Literature DB >> 34741469

Development and Validation of a Deep Learning Strategy for Automated View Classification of Pediatric Focused Assessment With Sonography for Trauma.

Aaron E Kornblith1,2,3, Newton Addo1,4, Ruolei Dong5,6, Robert Rogers7, Jacqueline Grupp-Phelan1,2, Atul Butte2,3, Pavan Gupta7, Rachael A Callcut7,8, Rima Arnaout3,4.   

Abstract

OBJECTIVE: Pediatric focused assessment with sonography for trauma (FAST) is a sequence of ultrasound views rapidly performed by clinicians to diagnose hemorrhage. A technical limitation of FAST is the lack of expertise to consistently acquire all required views. We sought to develop an accurate deep learning view classifier using a large heterogeneous dataset of clinician-performed pediatric FAST.
METHODS: We developed and conducted a retrospective cohort analysis of a deep learning view classifier on real-world FAST studies performed on injured children less than 18 years old in two pediatric emergency departments by 30 different clinicians. FAST was randomly distributed to training, validation, and test datasets, 70:20:10; each child was represented in only one dataset. The primary outcome was view classifier accuracy for video clips and still frames.
RESULTS: There were 699 FAST studies, representing 4925 video clips and 1,062,612 still frames, performed by 30 different clinicians. The overall classification accuracy was 97.8% (95% confidence interval [CI]: 96.0-99.0) for video clips and 93.4% (95% CI: 93.3-93.6) for still frames. Per view still frames were classified with an accuracy: 96.0% (95% CI: 95.9-96.1) cardiac, 99.8% (95% CI: 99.8-99.8) pleural, 95.2% (95% CI: 95.0-95.3) abdominal upper quadrants, and 95.9% (95% CI: 95.8-96.0) suprapubic.
CONCLUSION: A deep learning classifier can accurately predict pediatric FAST views. Accurate view classification is important for quality assurance and feasibility of a multi-stage deep learning FAST model to enhance the evaluation of injured children.
© 2021 American Institute of Ultrasound in Medicine.

Entities:  

Keywords:  abdominal injuries/diagnostic imaging; deep learning; machine learning; pediatric trauma; ultrasonography

Mesh:

Year:  2021        PMID: 34741469      PMCID: PMC9072593          DOI: 10.1002/jum.15868

Source DB:  PubMed          Journal:  J Ultrasound Med        ISSN: 0278-4297            Impact factor:   2.754


  27 in total

Review 1.  Artificial intelligence in ultrasound.

Authors:  Yu-Ting Shen; Liang Chen; Wen-Wen Yue; Hui-Xiong Xu
Journal:  Eur J Radiol       Date:  2021-04-12       Impact factor: 3.528

2.  Transfer Learning with Convolutional Neural Networks for Classification of Abdominal Ultrasound Images.

Authors:  Phillip M Cheng; Harshawn S Malhi
Journal:  J Digit Imaging       Date:  2017-04       Impact factor: 4.056

3.  Artificial intelligence and machine learning in emergency medicine.

Authors:  Jonathon Stewart; Peter Sprivulis; Girish Dwivedi
Journal:  Emerg Med Australas       Date:  2018-07-16       Impact factor: 2.151

Review 4.  Focused Assessment with Sonography in Trauma (FAST) in 2017: What Radiologists Can Learn.

Authors:  John R Richards; John P McGahan
Journal:  Radiology       Date:  2017-04       Impact factor: 11.105

5.  Randomized controlled clinical trial of point-of-care, limited ultrasonography for trauma in the emergency department: the first sonography outcomes assessment program trial.

Authors:  Lawrence A Melniker; Evan Leibner; Mark G McKenney; Peter Lopez; William M Briggs; Carol A Mancuso
Journal:  Ann Emerg Med       Date:  2006-03-24       Impact factor: 5.721

6.  An ensemble of neural networks provides expert-level prenatal detection of complex congenital heart disease.

Authors:  Rima Arnaout; Lara Curran; Yili Zhao; Jami C Levine; Erin Chinn; Anita J Moon-Grady
Journal:  Nat Med       Date:  2021-05-14       Impact factor: 53.440

7.  Peritoneal fluid localization on FAST examination in the pediatric trauma patient.

Authors:  Timothy E Brenkert; Cynthia Adams; Rebecca L Vieira; Rachel G Rempell
Journal:  Am J Emerg Med       Date:  2017-04-14       Impact factor: 2.469

8.  Validation of hand motion analysis as an objective assessment tool for the Focused Assessment with Sonography for Trauma examination.

Authors:  Markus T Ziesmann; Jason Park; Bertram Unger; Andrew W Kirkpatrick; Ashley Vergis; Chau Pham; David Kirschner; Sarvesh Logestty; Lawrence M Gillman
Journal:  J Trauma Acute Care Surg       Date:  2015-10       Impact factor: 3.313

9.  Fast and accurate view classification of echocardiograms using deep learning.

Authors:  Ali Madani; Ramy Arnaout; Mohammad Mofrad; Rima Arnaout
Journal:  NPJ Digit Med       Date:  2018-03-21

10.  Deep learning interpretation of echocardiograms.

Authors:  Amirata Ghorbani; David Ouyang; Abubakar Abid; Bryan He; Jonathan H Chen; Robert A Harrington; David H Liang; Euan A Ashley; James Y Zou
Journal:  NPJ Digit Med       Date:  2020-01-24
View more

北京卡尤迪生物科技股份有限公司 © 2022-2023.