Artificial Intelligence Systems for Identifying Sexual Abuse through Self-Figure Drawing
Rachel Lev-Wiesel & Ben Zaneti
Childhood sexual abuse is a worldwide prevalent phenomenon that has negative long term consequences for the victims their families and became a high economic toll on society. One of the main difficulties in coping with CSA is the reluctance of victims to disclose and the failure of professionals to detect it when there are usually no forensic evidence. Aiming to assist professionals in detecting CSA, an attempt to develop an artificial intelligence system that automatically and accurately will be able to detect childhood sexual abuse (CSA) experience out of self-figure drawings, was made. A data set of sexually abused versus unknown sexually abused consisting of 1377 self-figure drawings images: 771 abuse, 606 non-abuse, was curated.
Preliminary results showed that the system accurately detected 83% among the abused group and 76% as non-abused from the unknown sexually abused. These are results are very promising. Further research steps will be discussed.
Prof. Rachel Lev-Wizel. School of Creative Arts Therapies; Head, Emili Sagol Research Center. Research Areas: Trauma; Childhood sexual abuse; child abuse; Domestic violence; Holocaust; Use of drawings for diagnostic and therapeutic purposes.
sss