Literature DB >> 33819478

Diagnostic Accuracy and Failure Mode Analysis of a Deep Learning Algorithm for the Detection of Intracranial Hemorrhage.

Andrew F Voter1, Ece Meram2, John W Garrett2, John-Paul J Yu3.   

Abstract

OBJECTIVE: To determine the institutional diagnostic accuracy of an artificial intelligence (AI) decision support systems (DSS), Aidoc, in diagnosing intracranial hemorrhage (ICH) on noncontrast head CTs and to assess the potential generalizability of an AI DSS.
METHODS: This retrospective study included 3,605 consecutive, emergent, adult noncontrast head CT scans performed between July 1, 2019, and December 30, 2019, at our institution (51% female subjects, mean age of 61 ± 21 years). Each scan was evaluated for ICH by both a certificate of added qualification certified neuroradiologist and Aidoc. We determined the diagnostic accuracy of the AI model and performed a failure mode analysis with quantitative CT radiomic image characterization.
RESULTS: Of the 3,605 scans, 349 cases of ICH (9.7% of studies) were identified. The neuroradiologist and Aidoc interpretations were concordant in 96.9% of cases and the overall sensitivity, specificity, positive predictive value, and negative predictive value were 92.3%, 97.7%, 81.3%, and 99.2%, respectively, with positive predictive values unexpectedly lower than in previously reported studies. Prior neurosurgery, type of ICH, and number of ICHs were significantly associated with decreased model performance. Quantitative image characterization with CT radiomics failed to reveal significant differences between concordant and discordant studies. DISCUSSION: This study revealed decreased diagnostic accuracy of an AI DSS at our institution. Despite extensive evaluation, we were unable to identify the source of this discrepancy, raising concerns about the generalizability of these tools with indeterminate failure modes. These results further highlight the need for standardized study design to allow for rigorous and reproducible site-to-site comparison of emerging deep learning technologies.
Copyright © 2021 American College of Radiology. Published by Elsevier Inc. All rights reserved.

Entities:  

Keywords:  Artificial intelligence; decision support systems; generalizability; intracranial hemorrhage; noncontrast head CT

Mesh:

Year:  2021        PMID: 33819478      PMCID: PMC8349782          DOI: 10.1016/j.jacr.2021.03.005

Source DB:  PubMed          Journal:  J Am Coll Radiol        ISSN: 1546-1440            Impact factor:   6.240


  32 in total

1.  Implementation of the Canadian CT Head Rule and Its Association With Use of Computed Tomography Among Patients With Head Injury.

Authors:  Adam L Sharp; Brian Z Huang; Tania Tang; Ernest Shen; Edward R Melnick; Arjun K Venkatesh; Michael H Kanter; Michael K Gould
Journal:  Ann Emerg Med       Date:  2017-07-21       Impact factor: 5.721

2.  Practical guidelines for handling head and neck computed tomography artifacts for quantitative image analysis.

Authors:  Rachel B Ger; Daniel F Craft; Dennis S Mackin; Shouhao Zhou; Rick R Layman; A Kyle Jones; Hesham Elhalawani; Clifton D Fuller; Rebecca M Howell; Heng Li; R Jason Stafford; Laurence E Court
Journal:  Comput Med Imaging Graph       Date:  2018-09-15       Impact factor: 4.790

3.  Does image normalization and intensity resolution impact texture classification?

Authors:  Marcin Kociołek; Michał Strzelecki; Rafał Obuchowicz
Journal:  Comput Med Imaging Graph       Date:  2020-03-06       Impact factor: 4.790

Review 4.  Accuracy of deep learning for automated detection of pneumonia using chest X-Ray images: A systematic review and meta-analysis.

Authors:  Yuanyuan Li; Zhenyan Zhang; Cong Dai; Qiang Dong; Samireh Badrigilan
Journal:  Comput Biol Med       Date:  2020-07-14       Impact factor: 4.589

5.  Indications for surgical treatment of putaminal hemorrhage. Comparative study based on serial CT and time-course analysis.

Authors:  K Fujitsu; M Muramoto; Y Ikeda; Y Inada; I Kim; T Kuwabara
Journal:  J Neurosurg       Date:  1990-10       Impact factor: 5.115

Review 6.  Artificial intelligence and radiomics in pulmonary nodule management: current status and future applications.

Authors:  S Ather; T Kadir; F Gleeson
Journal:  Clin Radiol       Date:  2019-06-12       Impact factor: 2.350

7.  Modified ideal observer model (MIOM) for high-contrast and high-spatial resolution CT imaging tasks.

Authors:  Juan P Cruz-Bastida; Daniel Gomez-Cardona; John Garrett; Timothy Szczykutowicz; Guang-Hong Chen; Ke Li
Journal:  Med Phys       Date:  2017-07-18       Impact factor: 4.071

8.  Inconsistent Performance of Deep Learning Models on Mammogram Classification.

Authors:  Xiaoqin Wang; Gongbo Liang; Yu Zhang; Hunter Blanton; Zachary Bessinger; Nathan Jacobs
Journal:  J Am Coll Radiol       Date:  2020-02-14       Impact factor: 5.532

9.  Analysis of head CT scans flagged by deep learning software for acute intracranial hemorrhage.

Authors:  Daniel T Ginat
Journal:  Neuroradiology       Date:  2019-12-11       Impact factor: 2.804

Review 10.  Convolutional neural networks: an overview and application in radiology.

Authors:  Rikiya Yamashita; Mizuho Nishio; Richard Kinh Gian Do; Kaori Togashi
Journal:  Insights Imaging       Date:  2018-06-22
View more
  3 in total

1.  Semisupervised Training of a Brain MRI Tumor Detection Model Using Mined Annotations.

Authors:  Nathaniel C Swinburne; Vivek Yadav; Julie Kim; Ye R Choi; David C Gutman; Jonathan T Yang; Nelson Moss; Jacqueline Stone; Jamie Tisnado; Vaios Hatzoglou; Sofia S Haque; Sasan Karimi; John Lyo; Krishna Juluru; Karl Pichotta; Jianjiong Gao; Sohrab P Shah; Andrei I Holodny; Robert J Young
Journal:  Radiology       Date:  2022-01-18       Impact factor: 11.105

2.  Interpretable, not black-box, artificial intelligence should be used for embryo selection.

Authors:  Michael Anis Mihdi Afnan; Yanhe Liu; Vincent Conitzer; Cynthia Rudin; Abhishek Mishra; Julian Savulescu; Masoud Afnan
Journal:  Hum Reprod Open       Date:  2021-11-02

3.  Deep learning algorithm in detecting intracranial hemorrhages on emergency computed tomographies.

Authors:  Almut Kundisch; Alexander Hönning; Sven Mutze; Lutz Kreissl; Frederik Spohn; Johannes Lemcke; Maximilian Sitz; Paul Sparenberg; Leonie Goelz
Journal:  PLoS One       Date:  2021-11-29       Impact factor: 3.240

  3 in total

北京卡尤迪生物科技股份有限公司 © 2022-2023.