Fetal Plane Classification (AI Research)

Research project focused on developing deep learning models for automated classification of fetal ultrasound planes, improving diagnostic accuracy and reducing examination time.

Accurate identification of fetal planes in ultrasound images requires significant expertise and is subject to inter-observer variability. Automated classification could improve consistency and assist less experienced practitioners.

Developed convolutional neural network (CNN) architectures specifically designed for fine-grained classification of fetal ultrasound planes. The models achieve high accuracy in identifying standard planes used in prenatal screening, with explainability features to help clinicians understand the model's decisions.