Deep learning analysis of clinical course of primary nephrotic syndrome: Japan Nephrotic Syndrome Cohort Study (JNSCS).
Clinical course
Creatinine
Hematuria
Machine learning
Nephrotic syndrome
Prognosis
Proteinuria
Journal
Clinical and experimental nephrology
ISSN: 1437-7799
Titre abrégé: Clin Exp Nephrol
Pays: Japan
ID NLM: 9709923
Informations de publication
Date de publication:
Dec 2022
Dec 2022
Historique:
received:
24
03
2022
accepted:
16
07
2022
pubmed:
13
8
2022
medline:
19
11
2022
entrez:
12
8
2022
Statut:
ppublish
Résumé
Prognosis of nephrotic syndrome has been evaluated based on pathological diagnosis, whereas its clinical course is monitored using objective items and the treatment strategy is largely the same. We examined whether the entire natural history of nephrotic syndrome could be evaluated using objective common clinical items. Machine learning clustering was performed on 205 cases from the Japan Nephrotic Syndrome Cohort Study, whose clinical parameters, serum creatinine, serum albumin, dipstick hematuria, and proteinuria were traceable after kidney biopsy at 5 measured points up to 2 years. The clinical patterns of time-series data were learned using long short-term memory (LSTM)-encoder-decoder architecture, an unsupervised machine learning classifier. Clinical clusters were defined as Gaussian mixture distributions in a two-dimensional scatter plot based on the highest log-likelihood. Time-series data of nephrotic syndrome were classified into four clusters. Patients in the fourth cluster showed the increase in serum creatinine in the later part of the follow-up period. Patients in both the third and fourth clusters were initially high in both hematuria and proteinuria, whereas a lack of decline in the urinary protein level preceded the worsening of kidney function in fourth cluster. The original diseases of fourth cluster included all the disease studied in this cohort. Four kinds of clinical courses were identified in nephrotic syndrome. This classified clinical course may help objectively grasp the actual condition or treatment resistance of individual patients with nephrotic syndrome.
Sections du résumé
BACKGROUND
BACKGROUND
Prognosis of nephrotic syndrome has been evaluated based on pathological diagnosis, whereas its clinical course is monitored using objective items and the treatment strategy is largely the same. We examined whether the entire natural history of nephrotic syndrome could be evaluated using objective common clinical items.
METHODS
METHODS
Machine learning clustering was performed on 205 cases from the Japan Nephrotic Syndrome Cohort Study, whose clinical parameters, serum creatinine, serum albumin, dipstick hematuria, and proteinuria were traceable after kidney biopsy at 5 measured points up to 2 years. The clinical patterns of time-series data were learned using long short-term memory (LSTM)-encoder-decoder architecture, an unsupervised machine learning classifier. Clinical clusters were defined as Gaussian mixture distributions in a two-dimensional scatter plot based on the highest log-likelihood.
RESULTS
RESULTS
Time-series data of nephrotic syndrome were classified into four clusters. Patients in the fourth cluster showed the increase in serum creatinine in the later part of the follow-up period. Patients in both the third and fourth clusters were initially high in both hematuria and proteinuria, whereas a lack of decline in the urinary protein level preceded the worsening of kidney function in fourth cluster. The original diseases of fourth cluster included all the disease studied in this cohort.
CONCLUSIONS
CONCLUSIONS
Four kinds of clinical courses were identified in nephrotic syndrome. This classified clinical course may help objectively grasp the actual condition or treatment resistance of individual patients with nephrotic syndrome.
Identifiants
pubmed: 35962244
doi: 10.1007/s10157-022-02256-3
pii: 10.1007/s10157-022-02256-3
pmc: PMC9668942
doi:
Substances chimiques
Creatinine
AYI8EX34EU
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
1170-1179Subventions
Organisme : Ministry of Health, Labour and Welfare
ID : 2031693
Informations de copyright
© 2022. The Author(s).
Références
Cell. 2019 Apr 4;177(2):463-477.e15
pubmed: 30951672
Clin Exp Nephrol. 2012 Dec;16(6):903-20
pubmed: 23053590
Lupus. 2014 Oct;23(11):1124-32
pubmed: 24860121
J Thromb Haemost. 2014 Sep;12(9):1449-54
pubmed: 25040558
Am J Kidney Dis. 2004 Feb;43(2):368-82
pubmed: 14750104
Am J Med. 2014 Sep;127(9):871-7.e1
pubmed: 24838191
Sci Rep. 2016 May 18;6:26137
pubmed: 27188851
Kidney Int. 1993 Sep;44(3):638-42
pubmed: 8231039
Sci Rep. 2019 Mar 25;9(1):5104
pubmed: 30911057
Clin Exp Nephrol. 2018 Feb;22(1):55-60
pubmed: 28478493
Clin Exp Nephrol. 2016 Jun;20(3):342-70
pubmed: 27099136
Cochrane Database Syst Rev. 2012 Apr 18;(4):CD003964
pubmed: 22513919
Kidney Blood Press Res. 2017;42(6):1045-1052
pubmed: 29197864
Nephrol Dial Transplant. 2011 Feb;26(2):414-30
pubmed: 21068142
Kidney Int. 2004 Apr;65(4):1400-7
pubmed: 15086481
Nat Rev Nephrol. 2018 Jan;14(1):57-70
pubmed: 29176657
Clin Exp Nephrol. 2011 Aug;15(4):493-503
pubmed: 21437579
Clin Exp Nephrol. 2021 Aug;25(8):893-901
pubmed: 33768329
Sci Data. 2019 Jun 17;6(1):96
pubmed: 31209213
Clin Exp Nephrol. 2018 Dec;22(6):1266-1280
pubmed: 29679356