Prognostic prediction of idiopathic membranous nephropathy using interpretable machine learning.

BACKGROUND: Established prognostic models of idiopathic membranous nephropathy (IMN) were limited to traditional modeling methods and did not comprehensively consider clinical and pathological patient data. Based on the electronic medical record (EMR) system, machine learning (ML) was used to construct a risk prediction model for the prognosis of IMN. METHODS: Data from 418 patients with IMN were diagnosed by renal biopsy at the Fifth Clinical Medical College of Shanxi Medical University. Fifty-nine medical features of the patients could be obtained from EMR, and prediction models were established based on five ML algorithms. The area under the curve, recall rate, accuracy, and F1 were used to evaluate and compare the performances of the models. Shapley additive explanation (SHAP) was used to explain the results of the best-performing model. RESULTS: One hundred and seventeen patients (28.0%) with IMN experienced adverse events, 28 of them had compound outcomes (ESRD or double serum creatinine (SCr)), and 89 had relapsed. The gradient boosting machine (LightGBM) model had the best performance, with the highest AUC (0.892 ± 0.052, 95% CI 0.840-0.945), accuracy (0.909 ± 0.016), recall (0.741 ± 0.092), precision (0.906 ± 0.027), and F1 (0.905 ± 0.020). Recursive feature elimination with random forest and SHAP plots based on LightGBM showed that anti-phospholipase A2 receptor (anti-PLA2R), immunohistochemical immunoglobulin G4 (IHC IgG4), D-dimer (D-DIMER), triglyceride (TG), serum albumin (ALB), aspartate transaminase (AST), ß2-microglobulin (BMG), SCr, and fasting plasma glucose (FPG) were important risk factors for the prognosis of IMN. Increased risk of adverse events in IMN patients was correlated with high anti-PLA2R and low IHC IgG4. CONCLUSIONS: This study established a risk prediction model for the prognosis of IMN using ML based on clinical and pathological patient data. The LightGBM model may become a tool for personalized management of IMN patients.

Boosting the Utilization and Electrochemical Performances of Polyaniline by Forming a Binder-Free Nanoscale Coaxially Coated Polyaniline/Carbon Nanotube/Carbon Fiber Paper Hierarchical 3D Microstructure Composite as a Supercapacitor Electrode.

Nanoscale polyaniline (PANI) is formed on a hierarchical 3D microstructure carbon nanotubes (CNTs)/carbon fiber paper (CFP) substrate via a one-step electrochemical polymerization method. The chemical and structural properties of the binder-free PANI/CNTs/CFP electrode are characterized by field emission scanning electron microscopy, transmission electron microscopy, Fourier transform infrared spectroscopy, and Raman spectroscopy. The specific capacitance of PANI/CNTs/CFP tested in a symmetric two-electrode system reaches 731.6 mF·cm-2 (1354.7 F·g-1) at a current density of 1 mA·cm-2 (1.8 A·g-1). The symmetric supercapacitor device demonstrates excellent cycling performance up to 10,000 cycles with a capacitance retention of 81.4% at a current density of 1 mA·cm-2 (1.8 A·g-1). The results demonstrate that the binder-free CNTs/CFP composite is a strong backbone for depositing ultrathin PANI layers at a high mass loading. The hierarchical 3D microstructure PANI/CNTs/CFP provides enough space and transporting channels to form an efficient electrode-electrolyte interface for the supercapacitance reaction. The formed nanoscale PANI film coaxially coated on the sidewalls of CNTs enables efficient charge transfer and a shortened diffusion length. Hence, the utilization efficiency and electrochemical performances of PANI are significantly improved. The rational design strategy of a CNT-based binder-free hierarchical 3D microstructure can be used in preparing various advanced energy-storage electrodes for electrochemical energy-storage and conversion systems.