Using Artificial Intelligence to Improve the Accuracy of a Wrist-Worn, Noninvasive Glucose Monitor: A Pilot Study.

BACKGROUND: Self-monitoring of glucose is important to the successful management of diabetes; however, existing monitoring methods require a degree of invasive measurement which can be unpleasant for users. This study investigates the accuracy of a noninvasive glucose monitoring system that analyses spectral variations in microwave signals. METHODS: An open-label, pilot design study was conducted with four cohorts (N = 5/cohort). In each session, a dial-resonating sensor (DRS) attached to the wrist automatically collected data every 60 seconds, with a novel artificial intelligence (AI) model converting signal resonance output to a glucose prediction. Plasma glucose was measured in venous blood samples every 5 minutes for Cohorts 1 to 3 and every 10 minutes for Cohort 4. Accuracy was evaluated by calculating the mean absolute relative difference (MARD) between the DRS and plasma glucose values. RESULTS: Accurate plasma glucose predictions were obtained across all four cohorts using a random sampling procedure applied to the full four-cohort data set, with an average MARD of 10.3%. A statistical analysis demonstrates the quality of these predictions, with a surveillance error grid (SEG) plot indicating no data pairs falling into the high-risk zones. CONCLUSIONS: These findings show that MARD values approaching accuracies comparable to current commercial alternatives can be obtained from a multiparticipant pilot study with the application of AI. Microwave biosensors and AI models show promise for improving the accuracy and convenience of glucose monitoring systems for people with diabetes.

A uniform strategy of primary repair of tetralogy of Fallot: Transventricular approach results in low reoperation rate in the first decade.

OBJECTIVES: To review outcomes after a uniform strategy of transventricular repair of tetralogy of Fallot. METHODS: A total of 244 consecutive patients underwent transventricular primary repair of tetralogy of Fallot from 2004 to 2019. Median age at operation was 71 days; 57 (23%) patients were premature; 57 (23%) patients had low birth weight (<2.5 kg), and 40 (16%) had genetic syndromes. The diameter of pulmonary valve annulus, right pulmonary artery (PA), and left PA were 6.0 ± 1.8 mm (z score, -1.7 ± 1.3), 4.3 ± 1.4 mm (z score, -0.9 ± 1.2) and 4.1 ± 1.5 mm (z score, -0.5 ± 1.3). RESULTS: Three (1.2%) operative deaths were recorded. Ninety patients (37%) underwent transannular patching. Postoperative echocardiographic peak right ventricular outflow tract gradient decreased from 72 ± 27 mm Hg to 21 ± 16 mm Hg. Median intensive care unit and hospital stay were 3 and 7 days. The survival rate at 10 years was 94.6% ± 1.8%. Reintervention was required 86 times (55 catheter interventions) in 56 patients following tetralogy of Fallot repair. The freedom from all-cause reintervention rate at 10 years was 70.5% ± 3.6%. Cyanotic spells (hazard ratio, 2.14; 95% CI, 1.22-3.90; P < .01) and smaller pulmonary valve annulus z score (hazard ratio, 1.26; 95% CI, 1.01-1.59; P = .04) were associated with increasing risk of all reinterventions. Freedom from redo surgery for right ventricular outflow tract obstruction and right ventricular dilatation at 10 years were, respectively, 85.0% ± 3.1% and 98.7% ± 0.9%. Freedom from valve implantation was 96.7% ± 1.5% at 10 years. CONCLUSIONS: A uniform strategy of primary repair of tetralogy of Fallot through a transventricular approach resulted in low reoperation rate in the first decade. The need of pulmonary valve implantation was limited to <4% at 10 years.

A comparison of low birth weight among newborns of early adolescents, late adolescents, and adult mothers in the Peruvian Amazon.

To compare low birth weight (LBW: <2,500 g) between infants born to adolescent and adult mothers in Iquitos, Peru. A random sample of 4,467 records of women who delivered at the Hospital Apoyo Iquitos between 2005 and 2007 was collected from hospital birth registries. Multivariate logistic and linear regression analyses were performed to compare LBW in newborns of adolescents (10-14, 15-19 years) and adults (≥20 years) and then for primiparous mothers with a normal gestational age, adjusting for newborn sex, antenatal care, and location of the mother's residence. A total of 4,384 mothers had had a singleton live birth and 1,501 were primiparous with a normal gestational age. Early and late adolescents had significantly greater odds of having a LBW infant than adults (OR = 2.28, 95%CI: 1.09, 4.78; OR = 1.67, 95%CI: 1.30, 2.14, respectively). For primiparous mothers with a normal gestational age, the same was true only for early adolescents (OR = 3.07, 95%CI: 1.09, 8.61). There were significant differences in mean birth weight between adults (3178.7 g) and both adolescent age groups overall (10-14 years: 2848.9 g; 15-19 years: 2998.3 g) and for primiparous mothers with a normal gestational age (10-14 years: 2900.8 g; 15-19 years: 3059.2 g; ≥20 years: 3151.8 g). Results suggest there is an important difference between adolescent and adult mothers in terms of newborn birth weight, especially among early adolescents. Future research on LBW and possibly other adverse birth outcomes should consider early adolescents as a separate sub-group of higher risk.

Comparison of prenatal care coverage in early adolescents, late adolescents, and adult pregnant women in the Peruvian Amazon.

OBJECTIVE: To compare prenatal care coverage between adolescent (early and late) and adult pregnant women in Iquitos, Peru. METHODS: A random sample of 4467 birth records was collected. Multivariate analyses were performed to compare prenatal care coverage in all adolescent (10-14 years, 15-19 years) and adult (>or=20 years) age groups and then for primiparous women only. RESULTS: The mean number of visits was 5.0 for adolescents aged 10-14 years, 6.1 for adolescents aged 15-19 years, and 6.2 for women aged 20 years or older. For primiparous women, the means were 5.1, 6.2, and 6.8, for the respective age groups. Both the proportion attending and the number of prenatal visits were significantly lower in primiparous adolescents aged 10-14 years than in primiparous women aged 20 years or older (aOR 0.25; 95% CI, 0.10-0.62 and aRR 0.83; 95% CI, 0.74-0.94, respectively). CONCLUSION: All women attended prenatal care more frequently than the WHO's recommended 4 visits; however, early adolescents attended significantly less often than late adolescents or adult women. Further study of this inequality is warranted to adequately inform local health services.