CephTransX_net: An attention enhanced feature fusion network leveraging neighborhood rough set approach for...

The convergence of medical imaging, computer vision, and orthodontics has made automatic cephalometric landmark detection a pivotal area of research. Accurate cephalometric analysis is crucial in orthodontics, orthognathic and maxillofacial surgery for diagnosis, treatment planning, and monitoring craniofacial growth. In this research study, a multi-branch fused feature extraction network titled CephTransX_net is proposed to automatically predict landmark coordinates from cephalometric radiographs. The initial sequential branch enhances discriminative local feature learning and feature extraction through parallel feature fusion by integrating Convolved Pooled Normalized (CPN_B) and Gradient Optimized Multi-Path Bottleneck (GMB_B) blocks with Channel and Spatial Attention (CSAT_M) module. The Swin Transformer (ST_B) branch efficiently handles long-range dependencies and extracts global features in cephalometric radiographs. The multi-branch fused features along with features from skip connections of CPN_B and GMB_B blocks are concatenated using a Coordinate Attention module (CoAT_M) to captures the positional relationships between various landmark features. A Landmark Discriminative Deviation Factor (LDDF) is determined by applying the Neighborhood Rough Set (NRS) approach to analyse the surrounding features of each landmark by considering spatial relationships or similarity measures between the landmarks and neighboring regions. The Spatial Pyramid Pooling (SPP_L) layer incorporated in the final phase of CephTransX_net model extracts multi-scale features by pooling over sub-regions of varying sizes, enabling the network to capture both local and global context for precise cephalometric landmark identification. The CephTransX_net framework achieved an average Successful Detection Rates (SDR_s) of 88.71 % and 79.05 % in 2 mm using the 2015 International Symposium on Biomedical Imaging (ISBI) grand challenge dental X-ray analysis dataset. The effectiveness of the CephTransX_net model is evaluated using a private clinical dataset obtained from Solanki Dental Care Clinic in Sharjah, UAE, and attained an average SDR_s of 74.38 % in 2 mm precision range.