Automatic Hip Fracture Identification and Functional Subclassification with Deep Learning

Abstract

Purpose: Hip fractures are a common cause of morbidity and mortality. Automatic identification and classification of hip fractures using deep learning may improve outcomes by reducing diagnostic errors and decreasing time to operation. Methods: Hip and pelvic radiographs from 1118 studies were reviewed and 3034 hips were labeled via bounding boxes and classified as normal, displaced femoral neck fracture, nondisplaced femoral neck fracture, intertrochanteric fracture, previous ORIF, or previous arthroplasty. A deep learning-based object detection model was trained to automate the placement of the bounding boxes. A Densely Connected Convolutional Neural Network (DenseNet) was trained on a subset of the bounding box images, and its performance evaluated on a held out test set and by comparison on a 100-image subset to two groups of human observers: fellowshiptrained radiologists and orthopaedists, and senior residents in emergency medicine, radiology, and orthopaedics. Results: The binary accuracy for fracture of our model was 93.8% (95% CI, 91.3-95.8%), with sensitivity of 92.7% (95% CI, 88.7-95.6%), and specificity 95.0% (95% CI, 91.5-97.3%). Multiclass classification accuracy was 90.4% (95% CI, 87.4-92.9%). When compared to human observers, our model achieved at least expert-level classification under all conditions. Additionally, when the model was used as an aid, human performance improved, with aided resident performance approximating unaided fellowship-trained expert performance. Conclusions: Our deep learning model identified and classified hip fractures with at least expert-level accuracy, and when used as an aid improved human performance, with aided resident performance approximating that of unaided fellowship-trained attendings.

Detecting total hip replacement prosthesis design on plain radiographs using deep convolutional neural network

“…In this study, we implemented the DL method to automatically detect the design of THR implants from plain film radiographs. Other studies have applied DL methods on plain film radiographs for various orthopedic applications 9–17 . However, to the best of our knowledge, this is the first study to apply the DL method to automatically detect THR implant designs.…”

Comparing performance of deep convolutional neural network with orthopaedic surgeons on identification of total hip prosthesis design from plain radiographs

“…In this study, we implemented DL method to automatically detect the design of THR implants from plain film AP radiographs. Other studies have applied DL methods on plain film radiographs for various orthopedic applications [7][8][9][10][11][12][13][14][15][16] . However, to the best of our knowledge, our pilot study identifying three types of THR femoral implant designs 18 , and this study identifying nine types of THR femoral implant designs are the first applications of DL method to automatically detect THR femoral implant designs.…”

“…We hypothesized that deep learning (DL) based artificial intelligence algorithms could be trained to automatically identify hip implant designs from radiographic images. In recent years, DL methods have been applied to the interpretation of plain film radiographs with high degrees of success for identification and classification of orthopaedic fractures, staging knee osteoarthritis (OA) severity, and detection of aseptic loosening, to name a few [7][8][9][10][11][12][13][14][15][16][17] . In a previous pilot study, we successfully trained a DL method for the first time to classify a given THR radiograph into one of three possible femoral component designs 18 .…”

Assessment of a deep-learning system for fracture detection in musculoskeletal radiographs

“…We developed a deep-learning system for detecting fractures across the musculoskeletal system, trained it on data manually annotated by senior orthopedic surgeons and radiologists, and then evaluated the system’s ability to emulate them. Prior deep-learning systems for fracture detection are limited in scope to single bones, areas within a bone, specific anatomical regions (e.g., refs 9 – 11 ), or limited clinical settings (e.g., orthopedic settings—hand, wrist, ankle 12 ). Deep-learning methods have recently shown great promise at successfully addressing a wide variety of medical visual search tasks 13 – 16 , but have yet to tackle a common and heterogeneous clinical problem in medical imaging.…”

“…Although deep-learning methods have shown promise at addressing a variety of medical visual search tasks 13 – 15 , they have yet to tackle a heterogeneous clinical problem in medical imaging such as identifying fractures across 16 anatomical regions. Prior deep-learning studies for fracture detection typically have been limited to a single bone or anatomical region (e.g., wrist or hip 9 – 11 , 13 ) and the most similar study in scope is a deep-learning model that reports an overall AUC of 0.929 detecting abnormalities in only upper extremity musculoskeletal radiographs 18 . Thus, the present study has much broader clinical breadth than prior deep-learning systems for musculoskeletal radiographs.…”