No abstract

PurposeGender-based violence (GBV) has negative impact on the health and well-being of the survivors. The initial response can lessen the magnitude of the effect, and thus the health care sector's readiness to respond to GBV cases is important. This study aimed to explore the knowledge, attitude and its association to practice regarding GBV response among health care personnel.Design/methodology/approachA cross-sectional study was conducted in 48 public hospitals in Yangon, Myanmar, involving 398 participants including doctors and nurses, by a self-administered structured questionnaire.FindingsAmong 398 participants, most of the participants had moderate level of knowledge, attitude and supportive environmental factors. Only 12.8% had experienced GBV management. The respondents with a bachelor's degree were more likely to practice on health care management towards GBV cases than those who finished with a diploma degree. The medical officers and senior medical officers were more likely to practice than the junior nurses. The participants who had already attended the trainings had more practice than those who had not. Those who work in the regional hospitals were less likely to practice than those who work in the station-level hospital.Originality/valueThis paper explored the associated factors to health care personnel's practice of health care management towards GBV survivors in Myanmar which contains information about knowledge, attitude and supportive environmental factors. The study results can be used for a supportive data for health system strengthening the response of GBV cases in Myanmar.

No abstract

No abstract

Facial expression is the most challenging task in the field of computer vision. In this paper, an automatic facial expression recognition system from a still frontal posed image is presented. This system recognizes the human expression by observing the shape of the mouth. This paper uses color based segmentation followed by template matching for face detection and localization. For mouth segmentation, Canny_Template method is used. Orientation Histogram is used for feature extraction. Feed forward neural network is used as a classifier for classifying the expressions of supplied face into five basic expressions like surprise, neutral, sad, happy and angry. Experiments are carried out on Myanmar Facial Expression Database and give the correct performance in terms of 100% accuracy for training set and 70.71% accuracy for test set.