue: unknown, P. 129143 - 129143

Published: Dec. 1, 2024

Language: Английский

---