ABSTRACT
INTRODUCTION: In the UK, 1 in 50 children sustain a fractured bone yearly, yet studies have shown that 34% of children sustaining an injury do not have a visible fracture on initial radiographs. Wrist fractures are particularly difficult to identify because the growth plate poses diagnostic challenges when interpreting radiographs. METHODS: We developed Convolutional Neural Network (CNN) image recognition software to detect fractures in radiographs from children. A consecutive data set of 5,000 radiographs of the distal radius in children aged less than 19 years from 2014 to 2019 was used to train the CNN. In addition, transfer learning from a VGG16 CNN pretrained on non-radiological images was applied to improve generalisation of the network and the classification of radiographs. Hyperparameter tuning techniques were used to compare the model with the radiology reports that accompanied the original images to determine diagnostic test accuracy. RESULTS: The training set consisted of 2,881 radiographs with a fracture and 1,571 without; 548 radiographs were outliers. With additional augmentation, the final data set consisted of 15,498 images. The data set was randomly split into three subsets: training (70%), validation (10%) and test (20%). After training for 20 epochs, the diagnostic test accuracy was 85%. DISCUSSION: A CNN model is feasible in diagnosing paediatric wrist fractures. We demonstrated that this application could be utilised as a tool for improving diagnostic accuracy. Future work would involve developing automated treatment pathways for diagnosis, reducing unnecessary hospital visits and allowing staff redeployment to other areas.
Subject(s)
Fractures, Bone , Hand Injuries , Wrist Fractures , Wrist Injuries , Humans , Child , Artificial Intelligence , Proof of Concept Study , Neural Networks, Computer , Wrist Injuries/diagnostic imagingABSTRACT
Models that predict soft-tissue indentation forces have many important applications including estimation of interaction forces, palpation simulation, disease diagnosis, and robotic assistance. In many medical applications such as rehabilitation, clinical palpation, and manipulation of organs, characterizing soft-tissue properties mainly depends on the accurate estimation of indentation forces. A new indentation model for estimating circular indenter 'force-displacement' characteristics is presented in this paper. The proposed model is motivated by a 'force-displacement' soil-tool model and is computationally efficient. The main feature of the proposed model is that it can be used to predict the force variations for a variety of tools without the need for retuning the model parameters for each tool. A six-degree-of-freedom robot manipulator with force and position sensors is used to validate the indentation model. Measured force versus tool displacement data for lamb liver and kidney, for a variety of tool diameters, are presented and compared with the forces predicted by the model, showing good agreement (RMS error < 8 per cent).
