Sentiments prediction and thematic analysis for diabetes mobile apps using Embedded Deep Neural Networks and Latent Dirichlet Allocation.

The increasing reliance on mobile health for managing disease conditions has opened a new frontier in digital health, thus, the need for understanding what constitutes positive and negative sentiments of the various apps. This paper relies on Embedded Deep Neural Networks (E-DNN), Kmeans, and Latent Dirichlet Allocation (LDA) for predicting the sentiments of diabetes mobile apps users and identifying the themes and sub-themes of positive and negative sentimental users. A total of 38,640 comments from 39 diabetes mobile apps obtained from the google play store are analyzed and accuracy of 87.67 % ± 2.57 % was obtained from a 10-fold leave-one-out cross-validation. This accuracy is 2.95 % - 18.71 % better than other predominant algorithms used for sentiment analysis and 3.47 % - 20.17 % better than the results obtained by previous researchers. The study also identified the challenges of diabetes mobile apps usage to include safety and security issues, outdated information for diabetes management, clumsy user interface, and difficulty controlling operations. The positives of the apps are ease of operation, lifestyle management, effectiveness in communication and control, and data management capabilities.

Automatic user sentiments extraction from diabetes mobile apps - An evaluation of reviews with machine learning.

Using diabetes mobile apps for self-management of diabetes is one of the emerging strategies for controlling blood sugar levels and maintaining the wellness of patients with diabetes. This study aims to develop a strategy for thematically extracting user comments from diabetes mobile apps to understand the concern of patients with diabetes. Hence, 2678 user comments obtained from the Google Play Store are thematically analyzed with Non-negative Matrix Factorization (NMF) to identify the themes for describing positive, neutral, and negative sentiments. These themes are used as the ground truth for developing a 10-fold cross-validation ensemble Multilayer Artificial Neural Network (ANN) model following the Bag of Word (BOW) analysis of lemmatized user comments. The result shows that a total of 41.24% of positive sentimental users identified the diabetes mobile apps as Effective for Blood Sugar Monitoring (EBSM), 32.36% with neutral sentiments are mostly impressed by the Information Quality (IQ), whereas 40.81% of unhappy users are worried about the Poor Information Quality (PIQ). The prediction accuracy of the ANN model is 89%-97%, which is 5%-48% better than other predominant algorithms. It can be concluded from this study that diabetes mobile apps with a simple user interface, effective data storage and security, medication adherence, and doctor appointment scheduling are preferred by patients with diabetes.

Crowdsourcing Operational and Design Flaws of Diabetes Mobile Apps.

Unfortunately, many of the diabetes mobile apps have operational and design flaws that are debarring users from maximizing from the self-management paradigm. We, therefore, aim to identify the markers of operational and design flaws of diabetes mobile apps to facilitate a better user-centred design. e crowdsourced negative user review comments (rating score: 1-3) of 47 diabetes mobile apps from the google play store. A total of 781 negative user comments (rating score 1-3) from the apps are coded to identify and categorize the themes relating to the operational and design flaws. The operational and design flaws account for 50.32% of the challenges faced by the unhappy diabetes mobile apps users. Among them, 44.73% have issues with app crashing, 17.3% are concerned about device compatibility that inhibits seamless operations, 9.67% are worried about the problem of data uploading. Poor design is a worry to 19.29% of the users who complain of the crowded user interface, poor data management, poor analytics, difficulty scheduling doctors' appointments, and transferring data. More patients with diabetes can be encouraged to continue using diabetes mobile apps for self-management of diabetes through improved design and a pace-wise software advancement to match the ever-growing enhancements in android operating systems and telecommunication devices. This will help to counter most of the challenges identified in this study.

Preadmission assessment of extended length of hospital stay with RFECV-ETC and hospital-specific data.

BACKGROUND: Patients who exceed their expected length of stay in the hospital come at a cost to stakeholders in the healthcare sector as bed spaces are limited for new patients, nosocomial infections increase and the outcome for many patients is hampered due to multimorbidity after hospitalization. OBJECTIVES: This paper develops a technique for predicting Extended Length of Hospital Stay (ELOHS) at preadmission and their risk factors using hospital data. METHODS: A total of 91,468 records of patient's hospital information from a private acute teaching hospital were used for developing a machine learning algorithm relaying on Recursive Feature Elimination with Cross-Validation and Extra Tree Classifier (RFECV-ETC). The study implemented Synthetic Minority Oversampling Technique (SMOTE) and tenfold cross-validation to determine the optimal features for predicting ELOHS while relying on multivariate Logistic Regression (LR) for computing the risk factors and the Relative Risk (RR) of ELOHS at a 95% confidence level. RESULTS: An estimated 11.54% of the patients have ELOHS, which increases with patient age as patients < 18 years, 18-40 years, 40-65 years and ≥ 65 years, respectively, have 2.57%, 4.33%, 8.1%, and 15.18% ELOHS rates. The RFECV-ETC algorithm predicted preadmission ELOHS to an accuracy of 89.3%. Age is a predominant risk factors of ELOHS with patients who are > 90 years-PAG (> 90) {RR: 1.85 (1.34-2.56), P: < 0.001} having 6.23% and 23.3%, respectively, higher likelihood of ELOHS than patient 80-90 years old-PAG (80-90) {RR: 1.74 (1.34-2.38), P: < 0.001} and those 70-80 years old-PAG (70-80) {RR: 1.5 (1.1-2.05), P: 0.011}. Those from admission category-ADC (US1) {RR: 3.64 (3.09-4.28, P: < 0.001} are 14.8% and 70.5%, respectively, more prone to ELOHS compared to ADC (UC1) {RR: 3.17 (2.82-3.55), P: < 0.001} and ADC (EMG) {RR: 2.11 (1.93-2.31), P: < 0.001}. Patients from SES (low) {RR: 1.45 (1.24-1.71), P: < 0.001)} are 13.3% and 45% more susceptible to those from SES (middle) and SES (high). Admission type (ADT) such as AS2, M2, NEWS, S2 and others {RR: 1.37-2.77 (1.25-6.19), P: < 0.001} also have a high likelihood of contributing to ELOHS while the distance to hospital (DTH) {RR: 0.64-0.75 (0.56-0.82), P: < 0.001}, Charlson Score (CCI) {RR: 0.31-0.68 (0.22-0.99), P: < 0.001-0.043} and some VMO specialties {RR: 0.08-0.69 (0.03-0.98), P: < 0.001-0.035} have limited influence on ELOHS. CONCLUSIONS: Relying on the preadmission assessment of ELOHS helps identify those patients who are susceptible to exceeding their expected length of stay on admission, thus, making it possible to improve patients' management and outcomes.

A hybrid approach for risk stratification and predictive modelling of 30-days unplanned readmission of comorbid patients with diabetes.

OBJECTIVES: When comorbid patients with diabetes have 30-days Unplanned Readmission (URA), they attract more burdens to the healthcare system due to increased cost of treatment, insurance penalties to hospitals, and unavailable bed spaces for new patients. This paper, therefore, aims to develop a risk stratification and a predictive model for identifying patients at various risk severities of 30-days URA. METHODS: Patients records of comorbid patients with diabetes treated with different medications were collected from different hospitals and analysed with Principal Component Analysis (PCA) and Multivariate Logistic Regression (MLR) to determine the probability of 30-days URA, which is classified into very low, low, moderate, high, and very high. The risk classes are later modelled using ANOVA feature selection to identify the optimal predictors and the best random forest (RF) hyperparameters for 30-days URA risk stratification. Synthetic Minority Oversampling Technique (SMOTE) was used to balance the risk classes while employing a10-fold cross-validation. RESULTS: After analysing 17,933 episodes of comorbid diabetes patients' treatment, 10.71% are identified to have 30-days URA with 61.95% of patients at moderate risk, 35.5% at low risk, 2.25% at very low risk, 0.37% at high risk, and 0.08% at very high risk. The predictive accuracy of RF is: - recall: 0.947 ± 0.035, precision: 0.951 ± 0.033, F1-score: 0.947 ± 0.035, AUC: 0.994 ± 0.007 and Average Precision (AP) of 0.99. The predictive accuracies of the risk classes measured with F1-score are: - very low: 0.985 ± 0.019, low risk: 0.871 ± 0.079, moderate: 0.881 ± 0.093, high: 0.999 ± 0.003, and very high: 1.000 ± 0.00. CONCLUSION: This study identified the risk severity of comorbid patients with diabetes treated with different medications, making it easier to identify those that will be prioritized on hospitalization to minimize 30-days URA. By relying on the technique developed, vulnerable patients to 30-days URA can be given better post-discharge monitoring to build critical self-management skills that will minimize the cost of diabetes care and improve the quality of life.

Intelligent decision support with machine learning for efficient management of mechanical ventilation in the intensive care unit - A critical overview.

BACKGROUND: Effective management of Mechanical Ventilation (MV) is vital for reducing morbidity, mortality, and cost of healthcare. OBJECTIVE: This study aims to synthesize evidence for effective MV management through Intelligent decision support (IDS) with Machine Learning (ML). METHOD: Databases that include EBSCO, IEEEXplore, Google Scholar, SCOPUS, and the Web of Science were systematically searched to identify studies on IDS for effective MV management regarding Tidal Volume (TV), asynchrony, weaning, and other outcomes such as the risk of Prolonged Mechanical ventilation (PMV). The quality of the articles identified was assessed with a modified Joanna Briggs Institute (JBI) critical appraisal checklist for cross-sessional research. RESULTS: A total of 26 articles were identified for the study that has IDS for TV (n = 2, 7.8 %), asynchrony (n = 9, 34.6 %), weaning (n = 12, 46.2 %), and others (n = 3, 11.5 %). It was affirmed that implementing IDS in MV management will enhance seamless ICU patient management following the utilization of various Machine Learning (ML) algorithms in decision support. The studies relied on (n = 14) ML algorithms to predict the TV, asynchrony, weaning, risk of PMV and Positive End-Expiratory Pressure (PEEP) changes of 11-20262 ICU patients records with model inputs ranging from (n = 1) for timeseries analysis of TV to (n = 47) for weaning prediction. CONCLUSIONS: The small data size, poor study design, and result reporting, with the heterogeneity of techniques used in the various studies, hampered the development of a unified approach for managing MV efficiency in TV monitoring, asynchrony, and weaning predictions. Notwithstanding, the ensemble model was able to predict TV, asynchrony, and weaning to a higher accuracy than the other algorithms.

Intelligent therapeutic decision support for 30 days readmission of diabetic patients with different comorbidities.

The significance of medication therapy in managing comorbid diabetes is vital for maintaining the overall wellness of patients and reducing the cost of healthcare. Thus, using appropriate medication or medication combinations will be necessary for improved person-centred care and reduce complications associated with diagnosis and treatment. This study explains an intelligent decision support framework for managing 30 days unplanned readmission (30_URD) of comorbid diabetes using the Random Forest (RF) algorithm and Bayesian Network (BN) model. After the analysis of the medical records of 101,756 de-identified diabetic patients treated with 21 medications for 28 comorbidity combinations, the optimal medications for minimizing the likelihood of early readmissions were determined. This approach can help for identifying and managing most vulnerable patients thereby giving room to enhance post-discharge monitoring through clinical specialist supports to build critical-self management skills that will minimize the cost of diabetes care.