scholarly journals A Comparison of Feature Selection and Forecasting Machine Learning Algorithms for Predicting Glycaemia in Type 1 Diabetes Mellitus

2021 ◽  
Vol 11 (4) ◽  
pp. 1742
Author(s):  
Ignacio Rodríguez-Rodríguez ◽  
José-Víctor Rodríguez ◽  
Wai Lok Woo ◽  
Bo Wei ◽  
Domingo-Javier Pardo-Quiles
Keyword(s):  
Diabetes Mellitus ◽  
Machine Learning ◽  
Type 1 Diabetes ◽  
Feature Selection ◽  
Blood Glucose ◽  
Type 1 Diabetes Mellitus ◽  
Support Vector ◽  
Chronic Hyperglycemia ◽  
Predictive Algorithms

Type 1 diabetes mellitus (DM1) is a metabolic disease derived from falls in pancreatic insulin production resulting in chronic hyperglycemia. DM1 subjects usually have to undertake a number of assessments of blood glucose levels every day, employing capillary glucometers for the monitoring of blood glucose dynamics. In recent years, advances in technology have allowed for the creation of revolutionary biosensors and continuous glucose monitoring (CGM) techniques. This has enabled the monitoring of a subject’s blood glucose level in real time. On the other hand, few attempts have been made to apply machine learning techniques to predicting glycaemia levels, but dealing with a database containing such a high level of variables is problematic. In this sense, to the best of the authors’ knowledge, the issues of proper feature selection (FS)—the stage before applying predictive algorithms—have not been subject to in-depth discussion and comparison in past research when it comes to forecasting glycaemia. Therefore, in order to assess how a proper FS stage could improve the accuracy of the glycaemia forecasted, this work has developed six FS techniques alongside four predictive algorithms, applying them to a full dataset of biomedical features related to glycaemia. These were harvested through a wide-ranging passive monitoring process involving 25 patients with DM1 in practical real-life scenarios. From the obtained results, we affirm that Random Forest (RF) as both predictive algorithm and FS strategy offers the best average performance (Root Median Square Error, RMSE = 18.54 mg/dL) throughout the 12 considered predictive horizons (up to 60 min in steps of 5 min), showing Support Vector Machines (SVM) to have the best accuracy as a forecasting algorithm when considering, in turn, the average of the six FS techniques applied (RMSE = 20.58 mg/dL).

scholarly journals Feature Selection for Blood Glucose Level Prediction in Type 1 Diabetes Mellitus by Using the Sequential Input Selection Algorithm (SISAL)

Symmetry ◽  
2019 ◽  
Vol 11 (9) ◽  
pp. 1164 ◽  
Author(s):  
Rodríguez-Rodríguez ◽  
Rodríguez ◽  
González-Vidal ◽  
Zamora
Keyword(s):  
Diabetes Mellitus ◽  
Type 1 Diabetes ◽  
Feature Selection ◽  
Blood Glucose ◽  
Blood Glucose Level ◽  
Type 1 Diabetes Mellitus ◽  
Glucose Level ◽  
Selection Algorithm ◽  
Input Selection

Feature selection is a primary exercise to tackle any forecasting task. Machine learning algorithms used to predict any variable can improve their performance by lessening their computational effort with a proper dataset. Anticipating future glycemia in type 1 diabetes mellitus (DM1) patients provides a baseline in its management, and in this task, we need to carefully select data, especially now, when novel wearable devices offer more and more information. In this paper, a complete characterization of 25 diabetic people has been carried out, registering innovative variables like sleep, schedule, or heart rate in addition to other well-known ones like insulin, meal, and exercise. With this ground-breaking data compilation, we present a study of these features using the Sequential Input Selection Algorithm (SISAL), which is specially prepared for time series data. The results rank features according to their importance, regarding their relevance in blood glucose level prediction as well as indicating the most influential past values to be taken into account and distinguishing features with person-dependent behavior from others with a common performance in any patient. These ideas can be used as strategies to select data for predicting glycemia depending on the availability of computational power, required speed, or required accuracy. In conclusion, this paper tries to analyze if there exists symmetry among the different features that can affect blood glucose levels, that is, if their behavior is symmetric in terms of influence in glycemia.

scholarly journals A Comparison of Different Models of Glycemia Dynamics for Improved Type 1 Diabetes Mellitus Management with Advanced Intelligent Analysis in an Internet of Things Context

2020 ◽  
Vol 10 (12) ◽  
pp. 4381 ◽  
Author(s):  
Ignacio Rodríguez-Rodríguez ◽  
José-Víctor Rodríguez ◽  
José-María Molina-García-Pardo ◽  
Miguel-Ángel Zamora-Izquierdo ◽  
María-Teresa Martínez-Inglés
Keyword(s):  
Diabetes Mellitus ◽  
Type 1 Diabetes ◽  
Blood Glucose ◽  
Internet Of Things ◽  
Type 1 Diabetes Mellitus ◽  
High Volume ◽  
Daily Basis ◽  
Data Set ◽  
Chronic Hyperglycemia

The metabolic disease Type 1 Diabetes Mellitus (DM1) is caused by a reduction in the production of pancreatic insulin, which causes chronic hyperglycemia. Patients with DM1 are required to perform multiple blood glucose measurements on a daily basis to monitor their blood glucose dynamics through the use of capillary glucometers. In more recent times, technological developments have led to the development of cutting-edge biosensors and Continuous Glucose Monitoring (CGM) systems that can monitor patients’ blood glucose levels on a real-time basis. This offers medical providers access to glucose oscillations modeling interventions that can enhance DM1 treatment and management approaches through the use of novel disruptive technologies, such as Cloud Computing (CC), big data, Intelligent Data Analysis (IDA) and the Internet of Things (IoT). This work applies some advanced modeling techniques to a complete data set of glycemia-related biomedical features—obtained through an extensive, passive monitoring campaign undertaken with 25 DM1 patients under real-world conditions—in order to model glucose level dynamics through the proper identification of patterns. Hereby, four methods, which are run through CC due to the high volume of data collected, are applied and compared within an IoT context. The results show that Bayesian Regularized Neural Networks (BRNN) offer the best performance (0.83 R2) with a reduced Root Median Squared Error (RMSE) of 14.03 mg/dL.

scholarly journals Performance Analysis of Conventional Machine Learning Algorithms for Identification of Chronic Kidney Disease in Type 1 Diabetes Mellitus Patients

Diagnostics ◽  
2021 ◽  
Vol 11 (12) ◽  
pp. 2267
Author(s):  
Nakib Hayat Chowdhury ◽  
Mamun Bin Ibne Reaz ◽  
Fahmida Haque ◽  
Shamim Ahmad ◽  
Sawal Hamid Md Ali ◽  
...  
Keyword(s):  
Diabetes Mellitus ◽  
Machine Learning ◽  
Chronic Kidney Disease ◽  
Type 1 Diabetes ◽  
Kidney Disease ◽  
Type 1 Diabetes Mellitus ◽  
Gradient Boosting ◽  
Support Vector ◽  
Extreme Gradient Boosting

Chronic kidney disease (CKD) is one of the severe side effects of type 1 diabetes mellitus (T1DM). However, the detection and diagnosis of CKD are often delayed because of its asymptomatic nature. In addition, patients often tend to bypass the traditional urine protein (urinary albumin)-based CKD detection test. Even though disease detection using machine learning (ML) is a well-established field of study, it is rarely used to diagnose CKD in T1DM patients. This research aimed to employ and evaluate several ML algorithms to develop models to quickly predict CKD in patients with T1DM using easily available routine checkup data. This study analyzed 16 years of data of 1375 T1DM patients, obtained from the Epidemiology of Diabetes Interventions and Complications (EDIC) clinical trials directed by the National Institute of Diabetes, Digestive, and Kidney Diseases, USA. Three data imputation techniques (RF, KNN, and MICE) and the SMOTETomek resampling technique were used to preprocess the primary dataset. Ten ML algorithms including logistic regression (LR), k-nearest neighbor (KNN), Gaussian naïve Bayes (GNB), support vector machine (SVM), stochastic gradient descent (SGD), decision tree (DT), gradient boosting (GB), random forest (RF), extreme gradient boosting (XGB), and light gradient-boosted machine (LightGBM) were applied to developed prediction models. Each model included 19 demographic, medical history, behavioral, and biochemical features, and every feature’s effect was ranked using three feature ranking techniques (XGB, RF, and Extra Tree). Lastly, each model’s ROC, sensitivity (recall), specificity, accuracy, precision, and F-1 score were estimated to find the best-performing model. The RF classifier model exhibited the best performance with 0.96 (±0.01) accuracy, 0.98 (±0.01) sensitivity, and 0.93 (±0.02) specificity. LightGBM performed second best and was quite close to RF with 0.95 (±0.06) accuracy. In addition to these two models, KNN, SVM, DT, GB, and XGB models also achieved more than 90% accuracy.

scholarly journals Analysis of immune cell components and immune-related gene expression profiles in peripheral blood of patients with type 1 diabetes mellitus

2021 ◽  
Author(s):  
Jian Lin ◽  
Yuanhua Lu ◽  
Bizhou Wang ◽  
Ping Jiao ◽  
Jie Ma
Keyword(s):  
Gene Expression ◽  
Diabetes Mellitus ◽  
Type 1 Diabetes ◽  
Type 1 Diabetes Mellitus ◽  
Nk Cells ◽  
Immune Cells ◽  
Support Vector ◽  
Hub Genes ◽  
Immune Related Genes

Abstract Background Type 1 diabetes mellitus (T1DM) is a chronic autoimmune disease caused by severe loss of pancreatic β cells. Immune cells are key mediators of β cell destruction. This study attempted to investigate the role of immune cells and immune-related genes in the occurrence and development of T1DM. Methods The raw gene expression profile of the samples from 12 T1DM patients and 10 normal controls was obtained from Gene Expression Omnibus (GEO) database. Differentially expressed genes (DEGs) were identified by Limma package in R. The least absolute shrinkage and selection operator (LASSO) - support vector machines (SVM) were used to screen the hub genes. CIBERSORT algorithm was used to identify the different immune cells in distribution between T1DM and normal samples. Correlation of the hub genes and immune cells was analyzed by Spearman, and gene-GO-BP and gene-pathway interaction networks were constructed by Cytoscape plug-in ClueGO. Receiver operating characteristic (ROC) curves were used to assess diagnostic value of genes in T1DM. Results The 50 immune-related DEGs were obtained between the T1DM and normal samples. Then, the 50 immune-related DEGs were further screened to obtain the 5 hub genes. CIBERSORT analysis revealed that the distribution of plasma cells, resting mast cells, resting NK cells and neutrophils had significant difference between T1DM and normal samples. Natural cytotoxicity triggering receptor 3 (NCR3) was significantly related to the activated NK cells, M0 macrophages, monocytes, resting NK cells, and resting memory CD4+ T cells. Moreover, tumor necrosis factor (TNF) was significantly associated with naive B cell and naive CD4+ T cell. NCR3 [Area under curve (AUC) = 0.918] possessed a higher accuracy than TNF (AUC = 0.763) in diagnosis of T1DM. Conclusions The immune-related genes (NCR3 and TNF) and immune cells (NK cells) may play a vital regulatory role in the occurrence and development of T1DM, which possibly provide new ideas and potential targets for the immunotherapy of diabetes mellitus (DM).

Comparative analysis of different drug delivery methods of injectable hydrogel nanomaterials of insulin biomaterials via multiple daily injections and continuous subcutaneous insulin infusion in the treatment of type 1 diabetes mellitus in children

2021 ◽  
Vol 11 (7) ◽  
pp. 1154-1160
Author(s):  
Yan Sun ◽  
Haoshu Niu ◽  
Zhixia Wang ◽  
Ying Wang ◽  
Xuechun Li ◽  
...  
Keyword(s):  
Diabetes Mellitus ◽  
Drug Delivery ◽  
Type 1 Diabetes ◽  
Blood Glucose ◽  
Type 1 Diabetes Mellitus ◽  
Continuous Subcutaneous Insulin Infusion ◽  
Insulin Infusion ◽  
Insulin Dosage ◽  
Subcutaneous Insulin

The aim of this study was to investigate the difference between multiple daily injections (MDI) and continuous subcutaneous insulin infusion (CSII) in blood glucose control during the treatment of type 1 diabetes mellitus (T1DM) in children. under the nano-hydrogel delivery carrier. In order to improve the efficiency and therapeutic effect of the experiment, this paper adopts injectable nanomaterial-polymer composite hydrogel as drug delivery system to cooperate with insulin injection to improve the effective utilization of drugs. Eighty children diagnosed with T1DM by the department of Endocrinology, Genetics, and Metabolism of INNER MONGOLIA BAOGANG Hospital from October 2018 to December 2019 were selected as research subjects for this study. The children were randomly divided into MDI group (treated with MDI) and CSII group (treated with CSII), with 40 children in each group. The basic data of the children were compared, and changes in hemoglobin A1c (HbA1c) at admission and 1, 2, and 3 months after treatment were detected. During the detection, the blood glucose level, therapeutic time of blood glucose normalization, and daily insulin dosage were recorded. The HbA1c and fasting blood glucose (FBG) were followed up three months after discharge, and incidences of hypoglycemia in the two groups were observed. The results showed that the mean value of HbA1c in the MDI group was higher than that in the CSII group (P < 0.05). Each patient was assessed for the number of times their blood sugar was allowed to dip below normal levels; patients with less hypoglycemia had a higher rate of blood sugar control. The control rates of blood glucose in the MDI and CSII groups were 19.21% and 23.50%, respectively. The CSII group showed significantly higher blood glucose rates than the MDI group (P < 0.05). The therapeutic time of blood glucose normalization in the MDI group was significantly longer than that in the CSII group (P < 0.05). There was no significant difference in the average daily insulin dosage between the MDI and CSII groups (P > 0.05), which indicated that CSII therapy had significant advantages in reducing blood glucose in children with T1DM.

Programmed Cell Death-1 Monoclonal Antibody Therapy and Type 1 Diabetes Mellitus: A Review of the Literature

2019 ◽  
pp. 089719001985092 ◽  
Author(s):  
Kyle A. Farina ◽  
Michael P. Kane
Keyword(s):  
Diabetes Mellitus ◽  
Cell Death ◽  
Type 1 Diabetes ◽  
Blood Glucose ◽  
Programmed Cell Death ◽  
Type 1 Diabetes Mellitus ◽  
Inhibitor Therapy ◽  
Blood Glucose Monitoring ◽  
Programmed Cell Death 1

Two Food and Drug Administration-approved programmed cell death-1 (PD-1) inhibitors, nivolumab (Opdivo®), and pembrolizumab (Keytruda®), are indicated for treatment-resistant malignancies. Inhibition of PD-1 also inhibits T-cell peripheral tolerance, enhancing autoimmunity. Various autoimmune conditions have been reported with the use of these agents, including type 1 diabetes mellitus (T1DM). This article reviews literature regarding the development of T1DM in patients treated with PD-1 inhibitors and identifies strategies for the appropriate identification, monitoring, and follow-up of these patients. Published cases of T1DM related to PD-1 inhibitor therapy were identified using PubMed. Eighty-three identified publications were reviewed, of which 37 publications involving 42 cases of anti-PD-1 therapy-induced T1DM were identified. The average age of patients at presentation was 62 years and 59.5% were male. The mean number of PD-1 inhibitor doses received was 5, with a mean time to presentation of 11 weeks. Initial presentation of diabetic ketoacidosis was reported in 69% of cases, with an average blood glucose of 660 mg/dL and an average HbA1cof 8.7%. The exact mechanism PD-1 inhibitor therapy-induced T1DM is unknown. Blood glucose monitoring is recommended for all patients receiving anti-PD-1 therapy. Further research is needed to delineate the frequency of this adverse effect, as well as to evaluate potential risk factors and ideal management strategies.

Acute changes in blood glucose do not alter blood glutathione synthesis in adolescents with poorly controlled type 1 diabetes mellitus

Metabolism ◽  
2012 ◽  
Vol 61 (3) ◽  
pp. 373-378 ◽  
Author(s):  
Dominique Darmaun ◽  
Susan Welch ◽  
Shiela Smith ◽  
Shawn Sweeten ◽  
Nelly Mauras
Keyword(s):  
Diabetes Mellitus ◽  
Type 1 Diabetes ◽  
Blood Glucose ◽  
Type 1 Diabetes Mellitus ◽  
Glutathione Synthesis ◽  
Acute Changes
Sign in / Sign up

Export Citation Format

Share Document