Predicting outcomes of continuous renal replacement therapy using body composition monitoring: a deep-learning approach.

Fluid balance is a critical prognostic factor for patients with severe acute kidney injury (AKI) requiring continuous renal replacement therapy (CRRT). This study evaluated whether repeated fluid balance monitoring could improve prognosis in this clinical population. This was a multicenter retrospective study that included 784 patients (mean age, 67.8 years; males, 66.4%) with severe AKI requiring CRRT during 2017-2019 who were treated in eight tertiary hospitals in Korea. Sequential changes in total body water were compared between patients who died (event group) and those who survived (control group) using mixed-effects linear regression analyses. The performance of various machine learning methods, including recurrent neural networks, was compared to that of existing prognostic clinical scores. After adjusting for confounding factors, a marginal benefit of fluid balance was identified for the control group compared to that for the event group (p = 0.074). The deep-learning model using a recurrent neural network with an autoencoder and including fluid balance monitoring provided the best differentiation between the groups (area under the curve, 0.793) compared to 0.604 and 0.606 for SOFA and APACHE II scores, respectively. Our prognostic, deep-learning model underlines the importance of fluid balance monitoring for prognosis assessment among patients receiving CRRT.

Tackling the Challenges in Scene Graph Generation With Local-to-Global Interactions.

In this work, we seek new insights into the underlying challenges of the scene graph generation (SGG) task. Quantitative and qualitative analysis of the visual genome (VG) dataset implies: 1) ambiguity: even if interobject relationship contains the same object (or predicate), they may not be visually or semantically similar; 2) asymmetry: despite the nature of the relationship that embodied the direction, it was not well addressed in previous studies; and 3) higher-order contexts: leveraging the identities of certain graph elements can help generate accurate scene graphs. Motivated by the analysis, we design a novel SGG framework, Local-to-global interaction networks (LOGINs). Locally, interactions extract the essence between three instances of subject, object, and background, while baking direction awareness into the network by explicitly constraining the input order of subject and object. Globally, interactions encode the contexts between every graph component (i.e., nodes and edges). Finally, Attract and Repel loss is utilized to fine-tune the distribution of predicate embeddings. By design, our framework enables predicting the scene graph in a bottom-up manner, leveraging the possible complementariness. To quantify how much LOGIN is aware of relational direction, a new diagnostic task called Bidirectional Relationship Classification (BRC) is also proposed. Experimental results demonstrate that LOGIN can successfully distinguish relational direction than existing methods (in BRC task), while showing state-of-the-art results on the VG benchmark (in SGG task).

Recalibration and validation of the Charlson Comorbidity Index in an Asian population: the National Health Insurance Service-National Sample Cohort study.

Weights assigned to comorbidities in predicting mortality may vary based on the type of index disease and advances in the management of comorbidities. We aimed to develop a modified version of the Charlson Comorbidity Index (CCI) using an Asian nationwide database (mCCI-A), enabling the precise prediction of mortality rates in this population. The main data source used in this study was the National Health Insurance Service-National Sample Cohort (NHIS-NSC) obtained from the National Health Insurance database, which includes health insurance claims filed between January 1, 2002, and December 31, 2013, in Korea. Of the 1,025,340 individuals included in the NHIS-NSC, 570,716 patients who were hospitalized at least once were analyzed in this study. In total, 399,502 patients, accounting for 70% of the cohort, were assigned to the development cohort, and the remaining patients (n = 171,214) were assigned to the validation cohort. The mCCI-A scores were calculated by summing the weights assigned to individual comorbidities according to their relative prognostic significance determined by a multivariate Cox proportional hazard model. The modified index was validated in the same cohort. The Cox proportional hazard model provided reassigned severity weights for 17 comorbidities that significantly predicted mortality. Both the CCI and mCCI-A were correlated with mortality. However, compared with the CCI, the mCCI-A showed modest but significant increases in the c statistics. According to the analyses using continuous net reclassification improvement, the mCCI-A improved the net mortality risk reclassification by 44.0% (95% confidence intervals (CI), 41.6-46.5; p < 0.001). The mCCI-A facilitates better risk stratification of mortality rates in Korean inpatients than the CCI, suggesting that the mCCI-A may be a preferable index for use in clinical practice and statistical analyses in epidemiological studies.

Prediction of the Mortality Risk in Peritoneal Dialysis Patients using Machine Learning Models: A Nation-wide Prospective Cohort in Korea.

Herein, we aim to assess mortality risk prediction in peritoneal dialysis patients using machine-learning algorithms for proper prognosis prediction. A total of 1,730 peritoneal dialysis patients in the CRC for ESRD prospective cohort from 2008 to 2014 were enrolled in this study. Classification algorithms were used for prediction of N-year mortality including neural network. The survival hazard ratio was presented by machine-learning algorithms using survival statistics and was compared to conventional algorithms. A survival-tree algorithm presented the most accurate prediction model and outperformed a conventional method such as Cox regression (concordance index 0.769 vs 0.745). Among various survival decision-tree models, the modified Charlson Comorbidity index (mCCI) was selected as the best predictor of mortality. If peritoneal dialysis patients with high mCCI (>4) were aged ≥70.5 years old, the survival hazard ratio was predicted as 4.61 compared to the overall study population. Among the various algorithm using longitudinal data, the AUC value of logistic regression was augmented at 0.804. In addition, the deep neural network significantly improved performance to 0.841. We propose machine learning-based final model, mCCI and age were interrelated as notable risk factors for mortality in Korean peritoneal dialysis patients.

A Machine Learning Approach Using Survival Statistics to Predict Graft Survival in Kidney Transplant Recipients: A Multicenter Cohort Study.

Accurate prediction of graft survival after kidney transplant is limited by the complexity and heterogeneity of risk factors influencing allograft survival. In this study, we applied machine learning methods, in combination with survival statistics, to build new prediction models of graft survival that included immunological factors, as well as known recipient and donor variables. Graft survival was estimated from a retrospective analysis of the data from a multicenter cohort of 3,117 kidney transplant recipients. We evaluated the predictive power of ensemble learning algorithms (survival decision tree, bagging, random forest, and ridge and lasso) and compared outcomes to those of conventional models (decision tree and Cox regression). Using a conventional decision tree model, the 3-month serum creatinine level post-transplant (cut-off, 1.65 mg/dl) predicted a graft failure rate of 77.8% (index of concordance, 0.71). Using a survival decision tree model increased the index of concordance to 0.80, with the episode of acute rejection during the first year post-transplant being associated with a 4.27-fold increase in the risk of graft failure. Our study revealed that early acute rejection in the first year is associated with a substantially increased risk of graft failure. Machine learning methods may provide versatile and feasible tools for forecasting graft survival.

Superior outcomes of kidney transplantation compared with dialysis: An optimal matched analysis of a national population-based cohort study between 2005 and 2008 in Korea.

Data regarding kidney transplantation (KT) and dialysis outcomes are rare in Asian populations. In the present study, we evaluated the clinical outcomes associated with KT using claims data from the Korean national public health insurance program. Among the 35,418 adult patients with incident dialysis treated between 2005 and 2008 in Korea, 1539 underwent KT. An optimal balanced risk set matching was attempted to compare the transplant group with the control group in terms of the overall survival and major adverse cardiac event-free survival. Before matching, the dialysis group was older and had more comorbidities. After matching, there were no differences in age, sex, dialysis modalities, or comorbidities. Patient survival was significantly better in the transplant group than in the matched control group (P < 0.001). In addition, the transplant group showed better major adverse cardiac event-free survival than the dialysis group (P < 0.001; hazard ratio, 0.49; 95% confidence interval, 0.32-0.75). Korean patients with incident dialysis who underwent long-term dialysis had significantly more cardiovascular events and higher all-cause mortality rates than those who underwent KT. Thus, KT should be more actively recommended in Korean populations.