دورية أكاديمية
أعرض في EDS

Deep learning to estimate gestational age from fly-to cineloop videos: A novel approach to ultrasound quality control.

التفاصيل البيبلوغرافية
العنوان: Deep learning to estimate gestational age from fly-to cineloop videos: A novel approach to ultrasound quality control.
المؤلفون: Viswanathan AV; Department of Obstetrics and Gynecology, University of North Carolina School of Medicine, Chapel Hill, North Carolina, USA., Pokaprakarn T; Department of Obstetrics and Gynecology, University of North Carolina School of Medicine, Chapel Hill, North Carolina, USA.; Department of Biostatistics, University of North Carolina Gillings School of Global Public Health, Chapel Hill, North Carolina, USA., Kasaro MP; Department of Obstetrics and Gynecology, University of North Carolina School of Medicine, Chapel Hill, North Carolina, USA.; UNC Global Projects - Zambia LLC, Lusaka, Zambia., Shah HR; Department of Psychiatry, University of North Carolina School of Medicine, Chapel Hill, North Carolina, USA., Prieto JC; Department of Psychiatry, University of North Carolina School of Medicine, Chapel Hill, North Carolina, USA., Benabdelkader C; Department of Obstetrics and Gynecology, University of North Carolina School of Medicine, Chapel Hill, North Carolina, USA., Sebastião YV; Department of Obstetrics and Gynecology, University of North Carolina School of Medicine, Chapel Hill, North Carolina, USA., Sindano N; UNC Global Projects - Zambia LLC, Lusaka, Zambia., Stringer E; Department of Obstetrics and Gynecology, University of North Carolina School of Medicine, Chapel Hill, North Carolina, USA.; UNC Global Projects - Zambia LLC, Lusaka, Zambia., Stringer JSA; Department of Obstetrics and Gynecology, University of North Carolina School of Medicine, Chapel Hill, North Carolina, USA.; UNC Global Projects - Zambia LLC, Lusaka, Zambia.
المصدر: International journal of gynaecology and obstetrics: the official organ of the International Federation of Gynaecology and Obstetrics [Int J Gynaecol Obstet] 2024 Jun; Vol. 165 (3), pp. 1013-1021. Date of Electronic Publication: 2024 Jan 08.
نوع المنشور: Journal Article
اللغة: English
بيانات الدورية: Publisher: Wiley Country of Publication: United States NLM ID: 0210174 Publication Model: Print-Electronic Cited Medium: Internet ISSN: 1879-3479 (Electronic) Linking ISSN: 00207292 NLM ISO Abbreviation: Int J Gynaecol Obstet Subsets: MEDLINE
أسماء مطبوعة: Publication: 2017- : Malden, MA : Wiley
Original Publication: [New York, NY] Hoeber Medical Division, Harper & Row, [c1969-
مواضيع طبية MeSH: Ultrasonography, Prenatal*/standards , Ultrasonography, Prenatal*/methods , Deep Learning* , Gestational Age*, Humans ; Pregnancy ; Female ; Quality Control ; Video Recording ; Biometry/methods ; Pregnancy Trimester, Third ; Pregnancy Trimester, Second
مستخلص: Objective: Low-cost devices have made obstetric sonography possible in settings where it was previously unfeasible, but ensuring quality and consistency at scale remains a challenge. In the present study, we sought to create a tool to reduce substandard fetal biometry measurement while minimizing care disruption.
Methods: We developed a deep learning artificial intelligence (AI) model to estimate gestational age (GA) in the second and third trimester from fly-to cineloops-brief videos acquired during routine ultrasound biometry-and evaluated its performance in comparison to expert sonographer measurement. We then introduced random error into fetal biometry measurements and analyzed the ability of the AI model to flag grossly inaccurate measurements such as those that might be obtained by a novice.
Results: The mean absolute error (MAE) of our model (±standard error) was 3.87 ± 0.07 days, compared to 4.80 ± 0.10 days for expert biometry (difference -0.92 days; 95% CI: -1.10 to -0.76). Based on simulated novice biometry with average absolute error of 7.5%, our model reliably detected cases where novice biometry differed from expert biometry by 10 days or more, with an area under the receiver operating characteristics curve of 0.93 (95% CI: 0.92, 0.95), sensitivity of 81.0% (95% CI: 77.9, 83.8), and specificity of 89.9% (95% CI: 88.1, 91.5). These results held across a range of sensitivity analyses, including where the model was provided suboptimal truncated fly-to cineloops.
Conclusions: Our AI model estimated GA more accurately than expert biometry. Because fly-to cineloop videos can be obtained without any change to sonographer workflow, the model represents a no-cost guardrail that could be incorporated into both low-cost and commercial ultrasound devices to prevent reporting of most gross GA estimation errors.
(© 2024 International Federation of Gynecology and Obstetrics.)
References: Sci Rep. 2023 Feb 27;13(1):3322. (PMID: 36849625)
Curr Probl Diagn Radiol. 2017 May - Jun;46(3):210-215. (PMID: 28057388)
Front Glob Womens Health. 2022 Apr 05;3:880615. (PMID: 35449708)
Ultrasound Obstet Gynecol. 2009 Oct;34(4):387-94. (PMID: 19504627)
Cochrane Database Syst Rev. 2015 Jul 14;(7):CD007058. (PMID: 26171896)
Health Policy Plan. 2015 Feb;30(1):78-87. (PMID: 24357197)
Ultrasound Obstet Gynecol. 2016 Dec;48(6):719-726. (PMID: 26924421)
Int J Gynaecol Obstet. 2011 May;113(2):131-6. (PMID: 21315347)
NEJM Evid. 2022 May;1(5):. (PMID: 36875289)
Ultrasound Obstet Gynecol. 2009 Oct;34(4):395-403. (PMID: 19790099)
Int J Emerg Med. 2018 Feb 21;11(1):11. (PMID: 29468453)
Ultrasound Obstet Gynecol. 2019 Jun;53(6):715-723. (PMID: 31169958)
Radiology. 1984 Aug;152(2):497-501. (PMID: 6739822)
Trop Med Int Health. 2016 Mar;21(3):294-311. (PMID: 26683523)
J Digit Imaging. 2017 Apr;30(2):234-243. (PMID: 27896451)
Phys Med Biol. 2019 Sep 17;64(18):185010. (PMID: 31408850)
Int J Gynaecol Obstet. 2019 Jan;144(1):9-15. (PMID: 30267538)
Med Image Anal. 2017 Apr;37:22-36. (PMID: 28104551)
IEEE Trans Med Imaging. 2017 Nov;36(11):2204-2215. (PMID: 28708546)
Ultrasound Obstet Gynecol. 2018 Sep;52(3):332-339. (PMID: 28718938)
Ultrasound Obstet Gynecol. 2010 May;35(5):566-71. (PMID: 20183864)
Obstet Gynecol. 2017 May;129(5):e150-e154. (PMID: 28426621)
معلومات مُعتمدة: INV-003266 United States GATES Bill & Melinda Gates Foundation; OPP1191684 Bill and Melinda Gates Foundation; INV003266 Bill and Melinda Gates Foundation
فهرسة مساهمة: Keywords: artificial intelligence; biometry; deep learning; gestational age; quality control; ultrasound
تواريخ الأحداث: Date Created: 20240108 Date Completed: 20240512 Latest Revision: 20240630
رمز التحديث: 20240630
مُعرف محوري في PubMed: PMC11214162
DOI: 10.1002/ijgo.15321
PMID: 38189177
قاعدة البيانات: MEDLINE
الوصف
تدمد:1879-3479
DOI:10.1002/ijgo.15321