Background: This study aimed to evaluate the clinical impact of an artificial intelligence (Al)-based decision support system (DSS), Koios DS, on the analysis of ultrasound imaging and suspicious characteristics for thyroid nodule risk stratification Methods: A retrospective ultrasound study was conducted on all thyroid nodules with histological findings from June 2021 to December 2022 in a thyroid nodule clinic. The diagnostic performance of ultrasound imaging was evaluated by six readers on American College of Radiology Thyroid Imaging Reporting and Data System (ACR TI-RADS) before and after the use of the AI-based DSS, and by the AI itself. Results: A total of 172 patients (83.1% women) with a mean age of 52.3±15.3 years were evaluated. The mean maximum nodular diameter was 2.9±1.2 cm, with 11.0% being differentiated thyroid carcinomas. Among the nodules initially classified as ACR TI-RADS 3 and 4, Al reclassified 81.4% and 24.5% into lower risk categories, respectively.Receiver operating characteristic (ROC) curve analysis was performed to evaluate the diagnostic performance of the readers and the AI-based DSS versus histological diagnosis. There was an increase in the area under the receiver operating characteristic curve (AUROC) after the use of Al (0.776 vs. 0.817, p<0.001). The AI-based DSS improved the mean sensitivity (82.3% vs. 86.5%) and specificity (38.3% vs. 54.8%), produced a high negative predictive value (94.5% vs. 96.4%), and increased the positive predictive value (14.0% vs. 16.1%) and diagnostic precision (43.0% vs. 49.3%). Based on the ACR TI-RADS score, there was a significant improvement in interobserver agreement after the use of Al (r=0.741 for ultrasound imaging alone vs. 0.981 for ultrasound imaging and the AI-based DSS, p<0.001). Conclusion: The use of an AI-based DSS was associated with an overall improvement in the diagnostic efficacy of ultrasound imaging, based on the AUROC, as well as an increase in sensitivity, specificity, the negative and positive predictive value, and diagnostic accuracy. There was also a reduction in interobserver variability and an increase in the degree of concordance with the use of Al. Al reclassified more than half of the nodules with intermediate ACR TI-RADS into lower risk categories.

Objective: To evaluate the impact of glucose variability on the relationship between the GRI and other glycemic metrics in a cohort of pediatric and adult patients with type 1 diabetes (T1D) using intermittent scanning continuous glucose monitoring (isCGM). Methods: We performed a cross-sectional study of 202 patients with T1D under intensive insulin treatment (25.2% CSII) using isCGM. Clinical, metabolic, and glycemic metrics were collected, and the GRI was calculated with its hypoglycemia (CHypo) and hyperglycemia (CHyper) components. The correlation between the GRI and other classical glycometrics in relation to the coefficient of variation (CV) was evaluated. Results: A total of 202 patients were included (53% male; 67.8% adults) with a mean age of 28.6 ± 15.7 years and 12.5 ± 10.9 years of T1D evolution (TIR 59.0 ± 17.0%; CV 39.8 ± 8.0%; GMI 7.3 ± 1.1%). The mean GRI was 54.0 ± 23.3 with a CHypo and CHyper component of 5.7 ± 4.8 and 23.4 ± 14.3, respectively. A strong negative correlation was observed between the GRI and TIR (R = −0.917; R² = 0.840; p < 0.001), showing differences when dividing patients with low glycemic variability (CV < 36%) (R = −0.974; R² = 0.948; p < 0.001) compared to those with greater CV instability (≥36%) (R = −0.885; R² = 0.784; p < 0.001). The relationship of GRI with its two components was strongly positive with CHyper (R = 0.801; R² = 0.641; p < 0.001) and moderately positive with CHypo (R = 0.398; R² = 0.158; p < 0.001). When the GRI was evaluated with the rest of the classic glycemic metrics, a strong positive correlation was observed with HbA1c (R = 0.617; R² = 0.380; p < 0.001), mean glucose (R = 0.677; R² = 0.458; p < 0.001), glucose standard deviation (R = 0.778; R² = 0.605; p < 0.001), TAR > 250 (R = 0.801; R² = 0.641; p < 0.001), and TBR < 54 (R = 0.481; R² = 0.231; p < 0.001). Conclusions: The GRI correlated significantly with all the glycemic metrics analyzed, especially with the TIR. Glycemic variability (GV) significantly affected the correlation of the GRI with other parameters and should be taken into consideration.

Aims: To compare glycemic control and maternal–fetal outcomes of women with type 1 diabetes (T1D) using hybrid closed loop (HCL) vs. multiple daily insulin injections (MDI) plus continuous glucose monitoring (CGM). Methods: Multicenter prospective cohort study of pregnant women with T1D in Spain. We evaluated HbA1c and time spent within (TIR), below (TBR) and above (TAR) the pregnancy-specific glucose range 3.5–7.8 mmol/L. Adjusted models were performed for adverse pregnancy outcomes including baseline maternal characteristics and center. Results: 112 women were included (HCL n=59). Women in the HCL group had a longer duration of diabetes and higher rates of prepregnancy care. There were no between-group differences in HbA1c in any trimester. However, in the second trimester, MDI users had a greater decrease in HbA1c (-6.12±9.06 vs. -2.16 ±7.42 mmol/mol, p=0.031). No differences in TIR (3.5-7.8 mmol/L) and TAR were observed between HCL and MDI users, but with a higher total insulin dose in the second trimester (+0.13 IU/Kg/d). HCL therapy was associated with increased maternal weight gain during pregnancy (βadjusted 3.20 kg, 95%CI 0.90-5.50). Regarding neonatal outcomes, newborns of HCL users were more likely to have higher birthweight (βadjusted 279.0 g, 95% CI 39.5-518.5) and macrosomia (ORadjusted 3.18, 95% CI 1.05-9.67) compared to MDI users. These associations disappeared when maternal weight gain or third trimester HbA1c were included in the models. Conclusions: In a real-world setting, HCL users gained more weight during pregnancy and had larger newborns than MDI users, while achieving similar glycemic control in terms of HbA1c and TIR.

Vitamin D nonresponsive hypoparathyroidism is uncommonly seen in the clinical practice. The use of new treatment modalities, including teriparatide administration, provides an alternative requiring its validation. We report the first case of hypoparathyroidism refractory to vitamin D that was successfully controlled for 5 years by teriparatide treatment. A 53-year-old woman presented severe hypoparathyroidism after thyroidectomy. No therapeutic response was obtained with oral and i. v. calcium and magnesium, or even with 5 μg calcitriol/day. Digestive disorders were ruled out and heterologous parathyroid transplant was ineffective. Subcutaneous (s. c.) recombinant human PTH 1–34 (rhPTH-teriparatide) plus oral calcitriol, calcium, and magnesium, were partially effective, but effectiveness of 20 μg teriparatide lasted less than 4 h and stable calcemia was not possible even with 4–6 injections/day. Multipulse s. c. pump driven infusion of teriparatide achieved complete normalization of serum calcium, phosphate, magnesium, calciuria, and magnesuria with relatively low teriparatide doses (25–35 μg/day) after the first day of treatment. Effectiveness of this treatment modality was maintained for 5 years. The only significant side effect observed during these years was the development of subcutaneous nodules with occasional insufficient control of calcemia. A gain in bone mineral density was observed after the first year of treatment, which had remained stable and within normal values, thereafter until now. No abnormalities in bone scintigraphy were detected in the follow-up. Vitamin D unresponsive hypoparathyroidism maybe safely and effectively controlled at long term by s. c. multipulse pump treatment recombinant human PTH.