Deep Learning-Based Automated Detection of the Middle Cerebral Artery in Transcranial Doppler Ultrasound Examinations

• Published in: Ultrasound in medicine & biology Vol. 51; no. 9; pp. 1555 - 1563
• Main Authors: Lee, HyeonWoo, Shi, William, Mukaddim, Rashid Al, Brunelle, Elizabeth, Palisetti, Abhinav, Imaduddin, Syed M., Rajendram, Phavalan, Incontri, Diego, Lioutas, Vasileios-Arsenios, Heldt, Thomas, Raju, Balasundar I.
• Format: Journal Article
• Language: English
• Published: England Elsevier Inc 01.09.2025
• Subjects: Adult; Deep Learning; Female; Humans; Male; Middle Aged; Middle Cerebral Artery; Middle Cerebral Artery - diagnostic imaging; Object Detection; Radiology; Real-Time Detection; RT-DETR; Stroke - diagnostic imaging; Transcranial Doppler; Ultrasonography, Doppler, Transcranial - methods; Ultrasound Imaging; YOLOv10; Object Detection; YOLOv10; Real-Time Detection; RT-DETR; Ultrasound Imaging; Transcranial Doppler; Middle Cerebral Artery; Deep Learning
• ISSN: 0301-5629; 1879-291X; 1879-291X
• DOI: 10.1016/j.ultrasmedbio.2025.05.028

Transcranial Doppler (TCD) ultrasound has significant clinical value for assessing cerebral hemodynamics, but its reliance on operator expertise limits broader clinical adoption. In this work, we present a lightweight real-time deep learning-based approach capable of automatically identifying the middle cerebral artery (MCA) in TCD Color Doppler images. Two state-of-the-art object detection models, YOLOv10 and Real-Time Detection Transformers (RT-DETR), were investigated for automated MCA detection in real-time. TCD Color Doppler data (41 subjects; 365 videos; 61,611 frames) were collected from neurologically healthy individuals (n = 31) and stroke patients (n = 10). MCA bounding box annotations were performed by clinical experts on all frames. Model training consisted of pretraining utilizing a large abdominal ultrasound dataset followed by subsequent fine-tuning on acquired TCD data. Detection performance at the instance and frame levels, and inference speed were assessed through four-fold cross-validation. Inter-rater agreement between model and two human expert readers was assessed using distance between bounding boxes and inter-rater variability was quantified using the individual equivalence coefficient (IEC) metric. Both YOLOv10 and RT-DETR models showed comparable frame level accuracy for MCA presence, with F1 scores of 0.884 ± 0.023 and 0.884 ± 0.019 respectively. YOLOv10 outperformed RT-DETR for instance-level localization accuracy (AP: 0.817 vs. 0.780) and had considerably faster inference speed on a desktop CPU (11.6 ms vs. 91.14 ms). Furthermore, YOLOv10 showed an average inference time of 36 ms per frame on a tablet device. The IEC was −1.08 with 95 % confidence interval: [−1.45, −0.19], showing that the AI predictions deviated less from each reader than the readers’ annotations deviated from each other. Real-time automated detection of the MCA is feasible and can be implemented on mobile platforms, potentially enabling wider clinical adoption by less-trained operators in point-of-care settings.