Scientists have developed an innovative AI tool to help healthcare providers identify patients at risk of intimate partner violence (IPV), potentially years before they seek assistance. This machine learning model was trained on data from nearly 850 IPV survivors and over 5,200 control patients during regular hospital visits, as reported in the journal Nature. IPV, which can lead to severe injuries and mental health disorders, affected 18% of women according to a European Commission report. Current screening methods often fail due to fear and stigma surrounding disclosure. The AI systems analyzed structured hospital data, written notes, and combined information, achieving an 88% accuracy rate in risk identification. The tool allows for earlier intervention, flagging potential abuse up to three years before patients enter domestic abuse programs. Researchers aim to integrate this decision-support system into electronic medical records, enhancing real-time assessments for better public health outcomes.
Source link
Share
Read more