Profiling of kidney involvement in systemic lupus erythematosus by deep learning using the National Database of Designated Incurable Diseases of Japan.

BACKGROUND: Kidney involvement frequently occurs in systemic lupus erythematosus (SLE), and its clinical manifestations are complicated. We profiled kidney involvement in SLE patients using deep learning based on data from the National Database of Designated Incurable Diseases of Japan. METHODS: We analyzed the cross-sectional data of 1655 patients with SLE whose Personal Clinical Records were newly registered between 2015 and 2017. We trained an artificial neural network using clinical data, and the extracted characteristics were evaluated using an autoencoder. We tested the difference of population proportions to analyze the correlation between the presence or absence of kidney involvement and that of other clinical manifestations. RESULTS: Data of patients with SLE were compressed in a feature space in which the anti-double-stranded deoxyribonucleic acid (anti-dsDNA) antibody titer, antinuclear antibody titer, or white blood cell count contributed significantly to distinguishing patients. Many SLE manifestations were accompanied by kidney involvement, whereas in a subgroup of patients with high anti-dsDNA antibody titers and low antinuclear antibody titers, kidney involvement was positively and negatively correlated with hemolytic anemia and inflammatory manifestations, respectively. CONCLUSION: Although there are various combinations of SLE manifestations, our study revealed that some of them are specific to kidney involvement. SLE profiles extracted from the objective analysis will be useful for categorizing SLE manifestations.

Deep learning analysis of clinical course of primary nephrotic syndrome: Japan Nephrotic Syndrome Cohort Study (JNSCS).

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: 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: 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: 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.