Clara Mosquera-Lopez's research while affiliated with Oregon Health and Science University and other places

... This was accomplished using a KNN algorithm to train the digital twin to recognize correlations between glucose levels and heart rate patterns. This has allowed for a method of indirectly monitoring glucose concentrations through an easily worn fitness tracker watch [23]. ...

... By utilizing continuous glucose monitors, patients can transition from intermittent to continuous glucose monitoring every few minutes with a tiny flex sensor inserted (such as Abbott's FreeStyle Libre 2 sensor and FreeStyle LibreLink APP) or implanted under the skin, thereby obtaining more comprehensive data on glucose fluctuations (Galindo & Aleppo, 2020;Lee et al., 2021). AI algorithms (in particular machine learning algorithms), through temporal analysis of blood glucose values, insulin dosage, and various meal related nutritional factors, can assist patients in predicting trends in glucose fluctuations (blood glucose prediction or blood glucose interval prediction) and provide real-time corresponding coping suggestions (such as appropriate diet, exercise, or insulin settings advice) based on these trends, thereby helping patients better manage their blood glucose levels (Annuzzi et al., 2023;Jacobs et al., 2024). The features of smart-BGM technology could be especially useful for patients who lack knowledge and methods, or lack self-efficacy. ...

... The study applied a novel AI algorithm, called Evidential Neural Network, to forecast the probability of hypoglycemia within an interval of 8 hours overnight. 64 Dr Moshe Phillip described the successful trajectory of his research in implementing an AI-based decision support system for diabetes management, comprising support to both clinicians and patients in optimizing insulin therapy. 22 In particular, a non-inferiority trial demonstrated that the frequent insulin dose adjustments suggested by a decision support system had the same quality and safety as those provided by clinical experts. ...

... With the ubiquity of the smartphone [15], and the increasing popularity of wearable smart devices [16] such as the smartwatch and the smart ring, the use of such tools for monitoring pregnancy-related signs and symptoms and greatly transforming current knowledge of this important metamorphosis is a potential reality. While the body of research involving personal DHTs in pregnancy is small, a number of small preliminary studies have shown potential utility in using wearables for tracking activity, sleep, and physiologic metrics such as HR and HRV as predictors of pregnancy-related outcomes such as preterm birth and delivery readiness [17][18][19][20][21][22][23]. ...

... Innovative approaches to improving performance of closed-loop systems include incorporation of signals from other wearables such as a smartwatch app that detects eating behaviour [63] or a fitness sensor for detecting exercise/ activity [64]. The potential clinical benefits of integrating these devices remains to be determined in larger and longer trials. ...

... To predict insulin underdosing during digestion, we use a food prediction deep-learning model from the literature [45]. Our version of this model is a generalization where we model any source of glucose, not just food, but the details of it are beyond the scope of this paper. ...

... Thanks to large-scale and ongoing research studies, we can do a better job at predicting hypoglycemia risk now that we understand what factors play the biggest role. The present study by Mosquera-Lopez and colleagues revealed that the key risk factors significantly associated with hypoglycemia included glucose and body exposure to insulin at PA onset, the low blood glucose index in the 24 hours before PA, and the intensity and timing of the PA (24). Additional work by Bergford et al. (5) identified similar yet different hypoglycemia risk factors, including lower starting pre-exercise glucose levels, decreasing glucose levels before exercise, and higher levels of insulin on board or active insulin pre-exercise. ...

... Some decision support systems have integrated these types of guidelines into smartphone apps to make it easier for patients to follow the recommendations. 22 However, clinical-based guidelines may not be suitable for all people as they do not account for individual differences in insulin sensitivity, diet and exercise responses that can have a major impact on glucose levels during and following exercise. 23 Digital twins are mathematical model representations of real-world phenomena. ...

... Penelitian lain yang dilakukan oleh Lu et al. (2021) pada pasien dengan gangguan metabolism di Cina menunjukkan bahwa latihan fisik dapat menurunkan kadar glukosa darah baik pada orang normal ataupun pada pasien DM. Peyerapan glukosa oleh otot akan meningkat seiring dengan kegiatan atau latihan fisik yang dilakukan, sehingga hal inilah yang berperan dalam penurunan kadar glukosa darah (Asfaw and Dagne, 2022;Tyler et al., 2022;Liu et al., 2024). ...

... In addition, the authors in [232] performed their research in determining the heart condition based on its sounds and 20 VOLUME 4, 2016 This article has been accepted for publication in IEEE Access. This is the author's version which has not been fully edited and content may change prior to final publication. ...