Estimation of the Development of Depression and PTSD in Children Exposed to Sexual Abuse and Development of Decision Support Systems by Using Artificial Intelligence

UCUZ İ., ARI A., ÖZCAN Ö., TOPAKTAŞ Ö., Sarraf M., DOĞAN Ö.

JOURNAL OF CHILD SEXUAL ABUSE, cilt.31, sa.1, ss.73-85, 2022 (SSCI) identifier identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 31 Sayı: 1
  • Basım Tarihi: 2022
  • Doi Numarası: 10.1080/10538712.2020.1841350
  • Dergi Adı: JOURNAL OF CHILD SEXUAL ABUSE
  • Derginin Tarandığı İndeksler: Social Sciences Citation Index (SSCI), Scopus, Academic Search Premier, ASSIA, IBZ Online, Child Development & Adolescent Studies, CINAHL, Criminal Justice Abstracts, EBSCO Education Source, Educational research abstracts (ERA), EMBASE, Gender Studies Database, MEDLINE, Psycinfo, Social services abstracts, Sociological abstracts, Violence & Abuse Abstracts
  • Sayfa Sayıları: ss.73-85
  • Anahtar Kelimeler: Childhood sexual abuse, machine learning, artificial neural networks, depression, post-traumatic stress disorder, PSYCHIATRIC-DISORDERS, PHYSICAL ABUSE, DISCLOSURE, MALTREATMENT, BIOMARKERS, PSYCHOPATHOLOGY, VICTIMIZATION, OPPORTUNITIES, ADOLESCENTS, PREDICTORS
  • İnönü Üniversitesi Adresli: Evet

Özet

The most common diagnoses after childhood sexual abuse are Post-Traumatic Stress Disorder and depression. The aim of this study is to design a decision support system to help psychiatry physicians in the treatment of childhood sexual abuse. Computer aided decision support system (CADSS) based on ANN, which predicts the development of PTSD and Major Depressive Disorder, using different parameters of the act of abuse and patients was designed. The data of 149 girls and 21 boys who were victims of sexual abuse were included in the study. In the designed CADDS, the gender of the victim, the type of sexual abuse, the age of exposure, the duration until reporting, the time of abuse, the proximity of the abuser to the victim, number of sexual abuse, whether the child is exposed to threats and violence during the abuse, the person who reported the event, and the intelligence level of the victim are used as input parameters. The average accuracy values for all three designed systems were calculated as 99.2%. It has been shown that the system designed by using these data can be used safely in the psychiatric assessment process, in order to differentiate psychiatric diagnoses in the early post-abuse period.