Applicable Machine Learning Model for Predicting Contrast-induced Nephropathy Based on Pre-catheterization Variables.

Objective Contrast agents used for radiological examinations are an important cause of acute kidney injury (AKI). We developed and validated a machine learning and clinical scoring prediction model to stratify the risk of contrast-induced nephropathy, considering the limitations of current classical and machine learning models. Methods This retrospective study included 38,481 percutaneous coronary intervention cases from 23,703 patients in a tertiary hospital. We divided the cases into development and internal test sets (8:2). Using the development set, we trained a gradient boosting machine prediction model (complex model). We then developed a simple model using seven variables based on variable importance. We validated the performance of the models using an internal test set and tested them externally in two other hospitals. Results The complex model had the best area under the receiver operating characteristic (AUROC) curve at 0.885 [95% confidence interval (CI) 0.876-0.894] in the internal test set and 0.837 (95% CI 0.819-0.854) and 0.850 (95% CI 0.781-0.918) in two different external validation sets. The simple model showed an AUROC of 0.795 (95% CI 0.781-0.808) in the internal test set and 0.766 (95% CI 0.744-0.789) and 0.782 (95% CI 0.687-0.877) in the two different external validation sets. This was higher than the value in the well-known scoring system (Mehran criteria, AUROC=0.67). The seven precatheterization variables selected for the simple model were age, known chronic kidney disease, hematocrit, troponin I, blood urea nitrogen, base excess, and N-terminal pro-brain natriuretic peptide. The simple model is available at http://52.78.230.235:8081/Conclusions We developed an AKI prediction machine learning model with reliable performance. This can aid in bedside clinical decision making.

Antibody response in patients undergoing chronic hemodialysis post-severe acute respiratory syndrome coronavirus 2 vaccination: A prospective observational study.

Vaccination is important for patients undergoing hemodialysis (HD) to prevent coronavirus disease 2019 (COVID-19) infection since they are more vulnerable. However, they exhibit a weak response to vaccines, underscoring the importance of understanding whether antibodies are sufficiently produced and their durability post-COVID-19 vaccination. This prospective observational study assessed the antibody response of Korean patients undergoing HD for 1 year. We compared the antibody responses of patients undergoing HD to the COVID-19 vaccine with those of healthy volunteers from 2021 to 2022. The patient and control groups received 2 doses of ChAdOx1 nCoV-19 and mRNA-1273, respectively. Immunoglobulin G (IgG) and neutralizing antibody levels were measured weeks or months apart after 2 doses for 1 year using enzyme-linked immunosorbent and fluorescence-based competitive severe acute respiratory syndrome coronavirus 2 neutralizing assays, respectively. We analyzed the third dose's effect on the patient group by categorizing the group into patients who received the third dose and those who did not since it was initiated midway through the study. In the control group, we enrolled participants who had completed 3 doses of mRNA-1273 since almost all participants received the third dose. Thirty-two patients undergoing HD and 15 healthy participants who received 2 doses of ChAdOx1 nCoV-19 and 3 of mRNA-1273, respectively, were enrolled. Although antibody production was weaker in the patient group than in the control group (P < .001), patients showed an increase in IgG levels (0.408 ± 0.517 optical density (OD) pre-vaccination, 2.175 ± 1.241 OD in patients with 2 doses, and 2.134 ± 1.157 OD in patients with 3 doses 1 year after the second dose) and neutralizing antibodies (23 ± 8% pre-vaccination, 87 ± 23% in patients with 2 doses, and 89 ± 18% in patients with 3 doses 1 year after the second dose) post-vaccination (P < .001). In the patient group, 19 patients received a third dose (BNT162b2 or mRNA-1273); however, it did not increase the antibody levels (P = 1.000). Furthermore, the antibodies produced by the vaccination did not wane until 1 year. Two doses of vaccination resulted in a significant antibody response in patients undergoing HD, and antibody levels did not wane until 1 year.

Antibody response to COVID-19 vaccination in patients on chronic hemodialysis.

Purpose: Since patients on hemodialysis (HD) are known to be vulnerable to coronavirus disease 2019 (COVID-19), many studies were conducted regarding the effectiveness of the COVID-19 vaccine in HD patients in Western countries. Here, we assessed antibody response of HD patients for 6 months post-vaccination to identify the duration and effectiveness of the COVID-19 vaccine in the Asian population. Materials and Methods: We compared antibody response of the COVID-19 vaccine in HD patients with healthy volunteers. Patient and control groups had two doses of ChAdOx1 nCoV-19 and mRNA-1273, respectively. Immunoglobulin G (IgG) was measured before vaccination, 2 weeks after the first dose, 2 and 4 weeks, 3 and 6 months after the second dose. Neutralizing antibody was measured before vaccination and at 2 weeks, 3 and 6 months after second dose. Since the third dose was started in the middle of the study, we analyzed the effect of the third dose as well. Results: Although antibody production was weaker than the control group (n=22), the patient group (n=39) showed an increase in IgG and neutralizing antibody after two doses. And, 21/39 patients and 14/22 participants had a third dose (BNT162b2 or mRNA-1273 in the patient group, mRNA-1273 in the control group), and it did not affect antibody response in both group. Trend analysis showed IgG and neutralizing antibody did not decrease over time. Age, sex, and HD vintage did not affect antibody production in HD patients. Patients with higher body mass index displayed better seroresponse, while those on immunosuppressants showed poor seroresponse. Conclusion: Two doses of vaccination led to significant antibody response in HD patients, and the antibody did not wane until 6 months.

The importance of muscle mass in predicting intradialytic hypotension in patients undergoing maintenance hemodialysis.

BACKGROUND: Patients undergoing hemodialysis are susceptible to sarcopenia. As intracellular reservoirs of water, skeletal muscles are important contributors to intradialytic hypotension. This study was designed to determine the role of skeletal muscle mass in intradialytic hypotension. METHODS: In a cross-sectional study, the body composition of 177 patients was measured immediately after hemodialysis using bioelectrical impedance analysis. The parameters measured were skeletal muscle mass, intracellular and extracellular water contents, total body water, and cell-membrane functionality (in phase angle at 50 kHz). Data from laboratory tests, chest radiography, measurements of handgrip strength and mid-arm circumference, and questionnaires were collected. The main outcome was intradialytic hypotension, defined as more than two episodes of hypotension (systolic blood pressure of <90 mmHg) with intervention over the 3 months following enrollment. Logistic regression models including each parameter related to sarcopenia were compared with a clinical model. RESULTS: Patients with a low ratio of skeletal muscle mass to dry body weight (SMM/WT) had a higher rate of intradialytic hypotension (40.7%). Most low-SMM/WT patients were female, obese, diabetic, and had a lower handgrip strength compared with the other patients. In the high-SMM/WT group, the risk of intradialytic hypotension was lower, with an odds ratio of 0.08 (95% confidence interval [CI], 0.02-0.28) and adjusted odds ratio of 0.06 (95% CI, 0.01-0.29). CONCLUSION: Measurement and maintenance of skeletal muscle can help prevent intradialytic hypotension in frail patients undergoing hemodialysis.

Characterization of Medication Trends for Chronic Kidney Disease: Mineral and Bone Disorder Treatment Using Electronic Health Record-Based Common Data Model.

Chronic kidney disease-mineral bone disorder (CKD-MBD) is the most common complication in CKD patients. Although there is a consensus on treatment guidelines for CKD-MBD, it remains uncertain whether these treatment recommendations reflect actual practice. Therefore, the aim of this study was to investigate the CKD-MBD medication trend in real-world practice. This was a retrospective and observational study using a 12-year period database transformed into a common data model from three tertiary university hospitals. Study populations were subjects initially diagnosed as CKD. The date of diagnosis was designated as the index date. New patients were categorized year to year from 2008 to 2019 with a fixed observation period of 365 days to check the prescription of CKD-MBD medications including calcium-containing phosphate binder, noncalcium-containing phosphate binder, aluminium hydroxide, vitamin D receptor activator (VDRA), and cinacalcet. The numbers of CKD patients in the three hospitals were 7555, 2424, and 5351, respectively. The proportion for patients with CKD-MBD medication prescription decreased yearly regardless of hospital and CKD stage (p for trend < 0.05). The use of aluminium hydroxide disappeared steadily while the use of VDRA increased annually in all settings. Despite these changes in prescription patterns, the mean value for CKD-MBD-related serologic markers was almost within target range. The proportion of the population within the target value was not significantly changed. Irrespective of hospital and CKD stage, similar trends of prescription for CKD-MBD medications were observed in real-world practice. Further research with a distributed network study may be helpful to understand medication trends in CKD-MBD treatment.

Autonomous Assessment of Delamination Using Scarce Raw Structural Vibration and Transfer Learning.

Deep learning has helped achieve breakthroughs in a variety of applications; however, the lack of data from faulty states hinders the development of effective and robust diagnostic strategies using deep learning models. This work introduces a transfer learning framework for the autonomous detection, isolation, and quantification of delamination in laminated composites based on scarce low-frequency structural vibration data. Limited response data from an electromechanically coupled simulation model and from experimental testing of laminated composite coupons were encoded into high-resolution time-frequency images using SynchroExtracting Transforms (SETs). The simulated and experimental data were processed through different layers of pretrained deep learning models based on AlexNet, GoogleNet, SqueezeNet, ResNet-18, and VGG-16 to extract low- and high-level autonomous features. The support vector machine (SVM) machine learning algorithm was employed to assess how the identified autonomous features were able to assist in the detection, isolation, and quantification of delamination in laminated composites. The results obtained using these autonomous features were also compared with those obtained using handcrafted statistical features. The obtained results are encouraging and provide a new direction that will allow us to progress in the autonomous damage assessment of laminated composites despite being limited to using raw scarce structural vibration data.

Preexisting comorbidities are associated with the mortality rate as well as the predialysis adverse events in incident dialysis patients.

BACKGROUND: Optimal estimated glomerular filtration rate (eGFR) to start maintenance dialysis is controversial. Observational studies have reported that initiation of dialysis at high eGFRs is associated with worse postdialysis survival. METHODS: We retrospectively investigated 1,038 incident dialysis patients who started maintenance dialysis during 2010-2015. Patients were assessed for comorbidities and adverse events during the transitional period of dialysis initiation. Patients were classified as planned dialysis (PD) vs. unplanned dialysis (UD) according to indications for dialysis initiation. RESULTS: UD group comprised 352 patients (33.9%). Mean eGFR at dialysis initiation was higher in UD patients than PD patients (7.9 ± 5.1 vs. 5.9 ± 3.4 mL/min/1.73 m2, p < 0.001). Mean Davies comorbidity index in the UD group was higher (vs. PD group, 1.3 ± 1.0 vs. 0.9 ± 1.0, p < 0.001). Patients with more comorbidities experienced more ischemic heart disease (hazard ratio [HR], 4.36; 95% confidence interval [CI], 1.71-11.14) in the medium-risk group and HR of 8.84 (95% CI, 3.06-25.55) in the high-risk group (vs. low-risk group, p < 0.001)) during the predialysis period. High-risk group had increased postdialysis mortality (HR, 2.48; 95% CI, 1.46-4.20; p = 0.001). Adjusted HR of mortality was higher in the medium-risk group of UD patients (HR, 1.72; 95% CI, 1.16-2.56; p = 0.007). CONCLUSION: Patients with more comorbidities were at increased risk of predialysis ischemic heart disease and postdialysis mortality. UD patients in the medium-risk population had increased risk of postdialysis mortality. Dialysis start should be individualized by considering comorbidities.

ZnO Nanoparticle/Graphene Hybrid Photodetectors via Laser Fragmentation in Liquid.

By combining the enhanced photosensitive properties of zinc oxide nanoparticles and the excellent transport characteristics of graphene, UV-sensitive, solar-blind hybrid optoelectronic devices have been demonstrated. These hybrid devices offer high responsivity and gain, making them well suited for photodetector applications. Here, we report a hybrid ZnO nanoparticle/graphene phototransistor that exhibits a responsivity up to 4 × 104 AW-1 and gain of up to 1.3 × 105 with high UV wavelength selectivity. ZnO nanoparticles were synthesized by pulsed laser fragmentation in liquid to attain a simple, efficient, ligand-free method for nanoparticle fabrication. By combining simple fabrication processes with a promising device architecture, highly sensitive ZnO nanoparticle/graphene UV photodetectors were successfully demonstrated.

Light tunable plasmonic metasurfaces.

Self-assembled plasmonic metasurfaces are promising optical platforms to achieve accessible flat optics, due to their strong light-matter interaction, nanometer length scale precision, large area, light weight, and high-throughput fabrication. Here, using photothermal continuous wave laser lithography, we show the spectral and spatial tuning of metasurfaces comprised of a monolayer of ligand capped hexagonally packed gold nanospheres. To tune the spectral response of the metasurfaces, we show that by controlling the intensity of a laser focused onto the metasurface that the absorption peak can be reconfigured from the visible to near-infrared wavelength. The irreversible spectral tuning mechanism is attributed to photothermal modification of the surface morphology. Combining self-assembled metasurfaces with laser lithography, we demonstrate an optically thin (λ/42), spectrally selective plasmonic Fresnel zone plate. This work establishes a new pathway for creating flat, large area, frequency selective optical elements using self-assembled plasmonic metasurfaces and laser lithography.

Trabecular bone score may indicate chronic kidney disease-mineral and bone disorder (CKD-MBD) phenotypes in hemodialysis patients: a prospective observational study.

BACKGROUND: In the general population, the trabecular bone score (TBS) represents the bone microarchitecture and predicts fracture risk independent of bone mineral density (BMD). A few studies reported that TBS is significantly reduced in dialysis patients. Chronic kidney disease-mineral and bone disorder (CKD-MBD) are accompanied by increased fracture risk, cardiovascular morbidity, and mortality. We investigated whether TBS is associated with comorbidity related to CKD-MBD or frailty in hemodialysis patients. METHODS: In this prospective observational study, TBS was obtained using the TBS iNsight software program (Med-Imaps) with BMD dual energy x-ray absorptiometry (DXA) images (L1-L4) from prevalent hemodialysis patients. A Tilburg frailty indicator was used to evaluate frailty, and hand grip strength and bio-impedance (InBody) were measured. A patient-generated subjective global assessment (PG-SGA) was used for nutritional assessment. The history of cardiovascular events (CVE) and demographic, clinical, laboratory, and biomarker data were collated. We then followed up patients for the occurrence of CKD-MBD related complications. RESULTS: We enrolled 57 patients in total. The mean age was 56.8 ± 15.9 years (50.9% female). Prevalence of Diabetes mellitus (DM) was 40.4% and CVE was 36.8%. Mean TBS was 1.44 ± 0.10. TBS significantly reduced in the CVE group (1.38 ± 0.08 vs. 1.48 ± 0.10, p <  0.001). Multivariable regression analysis was conducted adjusting for age, sex, dialysis vintage, DM, CVE, albumin, intact parathyroid hormone, fibroblast growth factor 23, handgrip strength, and phosphate binder dose. Age (ß = - 0.030; p = 0.001) and CVE (ß = - 0.055; p = 0.024) were significant predictors of TBS. During the follow up period after TBS measurements (about 20 months), four deaths, seven incident fractures, and six new onset CVE were recorded. Lower TBS was associated with mortality (p = 0.049) or new onset fracture (p = 0.007, by log-rank test). CONCLUSION: Lower TBS was independently associated with increased age and CVE prevalence in hemodialysis patients. Mortality and fracture incidence were significantly higher in patients with lower TBS values. These findings suggest that TBS may indicate a phenotype of frailty and also a CKD-MBD phenotype reciprocal to CVE.