The Use of Big Data in Personalized Healthcare to Reduce Inventory Waste and Optimize Patient Treatment.

Precision medicine is emerging as an integral component in delivering care in the health system leading to better diagnosis and optimizing the treatment of patients. This growth is due to the new technologies in the data science field that have led to the ability to model complex diseases. Precision medicine is based on genomics and omics facilities that provide information about molecular proteins and biomarkers that could lead to discoveries for the treatment of patients suffering from various diseases. However, the main problems related to precision medicine are the ability to analyze, interpret, and integrate data. Hence, there is a lack of smooth transition from conventional to precision medicine. Therefore, this work reviews the limitations and discusses the benefits of overcoming them if big data tools are utilized and merged with precision medicine. The results from this review indicate that most of the literature focuses on the challenges rather than providing flexible solutions to adapt big data to precision medicine. As a result, this paper adds to the literature by proposing potential technical, educational, and infrastructural solutions in big data for a better transition to precision medicine.

Improvement of Dialysis Dosing Using Big Data Analytics.

OBJECTIVES: Large amounts of healthcare data are now generated via patient health records, records of diagnosis and treatment, smart devices, and wearables. Extracting insights from such data can transform healthcare from a traditional, symptom-driven practice into precisely personalized medicine. Dialysis treatments generate a vast amount of data, with more than 100 parameters that must be regulated for ideal treatment outcomes. When complications occur, understanding electrolyte parameters and predicting their outcomes to deliver the optimal dialysis dosing for each patient is a challenge. This study focused on refining dialysis dosing by utilizing emerging data from the growing number of dialysis patients to improve patients' quality of life and well-being. METHODS: Exploratory data analysis and data prediction approaches were performed to gather insights from patients' vital electrolytes on how to improve the patients' dialysis dosing. Four predictive models were constructed to predict electrolyte levels through various dialysis parameters. RESULTS: The decision tree model showed excellent performance and more accurate results than the support vector machine, linear regression, and neural network models. CONCLUSIONS: The predictive models identified that pre-dialysis blood urea nitrogen, pre-weight, dry weight, anticoagulation, and sex had the most significant effects on electrolyte concentrations. Such models could fine-tune dialysis dosing levels for the growing number of dialysis patients to improve each patient's quality of life, life expectancy, and well-being, and to reduce costs, efforts, and time consumption for both patients and physicians. The study's results need to be validated on a larger scale.

Survey for Computer-Aided Tools and Databases in Metabolomics.

Metabolomics has advanced from innovation and functional genomics tools and is currently a basis in the big data-led precision medicine era. Metabolomics is promising in the pharmaceutical field and clinical research. However, due to the complexity and high throughput data generated from such experiments, data mining and analysis are significant challenges for researchers in the field. Therefore, several efforts were made to develop a complete workflow that helps researchers analyze data. This paper introduces a review of the state-of-the-art computer-aided tools and databases in metabolomics established in recent years. The paper provides computational tools and resources based on functionality and accessibility and provides hyperlinks to web pages to download or use. This review aims to present the latest computer-aided tools, databases, and resources to the metabolomics community in one place.

Untargeted approach to investigating the metabolomics profile of type 2 diabetes emiratis.

Type 2 Diabetes (T2D) is expected to be the seventh most significant cause of death worldwide by 2030. Although research into its mechanism has received the attention it deserves, our understanding of T2D is still limited. This case-control study employs untargeted metabolomics to explore novel T2D plasma biomarkers in the Emirati population. Ninety-two UAE nationals were included in the cohort, with fifty T2D and forty-two non-T2D profiles. Participants were then stratified into three groups based on metabolic profiles, clinically verified diabetic status, and current HbA1c values: namely controlled diabetics, uncontrolled diabetics and prediabetics, and non-diabetics. The study identified fifteen significant differentially abundant metabolites between the uncontrolled diabetics group and the prediabetics or controlled diabetics group. Interestingly, some metabolites essential for the corticosteroid and thyroid signaling pathways were found to be significantly elevated in poorly controlled T2D, including cortisol, glycocholic acid, bile acids, thyroxine, and the tryptophan metabolite, 5-hydroxyindoleacetic acid. These findings align with those from prior western cohorts and suggest an intriguing linkage between T2D glycemic control and thyroid and adrenal signaling that may provide new diagnostic and prognostic indicators. RESEARCH SIGNIFICANCE: This study investigates the underlooked metabolomic role and correlation with T2D in the UAE population. The report indicates fifteen significant differentially abundant metabolites between on diabetics, uncontrolled diabetics and or controlled diabetics or prediabetics. This panel of metabolites such as thyroxine and corticosteroids should be considered further as potential diagnostic or prognostic biomarkers for T2D in the region.

Untargeted Metabolomic Plasma Profiling of Emirati Dialysis Patients with Diabetes versus Non-Diabetic: A Pilot Study.

Diabetic kidney disease (DKD) is a severe irreversible complication of diabetes mellitus that further disturbs glucose metabolism. Identifying metabolic changes in the blood may provide early insight into DKD pathogenesis. This study aims to determine blood biomarkers differentiating DKD from non-diabetic kidney disease in the Emirati population utilizing the LC-MS/MS platform. Blood samples were collected from hemodialysis subjects with and without diabetes to detect indicators of pathological changes using an untargeted metabolomics approach. Metabolic profiles were analyzed based on clinically confirmed diabetic status and current HbA1c values. Five differentially significant metabolites were identified based on the clinically confirmed diabetic status, including hydroxyprogesterone and 3,4-Dihydroxymandelic acid. Similarly, we identified seven metabolites with apparent differences between Dialysis Diabetic (DD) and Dialysis non-Diabetic (DND) groups, including isovalerylglycine based on HbA1c values. Likewise, the top three metabolic pathways, including Tyrosine metabolism, were identified following the clinically confirmed diabetic status. As a result, nine different metabolites were enriched in the identified metabolic pathways, such as 3,4-Dihydroxymandelic acid. As a result, eleven different metabolites were enriched, including Glycerol. This study provides an insight into blood metabolic changes related to DKD that may lead to more effective management strategies.

Prediction of COVID-19 Hospital Length of Stay and Risk of Death Using Artificial Intelligence-Based Modeling.

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is a highly infectious virus with overwhelming demand on healthcare systems, which require advanced predictive analytics to strategize COVID-19 management in a more effective and efficient manner. We analyzed clinical data of 2017 COVID-19 cases reported in the Dubai health authority and developed predictive models to predict the patient's length of hospital stay and risk of death. A decision tree (DT) model to predict COVID-19 length of stay was developed based on patient clinical information. The model showed very good performance with a coefficient of determination R 2 of 49.8% and a median absolute deviation of 2.85 days. Furthermore, another DT-based model was constructed to predict COVID-19 risk of death. The model showed excellent performance with sensitivity and specificity of 96.5 and 87.8%, respectively, and overall prediction accuracy of 96%. Further validation using unsupervised learning methods showed similar separation patterns, and a receiver operator characteristic approach suggested stable and robust DT model performance. The results show that a high risk of death of 78.2% is indicated for intubated COVID-19 patients who have not used anticoagulant medications. Fortunately, intubated patients who are using anticoagulant and dexamethasone medications with an international normalized ratio of <1.69 have zero risk of death from COVID-19. In conclusion, we constructed artificial intelligence-based models to accurately predict the length of hospital stay and risk of death in COVID-19 cases. These smart models will arm physicians on the front line to enhance management strategies to save lives.

Sustainable dynamic lot sizing models for cold products under carbon cap policy.

Amid the ever growing interest in operational supply chain models that incorporate environmental aspects as an integral part of the decision making process, this paper addresses the dynamic lot sizing problem of a cold product while accounting for carbon emissions generated during temperature-controlled storage and transportation activities. We present two mixed integer programming models to tackle the two cases where the carbon cap is imposed over the whole planning horizon versus the more stringent version of a cap per period. For the first model, a Lagrangian relaxation approach is proposed which provides a mean for comparing the operational cost and carbon footprint performance of the carbon tax and the carbon cap policies. Subsequently, a Bisection based algorithm is developed to solve the relaxed model and generate the optimal ordering policy. The second model, however, is solved via a dynamic programming based algorithm while respecting two established lower and upper bounds on the periodic carbon cap. The results of the computational experiments for the first model display a stepwise increase (decrease) in the total carbon emissions (operational cost) as the preset cap value is increased. A similar behavior is also observed for the second model with the exception that paradoxical increases in the total emissions are sometimes realized with slightly tighter values of the periodic cap.

IoT Based Predictive Maintenance Management of Medical Equipment.

Technological advancements are the main drivers of the healthcare industry as it has a high impact on delivering the best patient care. Recent years witnessed unprecedented growth in the number of medical equipment manufactured to aid high-quality patient care at a fast pace. With this growth of medical equipment, hospitals need to adopt optimal maintenance strategies that enhance the performance of their equipment and attempt to reduce their maintenance costs and effort. In this work, a Predictive Maintenance (PdM) approach is presented to help in failure diagnosis for critical equipment with various and frequent failure mode(s). The proposed approach relies on the understanding of the physics of failure, real-time collection of the right parameters using the Internet of Things (IoT) technology, and utilization of machine learning tools to predict and classify healthy and faulty equipment status. Moreover, transforming traditional maintenance into PdM has to be supported by an economic analysis to prove the feasibility and efficiency of transformation. The applicability of the approach was demonstrated using a case study from a local hospital in the United Arab Emirates (UAE) where the Vitros-Immunoassay analyzer was selected based on maintenance events and criticality assessment as a good candidate for transforming maintenance from corrective to predictive. The dominant failure mode is metering arm belt slippage due to wear out of belt and movement of pulleys which can be predicted using vibration signals. Vibration real data is collected using wireless accelerometers and transferred to a signal analyzer located on a cloud or local computer. Features extracted and selected are analyzed using Support Vector Machine (SVM) to detect the faulty condition. In terms of economics, the proposed approach proved to provide significant diagnostic and repair cost savings that can reach up to 25% and an investment payback period of one year. The proposed approach is scalable and can be used across medical equipment in large medical centers.