Real-time brain tumor detection in intraoperative ultrasound: From model training to deployment in the operating room.
Authors
Affiliations (11)
Affiliations (11)
- Department of Neurosurgery, Rio Hortega University Hospital, Dulzaina 2, Valladolid, 47014, Valladolid, Spain; Specialized Group in Biomedical Imaging and Computational Analysis (GEIBAC), Instituto de Investigacion Biosanitaria de Valladolid (IBioVALL), Dulzaina 2, Valladolid, 47014, Valladolid, Spain. Electronic address: [email protected].
- Department of Neurosurgery, Rio Hortega University Hospital, Dulzaina 2, Valladolid, 47014, Valladolid, Spain; Specialized Group in Biomedical Imaging and Computational Analysis (GEIBAC), Instituto de Investigacion Biosanitaria de Valladolid (IBioVALL), Dulzaina 2, Valladolid, 47014, Valladolid, Spain.
- Biomedical Engineering Group, University of Valladolid, P. de Belen 15, Valladolid, 47011, Valladolid, Spain.
- Department of Neurosurgery, Tata Memorial Centre, Homi Bhabha National Institute, Parel East, Mumbai, 400012, Maharashtra, India.
- UMR 1253, iBrain, Universit'e de Tours, Inserm, 10 Bd Tonnelle, Tours, 37000, France; Department of Neurosurgery, CHRU de Tours, 2 Bd Tonnelle, Tours, 37000, France.
- Department of Neurosurgery, ARNAS Civico Di Cristina Benfratelli Hospital, P.Za Leotta Nicola, Palermo, 90127, Italy.
- Department of Neurosurgery, Fondazione IRCCS Istituto Neurologico Carlo Besta, Via Celoria 11, Milan, 20133, Italy; Department of Pharmacological and Biomolecular Sciences, University of Milan, Via Festa del Perdono 7, Milan, 20122, Italy.
- Department of Neurosurgery, Fondazione IRCCS Istituto Neurologico Carlo Besta, Via Celoria 11, Milan, 20133, Italy.
- Department of Neurosurgery, Fondazione IRCCS Istituto Neurologico Carlo Besta, Via Celoria 11, Milan, 20133, Italy; Department of Oncology and Hematology-Oncology, Universit'a Degli Studi di Milano, Via Festa del Perdono 7, Milan, 20122, Italy; Department of Neurological Surgery, Johns Hopkins Medical School, 733 N Broadway, Baltimore, 21205, Maryland, USA.
- Department of Neurosurgery, Massachusetts General Hospital, Harvard Medical School, 55 Fruit St, Boston, 02114, Massachusetts, USA.
- Specialized Group in Biomedical Imaging and Computational Analysis (GEIBAC), Instituto de Investigacion Biosanitaria de Valladolid (IBioVALL), Dulzaina 2, Valladolid, 47014, Valladolid, Spain; Center for Biomedical Research in Network of Bioengineering, Biomaterials and Nanomedicine (CIBER-BBN), Av. Monforte de Lemos, 3-5, Valladolid, 47011, Spain; Institute for Research in Mathematics (IMUVA), University of Valladolid, P. de Belen 15, Valladolid, 47011, Spain.
Abstract
Intraoperative ultrasound (ioUS) is a valuable tool in brain tumor surgery due to its versatility, affordability, and seamless integration into the surgical workflow. However, its adoption remains limited, primarily because of the challenges associated with image interpretation and the steep learning curve required for effective use. This study aimed to enhance the interpretability of ioUS images by developing a real-time brain tumor detection system deployable in the operating room. We collected 2D ioUS images from the BraTioUS and ReMIND datasets, annotated with expert-refined tumor labels. Using the YOLO11 architecture and its variants, we trained object detection models to identify brain tumors. The dataset included 1732 images from 192 patients, divided into training, validation, and test sets. Data augmentation expanded the training set to 11,570 images. In the test dataset, YOLO11s achieved the best balance of precision and computational efficiency, with a mAP@50 of 0.95, mAP@50-95 of 0.65, and a processing speed of 34.16 frames per second. The proposed solution was prospectively validated in a cohort of 20 consecutively operated patients diagnosed with brain tumors. Neurosurgeons confirmed its seamless integration into the surgical workflow, with real-time predictions accurately delineating tumor regions. These findings highlight the potential of real-time object detection algorithms to enhance ioUS-guided brain tumor surgery, addressing key challenges in interpretation and providing a foundation for future development of computer vision-based tools for neuro-oncological surgery.