Abstract— Innovations in digital health and machine learning are changing the path of clinical health and care. People from many different geographies and cultures can benefit from the mobility of wearable devices and smartphones to monitor their health in a ubiquitous manner. This paper focuses on reviewing the digital health and machine learning technologies used in gestational diabetes ̵̶  a subtype of diabetes that occurs during pregnancy. Despite a large number of patients with gestational diabetes, only a handful of digital health applications have been deployed in clinical practice. This paper reviews sensor technologies in blood glucose monitoring devices and machine learning fused digital health innovations for gestational diabetes monitoring and management in both clinical and commercial settings. It is one of the first comprehensive reviews in this area to the best of our knowledge. In conclusion, there is a need to (1) develop digital health technologies and clinically interpretable machine learning methods for patients with gestational diabetes, assisting health professionals with treatment monitoring and planning; (2) adapt and develop clinically proven devices for patient self-management of health and well-being at the hospital and home settings thereby facilitating timely intervention; and (3) ensure innovations are affordable and sustainable for women everywhere. Data statement: this is a review manuscript that have not generated any new data. The views expressed are those of the authors and not necessarily those of InnoHK. This research was supported by the National Institute for Health Research (NIHR) Oxford Biomedical Research Centre. The views expressed are those of the authors and not necessarily those of the NHS, the NIHR or the Department of Health.

Objective: To determine the prevalence of pre-eclampsia and fetal growth restriction (FGR; <3rd centile) in women with pre-existing cardiac dysfunction. Design: Retrospective cohort study. Setting: Maternity units in UK and Australia. Population: Pregnant women with impaired left ventricular ejection fraction<55%. Methods: Routine clinical data, including medical history and pregnancy outcome were collected retrospectively. Main Outcome Measures: Pre-specified outcomes included pre-eclampsia and FGR prevalence in women with pre-existing cardiac impairment, compared with the general population; and the relationship between pregnancy outcome and pre-pregnancy cardiac phenotype. Results: In this cohort of 282 pregnancies, pre-eclampsia prevalence was not significantly increased (4.6% [95% C.I 2.2-7.0%] versus population prevalence of 4.6% [95% C.I. 2.7-8.2], p=0.99); 12/13 of these women had additional obstetric/medical risk factors. However, prevalences of preterm pre-eclampsia (<37 weeks) and FGR were increased (1.8% versus 0.7%, p=0.03; 15.2% versus 5.5%, p<0.001, respectively). Neither systolic nor diastolic function correlated with pregnancy outcome; however, left ventricular mass index (LVMi) weakly correlated with pre-eclampsia (5g/m2 increase: OR 1.18 [95% C.I. 1.01-1.38], p=0.04). Antenatal ß blockers (n=116) were associated with lower birthweight Z score (adjusted difference -0.33 [95% C.I. -0.63- -0.02], p=0.04). Conclusions: This study demonstrated a modest increase in preterm pre-eclampsia and significant increase in FGR in women with cardiac dysfunction. These results do not support a causal relationship between cardiac dysfunction and pre-eclampsia, especially accounting for the background risk status of the population. The mechanism underpinning the relationship between cardiac dysfunction and FGR merits further research but could be influenced by concomitant ß blocker use.