This study aimed to explore the role of artificial intelligence (AI) in predicting perinatal outcomes among women with COVID-19. Data was collected from hospitals in the Middle Euphrates and Southern regions of Iraq, with 152 pregnant patients included in the study. Patients were categorized into mild and severe infection groups, and their serum samples were analyzed for mineral levels (magnesium, copper, calcium, sodium, potassium, zinc, selenium, and iron) and immune factors (IL-6, IL-8, IL-32, IL-10, IL-18, IL-37, IL-38, IL-36, and IL-1). The findings revealed significant associations between specific mineral levels, immune factors, and perinatal outcomes. Mineral levels such as magnesium (75.5% mild infection, 80.9% severe infection), copper (68.2% mild infection, 64.3% severe infection), calcium ion (81.8% mild infection, 76.2% severe infection), sodium (70.9% mild infection, 69.0% severe infection), potassium (72.7% mild infection, 71.4% severe infection), zinc (61.8% mild infection, 54.8% severe infection), selenium (78.2% mild infection, 82.9% severe infection), and iron (74.5% mild infection, 68.3% severe infection) showed varying percentages associated with mild and severe infections. Immune factors such as IL-6 (32% mild infection, 21% severe infection), IL-8 (15% mild infection, 7% severe infection), IL-32 (24% mild infection, 9% severe infection), IL-10 (7% mild infection, no severe infection), IL-18 (13% mild infection, 11% severe infection) demonstrated varying percentages associated with perinatal outcomes, while other interleukins showed no changes in severe infections. These results highlight the potential of AI in predicting outcomes for pregnant women with COVID-19, which could aid in improving their management and care. Further research and validation of predictive models are recommended to enhance accuracy and applicability.