Matthew E. Levine

Personalized glucose forecasting for type 2 diabetes using data assimilation

Type 2 diabetes leads to premature death and reduced quality of life for 8% of Americans. Nutrition management is critical to maintaining glycemic control, yet it is difficult to achieve due to the high individual differences in glycemic response to nutrition. Anticipating glycemic impact of different meals can be challenging not only for individuals with diabetes, but also for expert diabetes educators. Personalized computational models that can accurately forecast an impact of a given meal on an individual’s blood glucose levels can serve as the engine for a new generation of decision support tools for individuals with diabetes. However, to be useful in practice, these computational engines need to generate accurate forecasts based on limited datasets consistent with typical self-monitoring practices of individuals with type 2 diabetes. This paper uses three forecasting machines: (i) data assimilation, a technique borrowed from atmospheric physics and engineering that uses Bayesian modeling to infuse data with human knowledge represented in a mechanistic model, to generate real-time, personalized, adaptable glucose forecasts; (ii) model averaging of data assimilation output; and (iii) dynamical Gaussian process model regression. The proposed data assimilation machine, the primary focus of the paper, uses a modified dual unscented Kalman filter to estimate states and parameters, personalizing the mechanistic models. Model selection is used to make a personalized model selection for the individual and their measurement characteristics. The data assimilation forecasts are empirically evaluated against actual postprandial glucose measurements captured by individuals with type 2 diabetes, and against predictions generated by experienced diabetes educators after reviewing a set of historical nutritional records and glucose measurements for the same individual. The evaluation suggests that the data assimilation forecasts compare well with specific glucose measurements and match or exceed in accuracy expert forecasts. We conclude by examining ways to present predictions as forecast-derived range quantities and evaluate the comparative advantages of these ranges.

Acute respiratory alkalosis associated with low minute ventilation in a patient with severe hypothyroidism

PurposePatients with severe hypothyroidism present unique challenges to anesthesiologists and demonstrate much increased perioperative risks. Overall, they display increased sensitivity to anesthetics, higher incidence of perioperative cardiovascular morbidity, increased risks for postoperative ventilatory failure and other physiological derangements. The previously described physiological basis for the increased incidence of postoperative ventiiatory failure in hypothyroid patients includes decreased central and peripheral ventiiatory responses to hypercarbia and hypoxia, muscle weakness, depressed central respiratory drive, and resultant alveolar hypoventilation. These ventiiatory failures are associated most frequently with severe hypoxia and carbon dioxide (CO2) retention. The purpose of this clinical report is to discuss an interesting and unique anesthetic presentation of a patient with severe hypothyroidism.Clinical featuresWe describe an unique presentation of ventilatory failure in a 58 yr old man with severe hypothyroidism. He had exceedingly low perioperative respiratory rate (3–4 bpm) and minute ventilation volume, and at the same time developed primary acute respiratory alkalosis and associated hypocarbia (PETCO2 =320-22 mmHg).ConclusionOur patient’s ventiiatory failure was based on unacceptably low minute ventilation and respiratory rate that was unable to sustain adequate oxygenation. His profoundly lowered basal metabolic rate and decreased CO2 production, resulting probably from severe hypothyroidism, may have resulted in development of acute respiratory alkalosis in spite of concurrently diminished minute ventilation.RésuméObjectifLes patients atteints d’hypothyroïdie sévère représentent tout un défi pour les anesthésiologistes étant donné les risques périopératoires très élevés. Dans l’ensemble, ils affichent une grande sensibilité aux anesthésiques, une forte incidence de morbidité cardiovasculaire périopératoire, des risques importants d’insuffisance ventilatoire postopératoire et d’autres désordres physiologiques. Les bases physiologiques, précédemment décrites, de l’incidence accrue d’insuffisance ventilatoire postopératoire chez ces patients comprennent des réponses ventilatoires centrales et périphériques diminuées à l’hypercapnie et à l’hypoxie, la faiblesse musculaire, la baisse de stimulation respiratoire centrale et l’hypoventilation alvéolaire qui en résulte. Ces défaillances ventilatoires sont le plus souvent associées à une hypoxie sévère et à une rétention de gaz carbonique (CO2). L’objectif du présent article est de commenter la présentation anesthésique unique et intéressante d’un patient atteint d’hypothyroïdie sévère.Éléments cliniquesIl s’agit d’un homme de 58 ans atteint d’hypothyroïdie sévère. Il présentait une fréquence respiratoire (3–4 bpm) et une ventilation-minute périopératoires extrêmement basses et il a, au même moment, développé une alcalose respiratoire primaire aiguë et une hypocarbie associée (PETCO2 ≈ 320–22 mmHg).ConclusionLa défaillance ventilatoire était basée sur d’inacceptables basses fréquence respiratoire et ventilationminute qui ne permettaient pas d’entretenir une oxygénation suffisante. La vitesse extrêmement faible du métabolisme basal et la production réduite de CO2, résultant probablement de l’hypothyroïdie sévère, ont pu conduire au développement d’une alcalose respiratoire aiguë malgré la diminution simultanée de la ventilation-minute.

From Personal Informatics to Personal Analytics: Investigating How Clinicians and Patients Reason About Personal Data Generated with Self-Monitoring in Diabetes

Diabetes self-management continues to present a significant challenge to millions of individuals around the world, as it often requires significant modifications to one’s lifestyle. The highly individual nature of the disease presents a need for each affected person to discover which daily activities have the most positive impact on one’s health and which are detrimental to it. Data collected with self-monitoring can help to reveal these relationships, however interpreting such data may be non-trivial. In this research we investigate how individuals with type 2 diabetes and their healthcare providers reason about data collected with self-monitoring and what computational methods can facilitate this process.

Pictures Worth a Thousand Words: Reflections on Visualizing Personal Blood Glucose Forecasts for Individuals with Type 2 Diabetes

Type 2 Diabetes Mellitus (T2DM) is a common chronic condition that requires management of ones lifestyle, including nutrition. Critically, patients often lack a clear understanding of how everyday meals impact their blood glucose. New predictive analytics approaches can provide personalized mealtime blood glucose forecasts. While communicating forecasts can be challenging, effective strategies for doing so remain little explored. In this study, we conducted focus groups with 13 participants to identify approaches to visualizing personalized blood glucose forecasts that can promote diabetes self-management and understand key styles and visual features that resonate with individuals with diabetes. Focus groups demonstrated that individuals rely on simple heuristics and tend to take a reactive approach to their health and nutrition management. Further, the study highlighted the need for simple and explicit, yet information-rich design. Effective visualizations were found to utilize common metaphors alongside words, numbers, and colors to convey a sense of authority and encourage action and learning.

Mechanistic machine learning: how data assimilation leverages physiologic knowledge using Bayesian inference to forecast the future, infer the present, and phenotype

Abstract We introduce data assimilation as a computational method that uses machine learning to combine data with human knowledge in the form of mechanistic models in order to forecast future states, to impute missing data from the past by smoothing, and to infer measurable and unmeasurable quantities that represent clinically and scientifically important phenotypes. We demonstrate the advantages it affords in the context of type 2 diabetes by showing how data assimilation can be used to forecast future glucose values, to impute previously missing glucose values, and to infer type 2 diabetes phenotypes. At the heart of data assimilation is the mechanistic model, here an endocrine model. Such models can vary in complexity, contain testable hypotheses about important mechanics that govern the system (eg, nutrition’s effect on glucose), and, as such, constrain the model space, allowing for accurate estimation using very little data.

A visual analytics approach for pattern-recognition in patient-generated data

Abstract Objective To develop and test a visual analytics tool to help clinicians identify systematic and clinically meaningful patterns in patient-generated data (PGD) while decreasing perceived information overload. Methods Participatory design was used to develop Glucolyzer, an interactive tool featuring hierarchical clustering and a heatmap visualization to help registered dietitians (RDs) identify associative patterns between blood glucose levels and per-meal macronutrient composition for individuals with type 2 diabetes (T2DM). Ten RDs participated in a within-subjects experiment to compare Glucolyzer to a static logbook format. For each representation, participants had 25 minutes to examine 1 month of diabetes self-monitoring data captured by an individual with T2DM and identify clinically meaningful patterns. We compared the quality and accuracy of the observations generated using each representation. Results Participants generated 50% more observations when using Glucolyzer (98) than when using the logbook format (64) without any loss in accuracy (69% accuracy vs 62%, respectively, p = .17). Participants identified more observations that included ingredients other than carbohydrates using Glucolyzer (36% vs 16%, p = .027). Fewer RDs reported feelings of information overload using Glucolyzer compared to the logbook format. Study participants displayed variable acceptance of hierarchical clustering. Conclusions Visual analytics have the potential to mitigate provider concerns about the volume of self-monitoring data. Glucolyzer helped dietitians identify meaningful patterns in self-monitoring data without incurring perceived information overload. Future studies should assess whether similar tools can support clinicians in personalizing behavioral interventions that improve patient outcomes.