Identifying patients at increased risk of hypoglycaemia in primary care: Development of a machine learning-based screening tool.

INTRODUCTION: In primary care, identifying patients with type 2 diabetes (T2D) who are at increased risk of hypoglycaemia is important for the prevention of hypoglycaemic events. We aimed to develop a screening tool based on machine learning to identify such patients using routinely available demographic and medication data. METHODS: We used a cohort study design and the Groningen Initiative to ANalyse Type 2 diabetes Treatment (GIANTT) medical record database to develop models for hypoglycaemia risk. The first hypoglycaemic event in the observation period (2007-2013) was the outcome. Demographic and medication data were used as predictor variables to train machine learning models. The performance of the models was compared with a model using additional clinical data using fivefold cross validation with the area under the receiver operator characteristic curve (AUC) as a metric. RESULTS: We included 13,876 T2D patients. The best performing model including only demographic and medication data was logistic regression with least absolute shrinkage and selection operator, with an AUC of 0.71. Ten variables were included (odds ratio): male gender (0.997), age (0.990), total drug count (1.012), glucose-lowering drug count (1.039), sulfonylurea use (1.62), insulin use (1.769), pre-mixed insulin use (1.109), insulin count (1.827), insulin duration (1.193), and antidepressant use (1.05). The proposed model obtained a similar performance to the model using additional clinical data. CONCLUSION: Using demographic and medication data, a model for identifying patients at increased risk of hypoglycaemia was developed using machine learning. This model can be used as a tool in primary care to screen for patients with T2D who may need additional attention to prevent or reduce hypoglycaemic events.

Machine-learning-based early prediction of end-stage renal disease in patients with diabetic kidney disease using clinical trials data.

AIM: To predict end-stage renal disease (ESRD) in patients with type 2 diabetes by using machine-learning models with multiple baseline demographic and clinical characteristics. MATERIALS AND METHODS: In total, 11 789 patients with type 2 diabetes and nephropathy from three clinical trials, RENAAL (n = 1513), IDNT (n = 1715) and ALTITUDE (n = 8561), were used in this study. Eighteen baseline demographic and clinical characteristics were used as predictors to train machine-learning models to predict ESRD (doubling of serum creatinine and/or ESRD). We used the area under the receiver operator curve (AUC) to assess the prediction performance of models and compared this with traditional Cox proportional hazard regression and kidney failure risk equation models. RESULTS: The feed forward neural network model predicted ESRD with an AUC of 0.82 (0.76-0.87), 0.81 (0.75-0.86) and 0.84 (0.79-0.90) in the RENAAL, IDNT and ALTITUDE trials, respectively. The feed forward neural network model selected urinary albumin to creatinine ratio, serum albumin, uric acid and serum creatinine as important predictors and obtained a state-of-the-art performance for predicting long-term ESRD. CONCLUSIONS: Despite large inter-patient variability, non-linear machine-learning models can be used to predict long-term ESRD in patients with type 2 diabetes and nephropathy using baseline demographic and clinical characteristics. The proposed method has the potential to create accurate and multiple outcome prediction automated models to identify high-risk patients who could benefit from therapy in clinical practice.

Predicting Deep Hypnotic State From Sleep Brain Rhythms Using Deep Learning: A Data-Repurposing Approach.

BACKGROUND: Brain monitors tracking quantitative brain activities from electroencephalogram (EEG) to predict hypnotic levels have been proposed as a labor-saving alternative to behavioral assessments. Expensive clinical trials are required to validate any newly developed processed EEG monitor for every drug and combinations of drugs due to drug-specific EEG patterns. There is a need for an alternative, efficient, and economical method. METHODS: Using deep learning algorithms, we developed a novel data-repurposing framework to predict hypnotic levels from sleep brain rhythms. We used an online large sleep data set (5723 clinical EEGs) for training the deep learning algorithm and a clinical trial hypnotic data set (30 EEGs) for testing during dexmedetomidine infusion. Model performance was evaluated using accuracy and the area under the receiver operator characteristic curve (AUC). RESULTS: The deep learning model (a combination of a convolutional neural network and long short-term memory units) trained on sleep EEG predicted deep hypnotic level with an accuracy (95% confidence interval [CI]) = 81 (79.2-88.3)%, AUC (95% CI) = 0.89 (0.82-0.94) using dexmedetomidine as a prototype drug. We also demonstrate that EEG patterns during dexmedetomidine-induced deep hypnotic level are homologous to nonrapid eye movement stage 3 EEG sleep. CONCLUSIONS: We propose a novel method to develop hypnotic level monitors using large sleep EEG data, deep learning, and a data-repurposing approach, and for optimizing such a system for monitoring any given individual. We provide a novel data-repurposing framework to predict hypnosis levels using sleep EEG, eliminating the need for new clinical trials to develop hypnosis level monitors.

Kidney Age Index (KAI): A novel age-related biomarker to estimate kidney function in patients with diabetic kidney disease using machine learning.

BACKGROUND AND OBJECTIVE: With aging, patients with diabetic kidney disease (DKD) show progressive decrease in kidney function. We investigated whether the deviation of biological age (BA) from the chronological age (CA) due to DKD can be used (denoted as Kidney Age Index; KAI) to quantify kidney function using machine learning algorithms. METHODS: Three large datasets were used in this study to develop KAI. The machine learning algorithms were trained on PREVEND dataset with healthy subjects (N = 7963) using 13 clinical markers to predict the CA. The trained model was then used to predict the BA of patients with DKD using RENAAL (N = 1451) and IDNT (N = 1706). The performance of four traditional machine learning algorithms were evaluated and the KAI = BA-CA was estimated for each patient. RESULTS: The neural network model achieved the best performance and predicted the CA of healthy subjects in PREVEND dataset with a mean absolute deviation (MAD) = 6.5 ± 3.5 years and pearson correlation = 0.62. Patients with DKD showed a significant higher KAI of 15.4 ± 11.8 years and 13.6 ± 12.3 years in RENAAL and IDNT datasets, respectively. CONCLUSIONS: Our findings suggest that for a given CA, patients with DKD shows excess BA when compared to their healthy counterparts due to disease severity. With further improvement, the proposed KAI can be used as a complementary easy-to-interpret tool to give a more inclusive idea into disease state.

A Classification Approach for Cancer Survivors from Those Cancer-Free, Based on Health Behaviors: Analysis of the Lifelines Cohort.

Health behaviors affect health status in cancer survivors. We hypothesized that nonlinear algorithms would identify distinct key health behaviors compared to a linear algorithm and better classify cancer survivors. We aimed to use three nonlinear algorithms to identify such key health behaviors and compare their performances with that of a logistic regression for distinguishing cancer survivors from those without cancer in a population-based cohort study. We used six health behaviors and three socioeconomic factors for analysis. Participants from the Lifelines population-based cohort were binary classified into a cancer-survivors group and a cancer-free group using either nonlinear algorithms or logistic regression, and their performances were compared by the area under the curve (AUC). In addition, we performed case-control analyses (matched by age, sex, and education level) to evaluate classification performance only by health behaviors. Data were collected for 107,624 cancer free participants and 2760 cancer survivors. Using all variables resulted an AUC of 0.75 ± 0.01, using only six health behaviors, the logistic regression and nonlinear algorithms differentiated cancer survivors from cancer-free participants with AUCs of 0.62 ± 0.01 and 0.60 ± 0.01, respectively. The main distinctive classifier was age. Though not relevant to classification, the main distinctive health behaviors were body mass index and alcohol consumption. In the case-control analyses, algorithms produced AUCs of 0.52 ± 0.01. No key health behaviors were identified by linear and nonlinear algorithms to differentiate cancer survivors from cancer-free participants in this population-based cohort.