(1) Background: nocturnal hypoglycemia (NH) is one of the most challenging side effects of multiple doses of insulin (MDI) therapy in type 1 diabetes (T1D). This work aimed to investigate the feasibility of a machine-learning-based prediction model to anticipate NH in T1D patients on MDI. (2) Methods: ten T1D adults were studied during 12 weeks. Information regarding T1D management, continuous glucose monitoring (CGM), and from a physical activity tracker were obtained under free-living conditions at home. Supervised machine-learning algorithms were applied to the data, and prediction models were created to forecast the occurrence of NH. Individualized prediction models were generated using multilayer perceptron (MLP) and a support vector machine (SVM). (3) Results: population outcomes indicated that more than 70% of the NH may be avoided with the proposed methodology. The predictions performed by the SVM achieved the best population outcomes, with a sensitivity and specificity of 78.75% and 82.15%, respectively. (4) Conclusions: our study supports the feasibility of using ML techniques to address the prediction of nocturnal hypoglycemia in the daily life of patients with T1D on MDI, using CGM and a physical activity tracker.

Aim: Although insulin resistance (IR) is a growing trait among type 1 diabetes (T1D) population, its relationship with atherosclerosis has been scarcely studied. We assessed the association between IR indexes and carotid atherosclerosis in T1D, a population at high cardiovascular disease (CVD) risk.Materials and Methods: We evaluated 191 participants with T1D and no prior CVD with at least one of the following criteria: ≥40 years old; diabetic nephropathy; or T1D duration ≥10 years harbouring ≥1 additional CVD risk factor. IR was assessed with the metabolic syndrome (MetS) harmonized definition proposed in 2009 and the estimated glucose disposal rate (eGDR), a T1D‐specific IR surrogate marker (lower values indicating higher IR). Standardized carotid ultrasonography was performed, recording intima‐media thickness (IMT), plaque presence and maximum height of plaque. Comparisons between patients according to their MetS status as well as concerning eGDR values were performed.Results: The participants' median age was 47.4 (41.1‐53.3) years and diabetes duration 25.7 (21.6‐32.5) years. Plaque prevalence was higher in patients with greater IR (49.1%, 29.1% and 20%, P = .001, for any plaque according to decreasing eGDR tertiles). Conversely, no statistically significant higher plaque prevalence was found in participants with MetS. In multivariate analyses (adjusted for general‐ and T1D‐specific risk factors, and statin treatment), MetS was associated with neither IMT nor plaque. On the contrary, eGDR was independently related to ≥2 plaques (P = .018) and maximum plaque height (P < .01).Conclusions: In T1D, IR assessed through eGDR but not by MetS definition was independently associated with plaque burden, a predictor of CVD.

Nocturnal hypoglycemia (NH) is one of the most challenging events for multiple dose insulin therapy (MDI) in people with type 1 diabetes (T1D). The goal of this study is to design a method to reduce the incidence of NH in people with T1D under MDI therapy, providing a decision-support system and improving confidence toward self-management of the disease considering the dataset used by Bertachi et al. Different machine learning (ML) algorithms, data sources, optimization metrics and mitigation measures to predict and avoid NH events have been studied. In addition, we have designed population and personalized models and studied the generalizability of the models and the influence of physical activity (PA) on them. Obtaining 30 g of rescue carbohydrates (CHO) is the optimal value for preventing NH, so it can be asserted that this is the value with which the time under 70 mg/dL decreases the most, with almost a 35% reduction, while increasing the time in the target range by 1.3%. This study supports the feasibility of using ML techniques to address the prediction of NH in patients with T1D under MDI therapy, using continuous glucose monitoring (CGM) and a PA tracker. The results obtained prove that BG predictions can not only be critical in achieving safer diabetes management, but also assist physicians and patients to make better and safer decisions regarding insulin therapy and their day-to-day lives.

Background: The optimal cardiovascular assessment of liver transplant (LT) candidates is unclear. We aimed to evaluate the performance of CT‐based coronary tests (coronary artery calcium score [CACS] and coronary CT angiography [CCTA]) and a modification of the CAD‐LT score (mCAD‐LT, excluding family history of CAD) to diagnose significant coronary artery disease (CAD) before LT and predict the incidence of post‐LT cardiovascular events (CVE). Methods: We retrospectively analysed a single‐centre cohort of LT candidates who underwent non‐invasive tests; invasive coronary angiography (ICA) was performed depending on the results of non‐invasive tests. mCAD‐LT was calculated in all patients. Results: Six‐hundred‐and‐thirty‐four LT candidates were assessed and 351 of them underwent LT. CACS, CCTA and ICA were performed in 245, 123 and 120 LT candidates, respectively. Significant CAD was found in 30% of patients undergoing ICA. The AUROCs of mCAD‐LT (.722) and CCTA (.654) were significantly higher than that of CACS (.502) to predict the presence of significant CAD. Specificity of the tests ranged between 31% for CCTA and 53% for CACS. Among patients who underwent LT, CACS ≥ 400 and mCAD‐LT were independently associated with the incidence of CVE; in patients who underwent CCTA before LT, significant CAD at CCTA also predicted post‐LT CVE. Conclusion: In this cohort, mCAD‐LT score and CT‐based tests detect the presence of significant CAD in LT candidates, although they tend to overestimate it. Both mCAD‐LT score and CT‐based tests classify LT recipients according to their risk of post‐LT CVE and can be used to improve post‐LT risk mitigation.