Literature DB >> 32623269

Deep learning evaluation of pelvic radiographs for position, hardware presence, and fracture detection.

Gene Kitamura1.   

Abstract

PURPOSE: Recent papers have shown the utility of deep learning in detecting hip fractures with pelvic radiographs, but there is a paucity of research utilizing deep learning to detect pelvic and acetabular fractures. Creating deep learning models also requires appropriately labeling x-ray positions and hardware presence. Our purpose is to train and test deep learning models to detect pelvic radiograph position, hardware presence, and pelvic and acetabular fractures in addition to hip fractures.
MATERIAL AND METHODS: Data was retrospectively acquired between 8/2009-6/2019. A subset of the data was split into 4 position labels and 2 hardware labels to create position labeling and hardware detecting models. The remaining data was parsed with these trained models, labeled based on 6 "separate" fracture patterns, and various fracture detecting models were created. A receiver operator characteristic (ROC) curve, area under the curve (AUC), and other output metrics were evaluated.
RESULTS: The position and hardware models performed well with AUC of 0.99-1.00. The AUC for proximal femoral fracture detection was as high as 0.95, which was in line with previously published research. Pelvic and acetabular fracture detection performance was as low as 0.70 for the posterior pelvis category and as high as 0.85 for the acetabular category with the "separate" fracture model.
CONCLUSION: We successfully created deep learning models that can detect pelvic imaging position, hardware presence, and pelvic and acetabular fractures with AUC loss of only 0.03 for proximal femoral fracture.
Copyright © 2020 Elsevier B.V. All rights reserved.

Entities:  

Keywords:  Artificial intelligence; Deep learning; Fracture; Machine learning; Radiographs

Mesh:

Year:  2020        PMID: 32623269      PMCID: PMC7483754          DOI: 10.1016/j.ejrad.2020.109139

Source DB:  PubMed          Journal:  Eur J Radiol        ISSN: 0720-048X            Impact factor:   3.528


  24 in total

1.  Detecting intertrochanteric hip fractures with orthopedist-level accuracy using a deep convolutional neural network.

Authors:  Takaaki Urakawa; Yuki Tanaka; Shinichi Goto; Hitoshi Matsuzawa; Kei Watanabe; Naoto Endo
Journal:  Skeletal Radiol       Date:  2018-06-28       Impact factor: 2.199

2.  Staged reconstruction of pelvic ring disruption: differences in morbidity, mortality, radiologic results, and functional outcomes between B1, B2/B3, and C-type lesions.

Authors:  P M Rommens; M H Hessmann
Journal:  J Orthop Trauma       Date:  2002-02       Impact factor: 2.512

3.  Computerized Bone Age Estimation Using Deep Learning Based Program: Evaluation of the Accuracy and Efficiency.

Authors:  Jeong Rye Kim; Woo Hyun Shim; Hee Mang Yoon; Sang Hyup Hong; Jin Seong Lee; Young Ah Cho; Sangki Kim
Journal:  AJR Am J Roentgenol       Date:  2017-09-12       Impact factor: 3.959

4.  Acetabular fractures: a 16-year prospective epidemiological study.

Authors:  A Laird; J F Keating
Journal:  J Bone Joint Surg Br       Date:  2005-07

5.  Epidemiology of hip fracture: Worldwide geographic variation.

Authors:  Dinesh K Dhanwal; Elaine M Dennison; Nick C Harvey; Cyrus Cooper
Journal:  Indian J Orthop       Date:  2011-01       Impact factor: 1.251

6.  Imaging of occult hip fractures: CT or MRI?

Authors:  Haroon Rehman; Rhys G E Clement; Fergus Perks; Timothy O White
Journal:  Injury       Date:  2016-03-03       Impact factor: 2.586

7.  Falls: epidemiology, pathophysiology, and relationship to fracture.

Authors:  Sarah D Berry; Ram R Miller
Journal:  Curr Osteoporos Rep       Date:  2008-12       Impact factor: 5.096

8.  Pelvic fractures: value of plain radiography in early assessment and management.

Authors:  J W Young; A R Burgess; R J Brumback; A Poka
Journal:  Radiology       Date:  1986-08       Impact factor: 11.105

Review 9.  A systematic review of hip fracture incidence and probability of fracture worldwide.

Authors:  J A Kanis; A Odén; E V McCloskey; H Johansson; D A Wahl; C Cooper
Journal:  Osteoporos Int       Date:  2012-03-15       Impact factor: 4.507

10.  Deep learning predicts hip fracture using confounding patient and healthcare variables.

Authors:  Marcus A Badgeley; John R Zech; Luke Oakden-Rayner; Benjamin S Glicksberg; Manway Liu; William Gale; Michael V McConnell; Bethany Percha; Thomas M Snyder; Joel T Dudley
Journal:  NPJ Digit Med       Date:  2019-04-30
View more
  4 in total

1.  Hanging protocol optimization of lumbar spine radiographs with machine learning.

Authors:  Gene Kitamura
Journal:  Skeletal Radiol       Date:  2021-02-15       Impact factor: 2.128

Review 2.  Applications of Machine Learning in Bone and Mineral Research.

Authors:  Sung Hye Kong; Chan Soo Shin
Journal:  Endocrinol Metab (Seoul)       Date:  2021-10-21

3.  Hybrid SFNet Model for Bone Fracture Detection and Classification Using ML/DL.

Authors:  Dhirendra Prasad Yadav; Ashish Sharma; Senthil Athithan; Abhishek Bhola; Bhisham Sharma; Imed Ben Dhaou
Journal:  Sensors (Basel)       Date:  2022-08-04       Impact factor: 3.847

4.  Deep learning accurately classifies elbow joint effusion in adult and pediatric radiographs.

Authors:  Jarno T Huhtanen; Mikko Nyman; Dorin Doncenco; Maral Hamedian; Davis Kawalya; Leena Salminen; Roberto Blanco Sequeiros; Seppo K Koskinen; Tomi K Pudas; Sami Kajander; Pekka Niemi; Jussi Hirvonen; Hannu J Aronen; Mojtaba Jafaritadi
Journal:  Sci Rep       Date:  2022-07-12       Impact factor: 4.996

  4 in total

北京卡尤迪生物科技股份有限公司 © 2022-2023.