Research overview (IMAGE) Kyushu University Caption As outlined in this figure, researchers from Kyushu University used blood metabolome and clinical data to clarify the components characteristic to hikikomori, and machine learning models were created for the classification, prediction of the severity, and stratification of hikikomori. Credit Kyushu University/Kato’s Hikikomori Lab Usage Restrictions Only for media and non-commercial use License Original content Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.