Quality control of elbow joint radiography using a YOLOv8-based artificial intelligence technology

Abstract

Background: To explore an artificial intelligence (AI) technology employing YOLOv8 for quality control (QC) on elbow joint radiographs.

Methods: From January 2022 to August 2023, 2643 consecutive elbow radiographs were collected and randomly assigned to the training, validation, and test sets in a 6:2:2 ratio. We proposed the anteroposterior (AP) and lateral (LAT) models to identify target detection boxes and key points on elbow radiographs using YOLOv8. These identifications were transformed into five quality standards: (1) AP elbow positioning coordinates (X_A and Y_A); (2) olecranon fossa positioning distance parameters (S₁₇ and S₂₇); (3) key points of joint space (Y₃, Y₄, Y₅ and Y₆); (4) LAT elbow positioning coordinates (X₂ and Y₂); and (5) flexion angle. Models were trained and validated using 2,120 radiographs. A test set of 523 radiographs was used for assessing the agreement between AI and physician and to evaluate clinical efficiency of models.

Results: The AP and LAT models demonstrated high precision, recall, and mean average precision for identifying boxes and points. AI and physicians showed high intraclass correlation coefficient (ICC) in evaluating: AP coordinates X_A (0.987) and Y_A (0.991); olecranon fossa parameters S₁₇ (0.964) and S₂₇ (0.951); key points Y₃ (0.998), Y₄ (0.997), Y₅ (0.998) and Y₆ (0.959); LAT coordinates X₂ (0.994) and Y₂ (0.986); and flexion angle (0.865). Compared to manual methods, using AI, QC time was reduced by 43% for AP images and 45% for LAT images (p < 0.001).

Conclusion: YOLOv8-based AI technology is feasible for QC of elbow radiography with high performance.

Relevance statement: This study proposed and validated a YOLOv8-based AI model for automated quality control in elbow radiography, obtaining high efficiency in clinical settings.

Key points: QC of elbow joint radiography is important for detecting diseases. Models based on YOLOv8 are proposed and perform well in image QC. Models offer objective and efficient solutions for QC in elbow joint radiographs.

Keywords: Artificial intelligence; Deep learning; Elbow joint; Quality control; Radiography.

Fig. 1

Flowchart of the model training, validation and test on the dataset. AP, Anteroposterior; LAT, Lateral

Fig. 2

Annotations of key points and auxiliary lines. a Anterioposterior view of the elbow joint. b Lateral view of the elbow joint. Auxiliary lines S₁₇ and S₂₇ in the anterioposterior view represent distances between points 1 and 7, and between points 2 and 7, respectively, along with the flexion angle α in the lateral view are shown

Fig. 3

The architecture of the YOLOv8 algorithm, which is divided into three parts, including backbone, neck, and head

Fig. 4

Visualization of the artificial intelligence (AI) models’ predictions and clinician’s annotation. The red key points and box are from the clinician’s annotations, and the blue key points and boxes are generated by the AI model. a Anterioposterior view, showing one target detection box and seven key points; b lateral view, showing one target detection box and three key points

Fig. 5

Scatter plots of correlations between AI predictions and clinician annotations in the test set. a ICC for X_A and Y_A in AP view; b ICC for S₁₇ and S₂₇ in AP view; c ICC for Y₃ and Y₄ in AP view; d ICC for Y₅ and Y₆ in AP view; e ICC for X₂ and Y₂ in LAT view; f ICC for flexion angle α in LAT view. AP, Anteroposterior; ICC, Intraclass correlation coefficient; LAT Lateral

Fig. 6

YOLOv8-QC GUI software and its application. a YOLOv8-QC GUI software. b QC results available to users. c Application of YOLOv8-QC GUI software. AP, Anteroposterior; LAT, Lateral; QC, Quality control; GUI, Graphical user interface