Detection of Hypoglycemia and Hyperglycemia Using Noninvasive Wearable Sensors: Electrocardiograms and Accelerometry

BACKGROUND

Monitoring glucose excursions is important in diabetes management. This can be achieved using continuous glucose monitors (CGMs). However, CGMs are expensive and invasive. Thus, alternative low-cost noninvasive wearable sensors capable of predicting glycemic excursions could be a game changer to manage diabetes.

METHODS

In this article, we explore two noninvasive sensor modalities, electrocardiograms (ECGs) and accelerometers, collected on five healthy participants over two weeks, to predict both hypoglycemic and hyperglycemic excursions. We extract 29 features encompassing heart rate variability features from the ECG, and time- and frequency-domain features from the accelerometer. We evaluated two machine learning approaches to predict glycemic excursions: a classification model and a regression model.

RESULTS

The best model for both hypoglycemia and hyperglycemia detection was the regression model based on ECG and accelerometer data, yielding 76% sensitivity and specificity for hypoglycemia and 79% sensitivity and specificity for hyperglycemia. This had an improvement of 5% in sensitivity and specificity for both hypoglycemia and hyperglycemia when compared with using ECG data alone.

CONCLUSIONS

Electrocardiogram is a promising alternative not only to detect hypoglycemia but also to predict hyperglycemia. Supplementing ECG data with contextual information from accelerometer data can improve glucose prediction.

Predictive model for microclimatic temperature and its use in mosquito population modeling

Mosquitoes transmit several infectious diseases that pose significant threat to human health. Temperature along with other environmental factors at breeding and resting locations play a role in the organismal development and abundance of mosquitoes. Accurate analysis of mosquito population dynamics requires information on microclimatic conditions at breeding and resting locations. In this study, we develop a regression model to characterize microclimatic temperature based on ambient environmental conditions. Data were collected by placing sensor loggers at resting and breeding locations such as storm drains across Houston, TX. Corresponding weather data was obtained from National Oceanic and Atmospheric Administration website. Features extracted from these data sources along with contextual information on location were used to develop a Generalized Linear Model for predicting microclimate temperatures. We also analyzed mosquito population dynamics for Aedes albopictus under ambient and microclimatic conditions using system dynamic (SD) modelling to demonstrate the need for accurate microclimatic temperatures in population models. The microclimate prediction model had an R² value of ~ 95% and average prediction error of ~ 1.5 °C indicating that microclimate temperatures can be reliably estimated from the ambient environmental conditions. SD model analysis indicates that some microclimates in Texas could result in larger populations of juvenile and adult Aedes albopictus mosquitoes surviving the winter without requiring dormancy.

Feature-Based Machine Learning Model for Real-Time Hypoglycemia Prediction

BACKGROUND

Hypoglycemia is a serious health concern in youth with type 1 diabetes (T1D). Real-time data from continuous glucose monitoring (CGM) can be used to predict hypoglycemic risk, allowing patients to take timely intervention measures.

METHODS

A machine learning model is developed for probabilistic prediction of hypoglycemia (<70 mg/dL) in 30- and 60-minute time horizons based on CGM datasets obtained from 112 patients over a range of 90 days consisting of over 1.6 million CGM values under normal living conditions. A comprehensive set of features relevant for hypoglycemia are developed and a parsimonious subset with most influence on predicting hypoglycemic risk is identified. Model performance is evaluated both with and without contextual information on insulin and carbohydrate intake.

RESULTS

The model predicted hypoglycemia with >91% sensitivity for 30- and 60-minute prediction horizons while maintaining specificity >90%. Inclusion of insulin and carbohydrate data yielded performance improvement for 60-minute but not for 30-minute predictions. Model performance was highest for nocturnal hypoglycemia (~95% sensitivity). Shortterm (less than one hour) and medium-term (one to four hours) features for good prediction performance are identified.

CONCLUSIONS

Innovative feature identification facilitated high performance for hypoglycemia risk prediction in pediatric youth with T1D. Timely alerts of impending hypoglycemia may enable proactive measures to avoid severe hypoglycemia and achieve optimal glycemic control. The model will be deployed on a patient-facing smartphone application in an upcoming pilot study.

993 Digital Monitoring of Paediatric Patients with Chronic Recurrent Multifocal Osteomyelitis at A National Orthopaedic Centre

Introduction

Chronic recurrent multifocal osteomyelitis (CRMO) is a paediatric inflammatory bone condition requiring close monitoring by a multi-disciplinary team throughout childhood. Many UK patients are seen at the Royal National Orthopaedic Hospital, Stanmore for more than a decade of specialist management. Prior to the recent COVID crisis, we recognised need for more frequent monitoring without the inconvenience of cross-country travel and so piloted an online system of monitoring CRMO.

Method

An online assessment form was created using the GDPR-compliant Qualtrics system after consultation with CRMO paediatricians and a specialist nurse. 3 medical students telephoned parents of 31 patients to pilot the assessment form. Responses indicating issues were then referred to the CNS for follow-up.

Results

Response data confirmed need for close monitoring with 26% of respondents being seen quarterly, 29% reporting pain greater than 6/10, 29% having side-effects and 30% finding current treatment ineffective. 74% of respondents found online assessment extremely/very useful and 58% preferred it to face-to-face appointments.

Conclusions

A digital CRMO monitoring tool can successfully identify patients with both new and ongoing issues. In the COVID and post-COVID era, digital tools can closely monitor patients with chronic conditions like CRMO whilst improving quality of life by reducing need for face-to-face appointments.

Improved Low-Glucose Predictive Alerts Based on Sustained Hypoglycemia: Model Development and Validation Study

BACKGROUND

Predictive alerts for impending hypoglycemic events enable persons with type 1 diabetes to take preventive actions and avoid serious consequences.

OBJECTIVE

This study aimed to develop a prediction model for hypoglycemic events with a low false alert rate, high sensitivity and specificity, and good generalizability to new patients and time periods.

METHODS

Performance improvement by focusing on sustained hypoglycemic events, defined as glucose values less than 70 mg/dL for at least 15 minutes, was explored. Two different modeling approaches were considered: (1) a classification-based method to directly predict sustained hypoglycemic events, and (2) a regression-based prediction of glucose at multiple time points in the prediction horizon and subsequent inference of sustained hypoglycemia. To address the generalizability and robustness of the model, two different validation mechanisms were considered: (1) patient-based validation (model performance was evaluated on new patients), and (2) time-based validation (model performance was evaluated on new time periods).

RESULTS

This study utilized data from 110 patients over 30-90 days comprising 1.6 million continuous glucose monitoring values under normal living conditions. The model accurately predicted sustained events with >97% sensitivity and specificity for both 30- and 60-minute prediction horizons. The false alert rate was kept to <25%. The results were consistent across patient- and time-based validation strategies.

CONCLUSIONS

Providing alerts focused on sustained events instead of all hypoglycemic events reduces the false alert rate and improves sensitivity and specificity. It also results in models that have better generalizability to new patients and time periods.

Risk management strategies following analysis of cataract negligence claims

INTRODUCTION

Clinical governance and risk management is very important in today's clinical practice. Cataract surgery is one of the most common procedures performed in the NHS, with around 200,000 operations per year. In order to help minimise the frequency of negligence claims, we performed a collaborative study to analyse claims relating to cataract surgery, dealt with by the defence organisations of England, Scotland, Wales, and Northern Ireland.

MATERIALS AND METHODS

All claims dealt with by the Medical Defence Union, the Medical Protection Society, and the Medical and Dental Defence Union of Scotland from January 1990 to December 1999, were analysed by three ophthalmologists with at least 5 years' speciality experience. Recurrent themes were identified and claims were grouped by major causative factor. The findings were discussed by a panel compromising the authors in conjunction with the defence unions and risk management strategies were designed.

RESULTS

There were 96 claims within the 10- year period analysed. Of these, the largest group (52) pertained to claims that related to accepted complications of cataract surgery. The remainder comprised two groups: 'Medical Errors' (anaesthetic, surgeon, and biometry) and 'Other Claims' comprising subjective complaints, pain and poor visual outcome. A total of 16 claims had been settled by May 2002, 45 are ongoing and 35 have closed without settlement.

CONCLUSIONS

The majority of claims pertained to well-recognised complications of cataract surgery. If these risks are adequately explained to the patient before surgery and if the care provided reaches a standard acceptable to a responsible body of professional opinion, all such claims should be defensible. Good visual outcome does not protect against litigation.

Synthesis and anti-HIV activity of some non-nucleoside 2,3-disubstituted quinazoline derivatives (Part-V)

Several [2-phenyl-4(3H)-oxo-3-quinazolinylamino]-N-substituted- arylacetamides (1a-j) have been synthesized and tested at the National Cancer Institute, Bethesda, Maryland, USA, for their anti-HIV activity against susceptible human host cells (CEM cell line) over a wide range of concentrations (6.35 x 10(-8) to 2.00 x 10(-4) M). The highest protection observed is 45.67%. The structures of these compounds have been established on the basis of elemental analysis and spectral data.

Optical Doppler tomographic imaging of fluid flow velocity in highly scattering media

An optical Doppler tomography (ODT) system that permits imaging of fluid flow velocity in highly scattering media is described. ODT combines Doppler velocimetry with the high spatial resolution of low-coherence optical interferometry to measure fluid flow velocity at discrete spatial locations. Tomographic imaging of particle flow velocity within a circular conduit submerged 1 mm below the surface in a highly scattering phantom of Intralipid is demonstrated.

Conversion of dihydroorotate to orotate in parasitic protozoa

The conversion of dihydroorotate to orotate, one of the key reactions in the de novo pyrimidine biosynthetic pathway, has been studied in a number of parasitic protozoa. Enzyme activities capable of carrying out this reaction were detected in six members of the Kinetoplastida (Trypanosoma brucei, Trypanosoma congolense, Trypanosoma vivax, Trypanosoma lewisi, Trypanosoma cruzi, Leishmania enriettii) and three members of the genus Plasmodium (P. knowlesi, P. berghei, P. gallinaceum). The mechanism of the reaction in the two groups of protozoa were quite distinct. In the Kinetoplastida, the enzyme is an hydroxylase which occurs in the soluble fraction of the cell and probably requires tetrahydrobiopterin for activity. In contrast, in Plasmodium, the enzyme is a dehydrogenase which is particulate, probably mitochondrial, and intimately connected to the electron transport chain to which it passes electrons directly, probably at the ubiquinone level. Neither activity is regulated by fully formed pyrimidines. The enzyme in Plasmodium is similar in mechanism to the isofunctional mammalian enzyme. However, since malarial ubiquinones are apparently different from those in the mammal and since menoctone, which is active in vivo in experimental malaria, is a good inhibitor of the malarial enzyme, it could represent a useful target for chemotherapeutic attack. The enzyme in the Kinetoplastida is quite distinct from that in the mammal so that it too apparently falls into this category, though none of the currently used antitrypanosomal drugs appears to block it activity at physiological concentrations.