AI segments masseter muscle on ultrasonography

Title: A deep learning approach for masseter muscle segmentation on ultrasonography

Authors: Gaye Keser, Ibrahim Sevki Bayrakdar, Filiz Namdar Pekiner, Özer Çelik, Kaan Orhan

Summary

This study aimed to develop and evaluate an artificial intelligence model using the PyTorch U-Net architecture for the automatic segmentation of the masseter muscle on ultrasonography images. The study utilized a retrospective dataset of 388 images, divided into training, verification, and test sets, and the annotations were verified by Oral and Maxillofacial Radiology experts. The model achieved perfect performance metrics (F1, sensitivity, and precision of 1.0), suggesting that this deep learning strategy can assist medical practitioners by reducing diagnostic time.

Key Words: Ultrasonography, Masseter muscle, Deep learning, Artificial intelligence, Segmentation

Extracted Data

Year: 2022

Modality: Ultrasonography (US)

Dataset: 388 ultrasonography images of adult masseter muscles

Dataset Split: Training 312; Validation 38; Test 38

Network Architecture: U-Net (PyTorch implementation)

Metrics: Sensitivity (Recall), Precision, F1-score

AP - Professional Qty: Unavailable (Source states "experts" in the plural form but specifies no exact quantity)

AP - Supervisor Presence: No

AP - Experience Level: Not specified

AP - Expertise Area: Oral and Maxillofacial Radiology

AP - Tool or System: CranioCatch labeling program

ML Task: Semantic Segmentation

Project Objective: Automatic segmentation of the masseter muscle on ultrasonography images

Clinical Relevance: Enables fast and reproducible assessment of masseter muscle morphology, supporting musculoskeletal diagnosis without ionizing radiation. It also enables the quantification of muscle size/thickness which is relevant to masticatory function, craniofacial functional processes, and the diagnosis of changes in face and neck muscle proportions.