Nondimensional diabetes indices for accurate diagnosis of diabetic subjects.

OBJECTIVES: Nondimensional indices or numbers can provide a generalized approach for integrating several biological parameters into one Nondimensional Physiological Index (NDPI) that can help characterize an abnormal state associated with a particular physiological system. In this paper, we have presented four Nondimensional Physiological Indices (NDI, DBI, DIN, CGMDI) for the accurate detection of diabetes subjects. METHODOLOGY: The NDI, DBI, and DIN diabetes indices are based on the Glucose-Insulin Regulatory System (GIRS) Model, represented by the governing differential equation of blood glucose concentration response to the glucose input rate. The solutions of this governing differential equation are employed to simulate the clinical data of the Oral Glucose Tolerance Test (OGTT), and thereby evaluate the GIRS model-system parameters, which are distinctly different for the normal and diabetic subjects. Then these GIRS model parameters are combined to form singular nondimensional indices: NDI, DBI, and DIN. When these indices are applied to the OGTT clinical data, we get significantly different values for normal and diabetic subjects. The DIN diabetes index is a more objective index involving extensive clinical studies, incorporating the GIRS model parameters as well as some key clinical-data markers (based on the information gained from the model clinical simulation and parametric identification). We have then developed another CGMDI diabetes index based on the GIRS model, for the assessment of diabetic subjects using the glucose levels measured by wearable continuous glucose monitoring (CGM) devices. CLINICAL STUDY AND RESULTS: For the DIN diabetes index, our clinical study comprised of 47 subjects (26 normal and 21 diabetics). After applying DIN to the OGTT data, a Distribution Plot of DIN was developed, displaying the ranges of DIN for (i) normal (i.e., non-diabetic) subjects with no risk of becoming diabetic, (ii) normal subjects at risk of becoming diabetic, (iii) borderline diabetic subjects who can become normal (with diet control and treatment), and (iv) distinctly diabetic subjects. This distribution plot is shown to distinctly separate normal subjects from diabetic subjects and also from subjects at risk of becoming diabetic. CONCLUSIONS: In this paper, we have developed several NDPIs in the form of novel nondimensional diabetes indices for the accurate detection of diabetes and diagnosis of diabetic subjects. These nondimensional diabetes indices can enable precision medical diagnostics of diabetes, and thereby also help to develop interventional guidelines for lowering glucose levels by means of insulin infusion. The novelty of our proposed CGMDI is that it utilizes the glucose value monitored by the CGM wearable device. In the future, an app can be developed to use the CGM data in the CGMDI to enable precision diabetes detection.

L-Tetrolet Pattern-Based Sleep Stage Classification Model Using Balanced EEG Datasets.

BACKGROUND: Sleep stage classification is a crucial process for the diagnosis of sleep or sleep-related diseases. Currently, this process is based on manual electroencephalogram (EEG) analysis, which is resource-intensive and error-prone. Various machine learning models have been recommended to standardize and automate the analysis process to address these problems. MATERIALS AND METHODS: The well-known cyclic alternating pattern (CAP) sleep dataset is used to train and test an L-tetrolet pattern-based sleep stage classification model in this research. By using this dataset, the following three cases are created, and they are: Insomnia, Normal, and Fused cases. For each of these cases, the machine learning model is tasked with identifying six sleep stages. The model is structured in terms of feature generation, feature selection, and classification. Feature generation is established with a new L-tetrolet (Tetris letter) function and multiple pooling decomposition for level creation. We fuse ReliefF and iterative neighborhood component analysis (INCA) feature selection using a threshold value. The hybrid and iterative feature selectors are named threshold selection-based ReliefF and INCA (TSRFINCA). The selected features are classified using a cubic support vector machine. RESULTS: The presented L-tetrolet pattern and TSRFINCA-based sleep stage classification model yield 95.43%, 91.05%, and 92.31% accuracies for Insomnia, Normal dataset, and Fused cases, respectively. CONCLUSION: The recommended L-tetrolet pattern and TSRFINCA-based model push the envelope of current knowledge engineering by accurately classifying sleep stages even in the presence of sleep disorders.