Development of Machine Learning Models for the Identification of Elevated Ketone Bodies During Hyperglycemia in Patients with Type 1 Diabetes.

Aims: Diabetic ketoacidosis (DKA) is a serious life-threatening condition caused by a lack of insulin, which leads to elevated plasma glucose and metabolic acidosis. Early identification of developing DKA is important to start treatment and minimize complications and risk of death. The aim of the present study is to develop and test prediction model(s) that gives an alarm about their risk of developing elevated ketone bodies during hyperglycemia. Methods: We analyzed data from 138 type 1 diabetes patients with measurements of ketone bodies and continuous glucose monitoring (CGM) data from over 30,000 days of wear time. We utilized a supervised binary classification machine learning approach to identify elevated levels of ketone bodies (≥0.6 mmol/L). Data material was randomly divided at patient level in 70%/30% (training/test) dataset. Logistic regression (LR) and random forest (RF) classifier were compared. Results: Among included patients, 913 ketone samples were eligible for modeling, including 273 event samples with ketone levels ≥0.6 mmol/L. An area under the receiver operating characteristic curve from the RF classifier was 0.836 (confidence interval [CI] 90%, 0.783-0.886) and 0.710 (CI 90%, 0.646-0.77) for the LR classifier. Conclusions: The novel approach for identifying elevated ketone levels in patients with type 1 diabetes utilized in this study indicates that CGM could be a valuable resource for the early prediction of patients at risk of developing DKA. Future studies are needed to validate the results.

The Effectiveness of Telemedicine Solutions in Type 1 Diabetes Management: A Systematic Review and Meta-analysis.

BACKGROUND: Telemedicine holds a potential to strengthen self-management support outside health care settings in the everyday management of type 1 diabetes (T1D). However, existing effectiveness reviews are older or include a relatively narrow focus on specific definitions of telemedicine or included databases. OBJECTIVE: To conduct a systematic review of the effectiveness of telemedicine solutions versus any comparator on diabetes-related outcomes among people with T1D. METHODS: Studies including adults (≥18 years) with T1D published before October 14, 2020, were eligible. Primary outcome was glycated hemoglobin (HbA1c, %). The Cochrane Library, PubMed, EMBASE, and CINAHL were searched. Meta-analysis based on the mean difference in HbA1c% was used to pool effects. The Cochrane tool was applied to assess risk-of-bias, and the certainty of evidence was graded using the GRADE approach. RESULTS: A total of 22 studies were included (with 1615 participants). Treatment effect for HbA1c% favored telemedicine (mean difference of -0.26% [95% confidence interval:-0.37% to -0.15%]) with moderate effect certainty. Heterogeneity was moderate (I2 = 33.30%). Although not significant, secondary outcomes were all in favor of telemedicine except number of severe hypoglycemic events and diabetes knowledge, but the certainty of the evidence for those outcomes was all low or very low. DISCUSSION: Reducing average HbA1c% levels are important to combat the risk of diabetic complications and premature death. However, the evidence mostly consist of small studies with a relative short duration and the estimated pooled effect is smaller than could be expected from quality improvement strategies in general for diabetes management. PROSPERO NUMBER: CRD42020123565.

Clinical implementation of an algorithm for predicting exacerbations in patients with COPD in telemonitoring: a study protocol for a single-blinded randomized controlled trial.

BACKGROUND: Acute exacerbations have a significant impact on patients with COPD by accelerating the decline in lung function leading to decreased health-related quality of life and survival time. In telehealth, health care professionals exercise clinical judgment over a physical distance. Telehealth has been implemented as a way to monitor patients more closely in daily life with an intention to intervene earlier when physical measurements indicate that health deteriorates. Several studies call for research investigating the ability of telehealth to automatically flag risk of exacerbations by applying the physical measurements that are collected as part of the monitoring routines to support health care professionals. However, more research is needed to further develop, test, and validate prediction algorithms to ensure that these algorithms improve outcomes before they are widely implemented in practice. METHOD: This trial tests a COPD prediction algorithm that is integrated into an existing telehealth system, which has been developed from the previous Danish large-scale trial, TeleCare North (NCT: 01984840). The COPD prediction algorithm aims to support clinical decisions by predicting the risk of exacerbations for patients with COPD based on selected physiological parameters. A prospective, parallel two-armed randomized controlled trial with approximately 200 participants with COPD will be conducted. The participants live in Aalborg municipality, which is located in the North Denmark Region. All participants are familiar with the telehealth system in advance. In addition to the participants' usual weekly monitored measurements, they are asked to measure their oxygen saturation two more times a week during the trial period. The primary outcome is the number of exacerbations defined as an acute hospitalization from baseline to follow-up. Secondary outcomes include changes in health-related quality of life measured by both the 12-Item Short Form Survey version 2 and EuroQol-5 Dimension Questionnaire as well as the incremental cost-effectiveness ratio. DISCUSSION: This trial seeks to explore whether the COPD prediction algorithm has the potential to support early detection of exacerbations in a telehealth setting. The COPD prediction algorithm may initiate timely treatment, which may decrease the number of hospitalizations. TRIAL REGISTRATION: NCT05218525 (pending at clinicaltrials.gov ) (date, month, year).

Using Case-Based Reasoning in a Learning System: A Prototype of a Pedagogical Nurse Tool for Evidence-Based Diabetic Foot Ulcer Care.

BACKGROUND: Currently, evidence-based learning systems to increase knowledge and evidence level of wound care are unavailable to wound care nurses in Denmark, which means that they need to learn about diabetic foot ulcers from experience and peer-to-peer training, or by asking experienced colleagues. Interactive evidence-based learning systems built on case-based reasoning (CBR) have the potential to increase wound care nurses' diabetic foot ulcer knowledge and evidence levels. METHOD: A prototype of a CBR-interactive, evidence-based algorithm-operated learning system calculates a dissimilarity score (DS) that gives a quantitative measure of similarity between a new case and cases stored in a case base in relation to six variables: necrosis, wound size, granulation, fibrin, dry skin, and age. Based on the DS, cases are selected by matching the six variables with the best predictive power and by weighing the impact of each variable according to its contribution to the prediction. The cases are ranked, and the six cases with the lowest DS are visualized in the system. RESULTS: Conventional education, that is, evidence-based learning material such as books and lectures, may be less motivating and pedagogical than peer-to-peer training, which is, however, often less evidence-based. The CBR interactive learning systems presented in this study may bridge the two approaches. Showing wound care nurses how individual variables affect outcomes may help them achieve greater insights into pathophysiological processes. CONCLUSION: A prototype of a CBR-interactive, evidence-based learning system that is centered on diabetic foot ulcers and related treatments bridges the gap between traditional evidence-based learning and more motivating and interactive learning approaches.

Preliminary Qualitative Evaluation of Patient-Related Perspectives Related to the Implementation of a Predictive Algorithm in a Telehealth System for COPD.

The aim of the present study was to evaluate patient-related perspectives from a five-week test of the implementation of a COPD prediction algorithm. The test intended to discover and avoid potential errors prior to testing the COPD prediction algorithm in a two-armed randomized controlled trial (RCT). The COPD prediction algorithm aims to predict exacerbations in COPD based on home measurements. In the present study, the algorithm was implemented in a currently deployed telehealth system. Five weeks after implementation, six interviews were conducted, including five interviews with patients with COPD and one interview with a specialized COPD nurse. The participants were overall satisfied with the telehealth system including the COPD prediction algorithm. However, technical issues must be addressed before the COPD prediction algorithm is ready to be tested in the RCT. Moreover, communication with the monitoring nurses should be clearer based on the COPD nurse's experiences. In conclusion, the participants were satisfied with the integration of the COPD prediction algorithm in the telehealth system. The identification of technical issues shows the importance of including a technical test period in a similar trial setup.

Assessment of Simple Bedside Wound Characteristics for a Prediction Model for Diabetic Foot Ulcer Outcomes.

BACKGROUND: Evidence-based learning systems built on prediction models can support wound care community nurses (WCCNs) during diabetic foot ulcer care sessions. Several prediction models in the area of diabetic foot ulcer healing have been developed, most built on cardiovascular measurement data. Two other data types are patient information (i.e. sex and hemoglobin A1c) and wound characteristics (i.e. wound area and wound duration); these data relate to the status of the diabetic foot ulcer and are easily accessible for WCCNs. The aim of the study was to assess simple bedside wound characteristics for a prediction model for diabetic foot ulcer outcomes. METHOD: Twenty predictor variables were tested. A pattern prediction model was used to forecast whether a given diabetic foot ulcer would (i) increase in size (or not) or (ii) decrease in size. Sensitivity, specificity, and area under the curve (AUC) in a receiver-operating characteristics curve were calculated. RESULTS: A total of 162 diabetic foot ulcers were included. In combination, the predictor variables necrosis, wound size, granulation, fibrin, dry skin, and age were most informative, in total an AUC of 0.77. CONCLUSIONS: Wound characteristics have potential to predict wound outcome. Future research should investigate implementation of the prediction model in an evidence-based learning system.