Bottom-up and top-down paradigms of artificial intelligence research approaches to healthcare data science using growing real-world big data.

OBJECTIVES: As the real-world electronic health record (EHR) data continue to grow exponentially, novel methodologies involving artificial intelligence (AI) are becoming increasingly applied to enable efficient data-driven learning and, ultimately, to advance healthcare. Our objective is to provide readers with an understanding of evolving computational methods and help in deciding on methods to pursue. TARGET AUDIENCE: The sheer diversity of existing methods presents a challenge for health scientists who are beginning to apply computational methods to their research. Therefore, this tutorial is aimed at scientists working with EHR data who are early entrants into the field of applying AI methodologies. SCOPE: This manuscript describes the diverse and growing AI research approaches in healthcare data science and categorizes them into 2 distinct paradigms, the bottom-up and top-down paradigms to provide health scientists venturing into artificial intelligent research with an understanding of the evolving computational methods and help in deciding on methods to pursue through the lens of real-world healthcare data.

Obstetric Hemorrhage Risk Associated with Novel COVID-19 Diagnosis from a Single-Institution Cohort in the United States.

OBJECTIVE: The study aimed to compare the quantitative blood loss (QBL) and hemorrhage-related outcomes of pregnant women with and without a coronavirus disease 2019 (COVID-19) diagnosis. STUDY DESIGN: This retrospective cohort study of all live deliveries at Boston Medical Center between April 1, 2020 and July 22, 2020 compares the outcomes of pregnant women with a laboratory-confirmed COVID-19 positive diagnosis and pregnant women without COVID-19. The primary outcomes are QBL and obstetric hemorrhage. The secondary outcomes analyzed were a maternal composite outcome that consisted of obstetric hemorrhage, telemetry-level (intermediate care unit) or intensive care unit, transfusion, length of stay greater than 5 days, or intraamniotic infection, and individual components of the maternal composite outcome. Groups were compared using Student's t-test, Chi-squared tests, or Fisher's exact. Logistic regression was used to adjust for confounding variables. RESULTS: Of 813 women who delivered a live infant between April 1 and July 22, 2020, 53 women were diagnosed with COVID-19 on admission to the hospital. Women with a COVID-19 diagnosis at their time of delivery were significantly more likely to identify as a race other than white (p = 0.01), to deliver preterm (p = 0.05), to be diagnosed with preeclampsia with severe features (p < 0.01), and to require general anesthesia (p < 0.01). Women diagnosed with COVID-19 did not have a significantly higher QBL (p = 0.64). COVID-19 positive pregnant patients had no increased adjusted odds of obstetric hemorrhage (adjusted odds ratio [aOR]: 0.41, 95% confidence interval [CI]: 0.17-1.04) and no increased adjusted odds of the maternal morbidity composite (aOR: 0.98, 95% CI: 0.50-1.93) when compared with those without a diagnosis of COVID-19. CONCLUSION: Pregnant women with COVID-19 diagnosis do not have increased risk for obstetric hemorrhage, increased QBL or risk of maternal morbidity compared with pregnant women without a COVID-19 diagnosis. Further research is needed to describe the impact of a COVID-19 diagnosis on maternal hematologic physiology and pregnancy outcomes. KEY POINTS: · Information about blood loss associated with peripartum COVID-19 is limited.. · COVID-19 diagnosis is not associated with increase in obstetric hemorrhage.. · COVID-19 diagnosis is not associated with increase in blood loss..