A Scalable Physician-level Deep Learning Algorithm Detects Universal Trauma on Pelvic Radiographs

Overview

Journal Nat Commun

Specialty Biology

Date 2021 Feb 17

PMID 33594071

Citations 39

Authors

Chi-Tung Cheng

Yirui Wang

Huan-Wu Chen

Po-Meng Hsiao

Chun-Nan Yeh

Chi-Hsun Hsieh

Shun Miao

Jing Xiao

Chien-Hung Liao

Le Lu

Affiliations

Soon will be listed here.

Abstract

Pelvic radiograph (PXR) is essential for detecting proximal femur and pelvis injuries in trauma patients, which is also the key component for trauma survey. None of the currently available algorithms can accurately detect all kinds of trauma-related radiographic findings on PXRs. Here, we show a universal algorithm can detect most types of trauma-related radiographic findings on PXRs. We develop a multiscale deep learning algorithm called PelviXNet trained with 5204 PXRs with weakly supervised point annotation. PelviXNet yields an area under the receiver operating characteristic curve (AUROC) of 0.973 (95% CI, 0.960-0.983) and an area under the precision-recall curve (AUPRC) of 0.963 (95% CI, 0.948-0.974) in the clinical population test set of 1888 PXRs. The accuracy, sensitivity, and specificity at the cutoff value are 0.924 (95% CI, 0.912-0.936), 0.908 (95% CI, 0.885-0.908), and 0.932 (95% CI, 0.919-0.946), respectively. PelviXNet demonstrates comparable performance with radiologists and orthopedics in detecting pelvic and hip fractures.

Citing Articles

Application and Prospects of Deep Learning Technology in Fracture Diagnosis.

Zhang J, Yang J, Wang X, Wang H, Zhou H, Yan Z Curr Med Sci. 2024; 44(6):1132-1140.

PMID: 39551854 DOI: 10.1007/s11596-024-2928-5.

Development and validation of deep learning models for identifying the brand of pedicle screws on plain spine radiographs.

Yao Y, Lin C, Chen H, Lin H, Hsiung W, Wang S JOR Spine. 2024; 7(3):e70001.

PMID: 39291095 PMC: 11406509. DOI: 10.1002/jsp2.70001.

Clinical validation of artificial intelligence-based preoperative virtual reduction for Neer 3- or 4-part proximal humerus fractures.

Jeon Y, Jung K, Kim M, Kim H, Yoon D, Park K BMC Musculoskelet Disord. 2024; 25(1):669.

PMID: 39192203 PMC: 11348784. DOI: 10.1186/s12891-024-07798-z.

Development and reporting of artificial intelligence in osteoporosis management.

Gatineau G, Shevroja E, Vendrami C, Gonzalez-Rodriguez E, Leslie W, Lamy O J Bone Miner Res. 2024; 39(11):1553-1573.

PMID: 39163489 PMC: 11523092. DOI: 10.1093/jbmr/zjae131.

How Can a Deep Learning Algorithm Improve Fracture Detection on X-rays in the Emergency Room?.

Reichert G, Bellamine A, Fontaine M, Naipeanu B, Altar A, Mejean E J Imaging. 2024; 7(7).

PMID: 39080893 PMC: 8321374. DOI: 10.3390/jimaging7070105.

References

Hallas P, Ellingsen T . Errors in fracture diagnoses in the emergency department--characteristics of patients and diurnal variation. BMC Emerg Med. 2006; 6:4. PMC: 1386703. DOI: 10.1186/1471-227X-6-4. View

Drukker K, Giger M, Joe B, Kerlikowske K, Greenwood H, Drukteinis J . Combined Benefit of Quantitative Three-Compartment Breast Image Analysis and Mammography Radiomics in the Classification of Breast Masses in a Clinical Data Set. Radiology. 2018; 290(3):621-628. PMC: 6394732. DOI: 10.1148/radiol.2018180608. View

Burlew C, Moore E, Stahel P, Geddes A, Wagenaar A, Pieracci F . Preperitoneal pelvic packing reduces mortality in patients with life-threatening hemorrhage due to unstable pelvic fractures. J Trauma Acute Care Surg. 2016; 82(2):233-242. PMC: 5250563. DOI: 10.1097/TA.0000000000001324. View

Chea P, Mandell J . Current applications and future directions of deep learning in musculoskeletal radiology. Skeletal Radiol. 2019; 49(2):183-197. DOI: 10.1007/s00256-019-03284-z. View

Urakawa T, Tanaka Y, Goto S, Matsuzawa H, Watanabe K, Endo N . Detecting intertrochanteric hip fractures with orthopedist-level accuracy using a deep convolutional neural network. Skeletal Radiol. 2018; 48(2):239-244. DOI: 10.1007/s00256-018-3016-3. View

Cheng C, Wang Y, Chen H, Hsiao P, Yeh C, Hsieh C . A scalable physician-level deep learning algorithm detects universal trauma on pelvic radiographs. Nat Commun. 2021; 12(1):1066. PMC: 7887334. DOI: 10.1038/s41467-021-21311-3. View

Chung S, Han S, Lee J, Oh K, Kim N, Yoon J . Automated detection and classification of the proximal humerus fracture by using deep learning algorithm. Acta Orthop. 2018; 89(4):468-473. PMC: 6066766. DOI: 10.1080/17453674.2018.1453714. View

Ebrahimighahnavieh M, Luo S, Chiong R . Deep learning to detect Alzheimer's disease from neuroimaging: A systematic literature review. Comput Methods Programs Biomed. 2019; 187:105242. DOI: 10.1016/j.cmpb.2019.105242. View

Lindsey R, Daluiski A, Chopra S, Lachapelle A, Mozer M, Sicular S . Deep neural network improves fracture detection by clinicians. Proc Natl Acad Sci U S A. 2018; 115(45):11591-11596. PMC: 6233134. DOI: 10.1073/pnas.1806905115. View

10.

Zhang L, Wang X, Yang D, Sanford T, Harmon S, Turkbey B . Generalizing Deep Learning for Medical Image Segmentation to Unseen Domains via Deep Stacked Transformation. IEEE Trans Med Imaging. 2020; 39(7):2531-2540. PMC: 7393676. DOI: 10.1109/TMI.2020.2973595. View

11.

Park S, Han K . Methodologic Guide for Evaluating Clinical Performance and Effect of Artificial Intelligence Technology for Medical Diagnosis and Prediction. Radiology. 2018; 286(3):800-809. DOI: 10.1148/radiol.2017171920. View

12.

Ureten K, Arslan T, Gultekin K, Demir A, Ozer H, Bilgili Y . Detection of hip osteoarthritis by using plain pelvic radiographs with deep learning methods. Skeletal Radiol. 2020; 49(9):1369-1374. DOI: 10.1007/s00256-020-03433-9. View

13.

Anwar S, Majid M, Qayyum A, Awais M, Alnowami M, Khan M . Medical Image Analysis using Convolutional Neural Networks: A Review. J Med Syst. 2018; 42(11):226. DOI: 10.1007/s10916-018-1088-1. View

14.

Cheng C, Ho T, Lee T, Chang C, Chou C, Chen C . Application of a deep learning algorithm for detection and visualization of hip fractures on plain pelvic radiographs. Eur Radiol. 2019; 29(10):5469-5477. PMC: 6717182. DOI: 10.1007/s00330-019-06167-y. View

15.

Nagendran M, Chen Y, Lovejoy C, Gordon A, Komorowski M, Harvey H . Artificial intelligence versus clinicians: systematic review of design, reporting standards, and claims of deep learning studies. BMJ. 2020; 368:m689. PMC: 7190037. DOI: 10.1136/bmj.m689. View

16.

Pfeifer R, Kalbas Y, Coimbra R, Leenen L, Komadina R, Hildebrand F . Indications and interventions of damage control orthopedic surgeries: an expert opinion survey. Eur J Trauma Emerg Surg. 2020; 47(6):2081-2092. DOI: 10.1007/s00068-020-01386-1. View

17.

Adams M, Chen W, Holcdorf D, McCusker M, Howe P, Gaillard F . Computer vs human: Deep learning versus perceptual training for the detection of neck of femur fractures. J Med Imaging Radiat Oncol. 2018; 63(1):27-32. DOI: 10.1111/1754-9485.12828. View

18.

Pinto A, Berritto D, Russo A, Riccitiello F, Caruso M, Belfiore M . Traumatic fractures in adults: missed diagnosis on plain radiographs in the Emergency Department. Acta Biomed. 2018; 89(1-S):111-123. PMC: 6179080. DOI: 10.23750/abm.v89i1-S.7015. View

19.

Patel U, Anwar A, Saleem S, Malik P, Rasul B, Patel K . Artificial intelligence as an emerging technology in the current care of neurological disorders. J Neurol. 2019; 268(5):1623-1642. DOI: 10.1007/s00415-019-09518-3. View

20.

Wang H, Jia H, Lu L, Xia Y . Thorax-Net: An Attention Regularized Deep Neural Network for Classification of Thoracic Diseases on Chest Radiography. IEEE J Biomed Health Inform. 2019; 24(2):475-485. DOI: 10.1109/JBHI.2019.2928369. View