Summary

This study introduces a novel unsupervised anomaly detection method for segmenting needle-like structures in real-time Magnetic Resonance Imaging (MRI). The approach effectively identifies needles and tumors in noisy images, improving brain biopsy procedures.