Artificial intelligence diagnostic model for multi-site fracture X-ray images of extremities based on deep convolutional neural networks

Abstract

Background: The rapid and accurate diagnosis of fractures is crucial for timely treatment of trauma patients. Deep learning, one of the most widely used forms of artificial intelligence (AI), is now commonly employed in medical imaging for fracture detection. This study aimed to construct a deep learning model using big data to recognize multiple-fracture X-ray images of extremity bones.

Methods: Radiographic imaging data of extremities were retrospectively collected from five hospitals between January 2017 and September 2020. The total number of people finally included was 25,635 and the total number of images included was 26,098. After labeling the lesions, the randomized method used 90% of the data as the training set to develop the fracture detection model, and the remaining 10% was used as the validation set to verify the model. The faster region convolutional neural networks (R-CNN) algorithm was adopted to construct diagnostic models for detection. The Dice coefficient was used to evaluate the image segmentation accuracy. The performances of detection models were evaluated with sensitivity, specificity, and area under the receiver operating characteristic curve (AUC).

Results: The free-response receiver operating characteristic (FROC) curve value was 0.886 and 0.843 for the detection of single and multiple fractures, respectively. Additionally, the effective identification AUC for all parts was higher than 0.920. Notably, the AUC for wrist fractures reached 0.952. The average accuracy in detecting bone fracture regions in the extremities was 0.865. When analyzing single and multiple lesions at the patient level, the sensitivity was 0.957 for patients with multiple lesions and 0.852 for those with single lesions. In the segmentation task, the training set (the data set used by the machine learning model to train and learn) and the validation set (the data set used to evaluate the performance of the model) reached 0.996 and 0.975, respectively.

Conclusions: The faster R-CNN training algorithm exhibits excellent performance in simultaneously identifying fractures in the hands, feet, wrists, ankles, radius and ulna, and tibia and fibula on X-ray images. It demonstrates high accuracy, low false-negative rates, and controllable false-positive rates. It can serve as a valuable screening tool.

Keywords: Deep learning; X-ray; artificial intelligence (AI); multi-site fracture of extremities.

Figure 1

The original image were screened and grouped. DICOM, Digital Imaging and Communications in Medicine.

Figure 2

Example image enhancement with gray-scale range stretching. x-axis represents the pixel value of the original image, y-axis represents the mapped image pixel range after image enhancement, where the black solid line represents no image enhancement, and the red solid line represents the result after image enhancement.

Figure 3

Example diagram of flip transformation.

Figure 4

Grayscale transformation example diagram (left 1: grayscale 0.8, left 2: grayscale 0.9, left 3: original image, left 4: grayscale 1.1, left 5: grayscale 1.2).

Figure 5

Model architecture of Unet partition network.

Figure 6

Network structure of fracture detection model. The red box indicates the detected fracture lesion area. FPN, feature pyramid network; RPN, region proposal network; ROI, region of interest; Reg is reg-layer (it predicts the coordinates of the proposal corresponding to the central anchor of the proposal); Cls is cls-layer (it determines whether a proposal is in the foreground or background).

Figure 7

Schematic diagram of the feature pyramid structure. Conv, convolution.

Figure 8

FROC curve and ROC curve. FROC, free-response receiver operating characteristic; AUC, area under the receiver operating characteristic curve; TPR, true positive rate; FPR, false positive rate; ROC, receiver operating characteristic.