Gradient-boosting threshold identification for developmental toxicity: A BPAF case-study framework for early-stage risk assessment.

A BPAF case-study framework identifies several environmental concentrations as preliminary risk flags using machine learning and spatial analytics, highlighting early-stage toxicity concerns.

• Why it matters: Limited toxicity data for BPA analogues hampers effective risk assessment and regulatory decision-making, emphasizing the need for innovative early-stage evaluation methods.
• What they did: Researchers developed an integrated assessment combining machine learning-derived benchmarks with spatial analytics to derive a screening-level reference dose and evaluate exposure patterns, analyzing global monitoring data.
• The result: Findings suggest certain sediment and water concentrations in China and Europe may pose preliminary risks, underscoring the importance of targeted monitoring and improved chemical oversight to protect environmental health.