In implant prosthesis treatment, the surgical guide of implant is used to ensure accurate implantation. However, such design heavily relies on the manual location of the implant position. When deep neural network has been proposed to assist the dentist in locating the implant position, most of them take a single slice as input, which do not fully explore 3D contextual information and ignores the influence of implant slope. In this paper, we design a Text Guided 3D Context and Slope Aware Triple Network (TCSloT) to integrate the perception of contextual information from multiple adjacent slices and awareness of variation of implant slopes. A Texture Variation Perception (TVP) module is correspondingly design to process the multiple slices and capture the texture variation among slices and a Slope-Aware Loss (SAL) is proposed to dynamically assign adaptive weights for the regression head. Additionally, we design a conditional text guidance (CTG) module to integrate the text condition (i.e., left, middle and right) from the CLIP to assist the implant position prediction. Extensive experiments on a dental implant dataset through five-fold cross-validation, demonstrated that the proposed TCSloT achieves superior performance than existing methods.

For cyclic LDPC codes, we propose to use their automorphism groups to improve the iterative decoding performance. The basic idea is to construct nonequivalent parity-check matrices via column permutations. Three types of iterative decoders are devised to take advantage of the code's automorphism group. In this paper we focus on cyclic LDPC codes defined by a circulant parity-check matrix and consider two known subgroups of the automorphism group of a cyclic code. For the large class of idempotent-based cyclic LDPC codes in the literature, we show that the two subgroups only provide equivalent parity-check matrices and thus cannot be harnessed for iterative decoding. Towards exploiting the automorphism group of a code, we propose a new class of cyclic LDPC codes based on pseudo-cyclic MDS codes with two information symbols, for which nonequivalent parity-check matrices are obtained. Simulation results show that for our constructed codes of short lengths, the automorphism group can significantly enhance the iterative decoding performance.

Based jointly on idempotents and modular Golomb rulers, we construct a class of nonbinary cyclic low-density parity-check (LDPC) codes. The defining parity-check matrix is a sparse circulant, on which we put two constraints: 1) the characteristic polynomial is an idempotent, 2) the nonzero elements of the first row are located on a modular Golomb ruler. We show that the second constraint forms a necessary and sufficient condition for the Tanner graph to have no cycles of length 4. The minimum distance of the code is proved equal to the column weight of the parity-check matrix plus one. A search algorithm is presented, with which we obtain some high rate codes with large minimum distances. The issue of code equivalence is also discussed. Simulation results show that the obtained codes perform well under iterative decoding.