Brief

Researchers have developed a novel context-aware hierarchical Bayesian model to improve IVF pregnancy rate predictions by incorporating laboratory environmental data. This model engineers 55 temporal features, such as thermal stability and humidity adherence, to capture incubator microenvironment dynamics. When applied to data from an Asian IVF clinic, these features reduced prediction error to 1.27%. The model also demonstrated its ability to share environmental effects across clinics, achieving an R2 of 0.86 and a 64% error reduction for a specific age group in a Northern European clinic. AI

Researchers have developed an attention-guided deep learning framework to improve the interpretability and accuracy of sperm morphology classification. By integrating a pre-trained EfficientNet-B0 model with a Convolutional Block Attention Module (CBAM), the system effectively focuses on critical sperm head features. This approach achieved high accuracy rates of 90.2% and 93.9% on public datasets, surpassing simpler models and providing visual explanations for its classifications. AI