MIDO GDM: an innovative artificial intelligence-based prediction model for the development of gestational diabetes in Mexican women

Given the barriers to early detection of gestational diabetes mellitus (GDM), this study aimed to develop an artificial intelligence (AI)-based prediction model for GDM in pregnant Mexican women. Data were retrieved from 1709 pregnant women who participated in the multicenter prospective cohort study 'Cuido mi embarazo'. A machine-learning-driven method was used to select the best predictive variables for GDM risk: age, family history of type 2 diabetes, previous diagnosis of hypertension, pregestational body mass index, gestational week, parity, birth weight of last child, and random capillary glucose. An artificial neural network approach was then used to build the model, which achieved a high level of accuracy (70.3%) and sensitivity (83.3%) for identifying women at high risk of developing GDM. This AI-based model will be applied throughout Mexico to improve the timing and quality of GDM interventions. Given the ease of obtaining the model variables, this model is expected to be clinically strategic, allowing prioritization of preventative treatment and promising a paradigm shift in prevention and primary healthcare during pregnancy. This AI model uses variables that are easily collected to identify pregnant women at risk of developing GDM with a high level of accuracy and precision.

Diagnostic Accuracy of Capillary Blood Glucometer Testing for Gestational Diabetes

PURPOSE

Few pregnant women in low-resource settings are screened for gestational diabetes mellitus (GDM) using the gold standard oral glucose tolerance test (OGTT). This study compared capillary blood glucose testing with 2-h plasma glucose measurements obtained using the 75-g OGTT to screen for GDM at primary healthcare clinics in Mexico.

PATIENTS AND METHODS

Pregnant women who participated in a previous prospective multicenter longitudinal cohort study and who had not been previously diagnosed with diabetes were included. Participants were evaluated using the plasmatic 2-h 75-g OGTT with simultaneous capillary blood glucose measurements using a glucometer. The study endpoint was the comparability of the glucometer results to the gold standard OGTT when collected simultaneously. Sensitivity, specificity, and area under the curve of the glucose measurements obtained for capillary blood compared with venous plasma (gold standard) were calculated to determine diagnostic accuracy.

RESULTS

The study included 947 pregnant women who had simultaneous glucose measurements available (blood capillary [glucometer] and venous blood OGTT). Overall, capillary blood glucose testing was very sensitive (89.47%); the specificity was 66.58% and the area under the curve (95% confidence interval) was 0.78 (0.74-0.81). The sensitivity, specificity and area under the curve of each capillary measurement were: 89.47%, 66.58% and 0.78 (0.74-0.82) for the fasting measurement, 91.53%, 93.24% and 0.92 (0.88-0.96) for the one-hour measurement, and 89.80%, 93.32%, 0.91 (0.87-0.95) for the second-hour measurement, respectively. No adverse events were reported.

CONCLUSION

Capillary OGTT is a valid alternative to the gold standard OGTT for screening of GDM in low-resource situations or in situations where there are other limitations to performing the OGTT as part of primary healthcare services.