Data mining technologies for blood glucose and diabetes management

Dynamic circadian fluctuations of glycemia in patients with type 2 diabetes mellitus

BACKGROUND

Diabetes mellitus (DM) has glucose variability that is of such relevance that the appearance of vascular complications in patients with DM has been attributed to hyperglycemic and dysglycemic events. It is known that T1D patients mainly have glycemic variability with a specific oscillatory pattern with specific circadian characteristics for each patient. However, it has not yet been determined whether an oscillation pattern represents the variability of glycemic in T2D. This is why our objective is to determine the characteristics of glycemic oscillations in T2D and generate a robust predictive model.

RESULTS

Showed that glycosylated hemoglobin, glycemia, and body mass index were all higher in patients with T2D than in controls (all p < 0.05). In addition, time in hyperglycemia and euglycemia was markedly higher and lower in the T2D group (p < 0.05), without significant differences for time in hypoglycemia. Standard deviation, coefficient of variation, and total power of glycemia were significantly higher in the T2D group than Control group (all p < 0.05). The oscillatory patterns were significantly different between groups (p = 0.032): the control group was mainly distributed at 2-3 and 6 days, whereas the T2D group showed a more homogeneous distribution across 2-3-to-6 days.

CONCLUSIONS

The predictive model of glycemia showed that it is possible to accurately predict hyper- and hypoglycemia events. Thus, T2D patients exhibit specific oscillatory patterns of glycemic control, which are possible to predict. These findings may help to improve the treatment of DM by considering the individual oscillatory patterns of patients.

Oscillatory pattern of glycemic control in patients with diabetes mellitus

Daily glucose variability is higher in diabetic mellitus (DM) patients which has been related to the severity of the disease. However, it is unclear whether glycemic variability displays a specific pattern oscillation or if it is completely random. Thus, to determine glycemic variability pattern, we measured and analyzed continuous glucose monitoring (CGM) data, in control subjects and patients with DM type-1 (T1D). CGM data was assessed for 6 days (day: 08:00-20:00-h; and night: 20:00-08:00-h). Participants (n = 172; age = 18-80 years) were assigned to T1D (n = 144, females = 65) and Control (i.e., healthy; n = 28, females = 22) groups. Anthropometry, pharmacologic treatments, glycosylated hemoglobin (HbA1c) and years of evolution were determined. T1D females displayed a higher glycemia at 10:00-14:00-h vs. T1D males and Control females. DM patients displays mainly stationary oscillations (deterministic), with circadian rhythm characteristics. The glycemia oscillated between 2 and 6 days. The predictive model of glycemia showed that it is possible to predict hyper and hypoglycemia (R² = 0.94 and 0.98, respectively) in DM patients independent of their etiology. Our data showed that glycemic variability had a specific oscillation pattern with circadian characteristics, with episodes of hypoglycemia and hyperglycemia at day phases, which could help therapeutic action for this population.

Toward Detecting Infection Incidence in People With Type 1 Diabetes Using Self-Recorded Data (Part 1): A Novel Framework for a Personalized Digital Infectious Disease Detection System

BACKGROUND

Type 1 diabetes is a chronic condition of blood glucose metabolic disorder caused by a lack of insulin secretion from pancreas cells. In people with type 1 diabetes, hyperglycemia often occurs upon infection incidences. Despite the fact that patients increasingly gather data about themselves, there are no solid findings that uncover the effect of infection incidences on key parameters of blood glucose dynamics to support the effort toward developing a digital infectious disease detection system.

OBJECTIVE

The study aims to retrospectively analyze the effect of infection incidence and pinpoint optimal parameters that can effectively be used as input variables for developing an infection detection algorithm and to provide a general framework regarding how a digital infectious disease detection system can be designed and developed using self-recorded data from people with type 1 diabetes as a secondary source of information.

METHODS

We retrospectively analyzed high precision self-recorded data of 10 patient-years captured within the longitudinal records of three people with type 1 diabetes. Obtaining such a rich and large data set from a large number of participants is extremely expensive and difficult to acquire, if not impossible. The data set incorporates blood glucose, insulin, carbohydrate, and self-reported events of infections. We investigated the temporal evolution and probability distribution of the key blood glucose parameters within a specified timeframe (weekly, daily, and hourly).

RESULTS

Our analysis demonstrated that upon infection incidence, there is a dramatic shift in the operating point of the individual blood glucose dynamics in all the timeframes (weekly, daily, and hourly), which clearly violates the usual norm of blood glucose dynamics. During regular or normal situations, higher insulin and reduced carbohydrate intake usually results in lower blood glucose levels. However, in all infection cases as opposed to the regular or normal days, blood glucose levels were elevated for a prolonged period despite higher insulin and reduced carbohydrates intake. For instance, compared with the preinfection and postinfection weeks, on average, blood glucose levels were elevated by 6.1% and 16%, insulin (bolus) was increased by 42% and 39.3%, and carbohydrate consumption was reduced by 19% and 28.1%, respectively.

CONCLUSIONS

We presented the effect of infection incidence on key parameters of blood glucose dynamics along with the necessary framework to exploit the information for realizing a digital infectious disease detection system. The results demonstrated that compared with regular or normal days, infection incidence substantially alters the norm of blood glucose dynamics, which are quite significant changes that could possibly be detected through personalized modeling, for example, prediction models and anomaly detection algorithms. Generally, we foresee that these findings can benefit the efforts toward building next generation digital infectious disease detection systems and provoke further thoughts in this challenging field.

A Reinforcement Learning-Based Method for Management of Type 1 Diabetes: Exploratory Study

Background

Type 1 diabetes mellitus (T1DM) is characterized by chronic insulin deficiency and consequent hyperglycemia. Patients with T1DM require long-term exogenous insulin therapy to regulate blood glucose levels and prevent the long-term complications of the disease. Currently, there are no effective algorithms that consider the unique characteristics of T1DM patients to automatically recommend personalized insulin dosage levels.

Objective

The objective of this study was to develop and validate a general reinforcement learning (RL) framework for the personalized treatment of T1DM using clinical data.

Methods

This research presents a model-free data-driven RL algorithm, namely Q-learning, that recommends insulin doses to regulate the blood glucose level of a T1DM patient, considering his or her state defined by glycated hemoglobin (HbA_1c) levels, body mass index, engagement in physical activity, and alcohol usage. In this approach, the RL agent identifies the different states of the patient by exploring the patient’s responses when he or she is subjected to varying insulin doses. On the basis of the result of a treatment action at time step t, the RL agent receives a numeric reward, positive or negative. The reward is calculated as a function of the difference between the actual blood glucose level achieved in response to the insulin dose and the targeted HbA_1c level. The RL agent was trained on 10 years of clinical data of patients treated at the Mass General Hospital.

Results

A total of 87 patients were included in the training set. The mean age of these patients was 53 years, 59% (51/87) were male, 86% (75/87) were white, and 47% (41/87) were married. The performance of the RL agent was evaluated on 60 test cases. RL agent–recommended insulin dosage interval includes the actual dose prescribed by the physician in 53 out of 60 cases (53/60, 88%).

Conclusions

This exploratory study demonstrates that an RL algorithm can be used to recommend personalized insulin doses to achieve adequate glycemic control in patients with T1DM. However, further investigation in a larger sample of patients is needed to confirm these findings.

How Knowledge Emerges From Artificial Intelligence Algorithm and Data Visualization for Diabetes Management

BACKGROUND

Self-monitoring blood glucose (SMBG) is facilitated by application available to analyze these data. They are mainly based on descriptive statistical analyses. In this study, we are proposing a method inspired by artificial intelligence algorithm for displaying glycemic data in an intelligible way with high-level information that is compatible with the short duration allocated to medical visits.

METHOD

We propose a display method based on a numerical glycemic data conversion using a qualitative color scale that exhibits the patient's overall glycemic state. Moreover, a machine learning algorithm inputs these displays to exhibit recurrent glycemic pattern over configurable extended time period.

RESULTS

A demonstrator of our method, output as a glycemic map, could be used by the physician during quarterly patient consultations. We have tested this methodology retrospectively on a database in order to observe the behavior of our algorithm. In some data files we were able to highlight some of the glycemic patterns characteristics that remain invisible on the tabular representations or through the use of descriptive statistic. In a next step the interpretation will have to be done by physicians to confirm they underlie knowledge.

CONCLUSIONS

Our approach with artificial intelligence algorithm paired up with graphical color display allow a large database fast analysis to provide insights on diabetes knowledge. The next steps are first to set up a clinical trial to validate this methodology with dedicated patients and physicians then we will adapt our methodology for the huge data sets generated by continuous glycemic measurement (CGM) devices.

Big data analytics for preventive medicine

Medical data is one of the most rewarding and yet most complicated data to analyze. How can healthcare providers use modern data analytics tools and technologies to analyze and create value from complex data? Data analytics, with its promise to efficiently discover valuable pattern by analyzing large amount of unstructured, heterogeneous, non-standard and incomplete healthcare data. It does not only forecast but also helps in decision making and is increasingly noticed as breakthrough in ongoing advancement with the goal is to improve the quality of patient care and reduces the healthcare cost. The aim of this study is to provide a comprehensive and structured overview of extensive research on the advancement of data analytics methods for disease prevention. This review first introduces disease prevention and its challenges followed by traditional prevention methodologies. We summarize state-of-the-art data analytics algorithms used for classification of disease, clustering (unusually high incidence of a particular disease), anomalies detection (detection of disease) and association as well as their respective advantages, drawbacks and guidelines for selection of specific model followed by discussion on recent development and successful application of disease prevention methods. The article concludes with open research challenges and recommendations.

A review of personalized blood glucose prediction strategies for T1DM patients

This paper presents a methodological review of models for predicting blood glucose (BG) concentration, risks and BG events. The surveyed models are classified into three categories, and they are presented in summary tables containing the most relevant data regarding the experimental setup for fitting and testing each model as well as the input signals and the performance metrics. Each category exhibits trends that are presented and discussed. This document aims to be a compact guide to determine the modeling options that are currently being exploited for personalized BG prediction.

Mining data when technology is applied to support patients and professional on the control of chronic diseases: the experience of the METABO platform for diabetes management

This chapter provides an overview of how healthcare institution could benefit from the usage of technologies and personal health systems. Clinical, Usage and Technical data are mined in different ways and with different methods to support users (patients, health professionals and informal caregivers) in taking decisions. As a case study, the solutions and the techniques adopted in a research project focused on the delivery of technologies to improve diabetes management are described.

Big Data Technologies: New Opportunities for Diabetes Management

The so-called big data revolution provides substantial opportunities to diabetes management. At least 3 important directions are currently of great interest. First, the integration of different sources of information, from primary and secondary care to administrative information, may allow depicting a novel view of patient's care processes and of single patient's behaviors, taking into account the multifaceted nature of chronic care. Second, the availability of novel diabetes technologies, able to gather large amounts of real-time data, requires the implementation of distributed platforms for data analysis and decision support. Finally, the inclusion of geographical and environmental information into such complex IT systems may further increase the capability of interpreting the data gathered and extract new knowledge from them. This article reviews the main concepts and definitions related to big data, it presents some efforts in health care, and discusses the potential role of big data in diabetes care. Finally, as an example, it describes the research efforts carried on in the MOSAIC project, funded by the European Commission.

Co-medication of pravastatin and paroxetine-a categorical study

Electronic Medical Records (EMRs) are wealthy storehouses of patient information, to which data mining techniques can be prudently applied to reveal clinically significant patterns. Detecting patterns in drug-drug interactions, leading to adverse drug reactions is a powerful application of EMR data mining. Adverse effects of drug treatments can be investigated by mining clinical laboratory tests data which are reliable indicators of abnormal physiological functions. We report here the co-medication effects of pravastatin (HMG-CoA reductase inhibitor) and paroxetine (selective serotonin reuptake inhibitor (SSRI) anti-depressant) on significant clinical parameters, identified through a data mining analysis conducted on the Allscripts data warehouse. We found that the concomitant drug treatments of pravastatin and paroxetine increased the mean values of glucose serum from 113.2 to 132.1 mg/dL and international normalized ratio (INR) from 2.18 to 2.52, respectively. It also decreased the mean values of estimated glomerular filtration rate (eGFR) from 43 to 37 mL/min/1.73 m(3) and blood CO2 levels from 24.8 to 23.9 mEq/L respectively. Our findings indicate that co-medication of pravastatin and paroxetine might have significant impact on blood anti-coagulation, kidney function, and glucose homeostasis. Our methodology can be applied to any EMR data set to reveal co-medication effects of any drug pairs.

A systematic review of IT for diabetes self-management: are we there yet?

BACKGROUND

Recent advances in information technology (IT) coupled with the increased ubiquitous nature of information technology (IT) present unique opportunities for improving diabetes self-management. The objective of this paper is to determine, in a systematic review, how IT has been used to improve self-management for adults with Type 1 and Type 2 diabetes.

METHODS

The review covers articles extracted from relevant databases using search terms related information technology and diabetes self-management published after 1970 until August 2012. Additional articles were extracted using the citation map in Web of Science. Articles representing original research describing the use of IT as an enabler for self-management tasks performed by the patient are included in the final analysis.

RESULTS

Overall, 74% of studies showed some form of added benefit, 13% articles showed no-significant value provided by IT, and 13% of articles did not clearly define the added benefit due to IT. Information technologies used included the Internet (47%), cellular phones (32%), telemedicine (12%), and decision support techniques (9%). Limitations and research gaps identified include usability, real-time feedback, integration with provider electronic medical record (EMR), as well as analytics and decision support capabilities.

CONCLUSION

There is a distinct need for more comprehensive interventions, in which several technologies are integrated in order to be able to manage chronic conditions such as diabetes. Such IT interventions should be theoretically founded and should rely on principles of user-centered and socio-technical design in its planning, design and implementation. Moreover, the effectiveness of self-management systems should be assessed along multiple dimensions: motivation for self-management, long-term adherence, cost, adoption, satisfaction and outcomes as a final result.

Designing a Human T-Lymphotropic Virus Type 1 (HTLV-I) Diagnostic Model using the Complete Blood Count

OBJECTIVE(S)

Infection caused by Human T-Lymphotropic Virus Type 1 (HTLV-I) can be observed in some areas of Iran in form of endemic. Most of the cases are asymptomatic, and few cases progress to malignancies and neural diseases. Designing and implementing a model to screen people especially in endemic regions can help timely detection of infected people and improve the prognosis of the disease.

MATERIALS AND METHODS

In this study, results of the complete blood count (CBC-diff) for 599 healthy people and the patients with different types of Leukemia and HTLV-I have been examined. Modeling was made using CHAID method. The final model was carried out based on the number of white blood cells (WBC), platelets, and percentages of eosinophils.

RESULTS

The accuracy of the final model was 91%. By applying this model to the CBC-diff results of people without symptoms or miscellaneous patients in endemic regions of our country, disease carriers can be identified and referred for supplementary tests.

CONCLUSION

With regard to the prevalence of different complications in infected people, these individuals can be identified earlier, leading to the improvement of the prognosis of this disease and the increase of the health status especially in endemic regions.

AUTMINER: A SYSTEM FOR EXTRACTING ASD-RELATED GENES USING TEXT MINING

Autism spectrum disorders (ASD) represent a group of developmental disorders with strong genetic underpinnings. To explore the genetic complexity of ASD, we developed AutMiner (), a public web-portal for the collection of genes linked to ASD, and the implementation of an autism-centre network. AutMiner extracts candidate genes associated with ASD using text mining from 9276 abstracts. Compared to other recent systems, gene entries are richer to provide a reference for clinical geneticists. AutMiner also constructs ASD-related network consisting of autism-gene network and gene-gene network. To the best of our knowledge, this is the first web example of ASD-related network. The major focus of AutMiner is to offer a valuable reference tool for clinical geneticists in establishing and implementing effective genetic screening programmes for those patients with ASD.

Data-mining technologies for diabetes: a systematic review

BACKGROUND

The objective of this study is to conduct a systematic review of applications of data-mining techniques in the field of diabetes research.

METHOD

We searched the MEDLINE database through PubMed. We initially identified 31 articles by the search, and selected 17 articles representing various data-mining methods used for diabetes research. Our main interest was to identify research goals, diabetes types, data sets, data-mining methods, data-mining software and technologies, and outcomes.

RESULTS

The applications of data-mining techniques in the selected articles were useful for extracting valuable knowledge and generating new hypothesis for further scientific research/experimentation and improving health care for diabetes patients. The results could be used for both scientific research and real-life practice to improve the quality of health care diabetes patients.

CONCLUSIONS

Data mining has played an important role in diabetes research. Data mining would be a valuable asset for diabetes researchers because it can unearth hidden knowledge from a huge amount of diabetes-related data. We believe that data mining can significantly help diabetes research and ultimately improve the quality of health care for diabetes patients.