Human Gender Prediction Based on Deep Transfer Learning from Panoramic Dental Radiograph Images

Panoramic Dental Radiography (PDR) image processing is one of the most extensively used manual methods for gender determination in forensic medicine. With assistance PDR images, a person's biological can be performed through analyzing skeletal structures expressing sexual dimorphism. Manual approaches require wide range mandibular parameter measurements metric units. Besides being time-consuming, these also necessitate employment experienced professionals. In this context, deep learning models are widely utilized auto-analysis radiological images nowadays, owing to their high speed, accuracy, and stability. our study, data set consisting 24,000 dental panoramic was prepared binary classification, transfer method accelerate training increase performance proposed DenseNet121 model. method, instead starting process from scratch, existing patterns learned beforehand were used. Extensive comparisons made using (DTL) VGG16, ResNet50, EfficientNetB6 assess classification model images. According findings comparative analysis, outperformed other by achieving success rate 97.25% classification.