Broad Learning Enhanced ¹ H-MRS for Early Diagnosis of Neuropsychiatric Systemic Lupus Erythematosus.

Publisher: Hindawi Country of Publication: United States NLM ID: 101277751 Publication Model: eCollection Cited Medium: Internet ISSN: 1748-6718 (Electronic) Linking ISSN: 1748670X NLM ISO Abbreviation: Comput Math Methods Med Subsets: MEDLINE

Publication: 2011-2024 : New York : Hindawi
Original Publication: London : Taylor & Francis, c2006-

Support Vector Machine*; Diagnosis, Computer-Assisted/*methods ; Lupus Vasculitis, Central Nervous System/*diagnostic imaging ; Proton Magnetic Resonance Spectroscopy/*statistics & numerical data; Adult ; Brain/diagnostic imaging ; Brain/metabolism ; Case-Control Studies ; Computational Biology ; Diagnosis, Computer-Assisted/statistics & numerical data ; Early Diagnosis ; Female ; Functional Neuroimaging/statistics & numerical data ; Humans ; Lupus Vasculitis, Central Nervous System/metabolism ; Magnetic Resonance Imaging/statistics & numerical data ; Male ; Retrospective Studies

In this paper, we explore the potential of using the multivoxel proton magnetic resonance spectroscopy ( ¹ H-MRS) to diagnose neuropsychiatric systemic lupus erythematosus (NPSLE) with the assistance of a support vector machine broad learning system (BL-SVM). We retrospectively analysed 23 confirmed patients and 16 healthy controls, who underwent a 3.0 T magnetic resonance imaging (MRI) sequence with multivoxel ¹ H-MRS in our hospitals. One hundred and seventeen metabolic features were extracted from the multivoxel ¹ H-MRS image. Thirty-three metabolic features selected by the Mann-Whitney U test were considered to have a statistically significant difference ( p < 0.05). However, the best accuracy achieved by conventional statistical methods using these 33 metabolic features was only 77%. We turned to develop a support vector machine broad learning system (BL-SVM) to quantitatively analyse the metabolic features from ¹ H-MRS. Although not all the individual features manifested statistics significantly, the BL-SVM could still learn to distinguish the NPSLE from the healthy controls. The area under the receiver operating characteristic curve (AUC), the sensitivity, and the specificity of our BL-SVM in predicting NPSLE were 95%, 95.8%, and 93%, respectively, by 3-fold cross-validation. We consequently conclude that the proposed system effectively and efficiently working on limited and noisy samples may brighten a noinvasive in vivo instrument for early diagnosis of NPSLE.
Competing Interests: The authors declare that they have no conflict of interest.
(Copyright © 2020 Yan Li et al.)

J Neuroimmunol. 2009 Jul 25;212(1-2):3-9. (PMID: 19500858)
Eur Neuropsychopharmacol. 2019 Dec;29(2):222-234. (PMID: 30558824)
Nature. 2015 May 28;521(7553):436-44. (PMID: 26017442)
J Autoimmun. 2016 Nov;74:41-72. (PMID: 27427403)
Clin Exp Ophthalmol. 2005 Oct;33(5):548-9. (PMID: 16181294)
Lupus. 2020 Apr;29(5):499-504. (PMID: 32138582)
Eur Radiol. 2019 Jun;29(6):2792-2801. (PMID: 30569184)
Magn Reson Med. 2019 Aug;82(2):527-550. (PMID: 30919510)
ACS Chem Neurosci. 2020 Nov 12;:. (PMID: 33179901)
Front Med (Lausanne). 2018 Dec 04;5:340. (PMID: 30564579)
J Nanobiotechnology. 2019 Dec 17;17(1):123. (PMID: 31847857)
Comput Biol Med. 2017 Jun 1;85:1-6. (PMID: 28412572)
IEEE Trans Med Imaging. 2018 Oct;37(10):2290-2297. (PMID: 29994303)
Mod Rheumatol. 2020 Mar;30(2):321-326. (PMID: 30836042)
IEEE Trans Med Imaging. 2018 Jan;37(1):251-261. (PMID: 28866485)
Korean J Intern Med. 2016 Nov;31(6):1171-1177. (PMID: 27017392)
IEEE/ACM Trans Comput Biol Bioinform. 2019 Nov 15;PP:. (PMID: 31751282)
Lupus. 2014;23(1):10-9. (PMID: 24192079)
IEEE Trans Neural Netw Learn Syst. 2018 Jan;29(1):10-24. (PMID: 28742048)
Lupus. 2003;12(12):897-902. (PMID: 14714908)
NMR Biomed. 2015 Mar;28(3):376-83. (PMID: 25615718)
Arthritis Rheum. 1999 Apr;42(4):599-608. (PMID: 10211873)
Acad Radiol. 2013 Oct;20(10):1286-96. (PMID: 24029061)
Neurosci Biobehav Rev. 2020 Jan;108:854-865. (PMID: 31838195)
Sci Rep. 2016 Jan 13;6:19026. (PMID: 26758023)
Arthritis Rheumatol. 2015 Jun;67(6):1598-610. (PMID: 25709106)
Nat Rev Dis Primers. 2016 Jun 16;2:16039. (PMID: 27306639)
J Neurol Sci. 2019 Jun 15;401:19-26. (PMID: 31005759)
Clin Exp Rheumatol. 2018 Nov-Dec;36(6):1003-1007. (PMID: 29846157)
Ann Rheum Dis. 2018 Mar;77(3):441-448. (PMID: 29259047)
Autoimmun Rev. 2018 Jan;17(1):36-43. (PMID: 29108821)
Autoimmun Rev. 2019 Apr;18(4):426-432. (PMID: 30763633)
Iran J Radiol. 2016 May 23;13(3):e32927. (PMID: 27878065)
Neurology. 2020 Mar 17;94(11):e1147-e1156. (PMID: 31919113)
Lupus. 2014 Sep;23(10):1073-8. (PMID: 25118052)
Front Neurol. 2018 Oct 23;9:901. (PMID: 30405523)
Bipolar Disord. 2016 Nov;18(7):583-590. (PMID: 27870506)
IEEE Trans Med Imaging. 2019 Jan;38(1):312-321. (PMID: 30106676)
Lupus. 2004;13(3):149-58. (PMID: 15119542)
Inflammopharmacology. 2019 Jun;27(3):511-520. (PMID: 30911862)
IEEE Trans Neural Netw Learn Syst. 2020 Oct 20;PP:. (PMID: 33079681)
Arthritis Rheumatol. 2016 Aug;68(8):1945-54. (PMID: 26894522)
Clin Rheumatol. 2016 Oct;35(10):2449-56. (PMID: 27527090)
Arthritis Rheumatol. 2019 Jan;71(1):33-42. (PMID: 29927108)
Eur Child Adolesc Psychiatry. 2020 Feb 1;:. (PMID: 32008167)
Arthritis Rheum. 2010 Jul;62(7):2055-63. (PMID: 20309864)
Neuroimage Clin. 2019;23:101835. (PMID: 31035232)
Lupus. 2019 Aug;28(9):1128-1133. (PMID: 31315530)

Date Created: 20201210 Date Completed: 20210827 Latest Revision: 20210827

20240628

PMC7704182

10.1155/2020/8874521

33299467