Background: Three-dimensional facial soft tissue landmark prediction is an important tool in dentistry, for which several methods have been developed in recent years, including a deep learning algorithm which relies on converting 3D models into 2D maps, which results in the loss of information and precision. Methods: This study proposes a neural network architecture capable of directly predicting landmarks from a 3D facial soft tissue model. Firstly, the range of each organ is obtained by an object detection network. Secondly, the prediction networks obtain landmarks from the 3D models of different organs. Results: The mean error of this method in local experiments is $2.62\pm 2.39$, which is lower than that in other machine learning algorithms or geometric information algorithms. Additionally, over 72% of the mean error of test data falls within $\pm 2.5$ mm, and 100% falls within 3 mm. Moreover, this method can predict 32 landmarks, which is higher than any other machine learning-based algorithm. Conclusions: According to the results, the proposed method can precisely predict a large number of 3D facial soft tissue landmarks, which gives the feasibility of directly using 3D models for prediction.

Improper orthodontic treatment can disrupt alveolar bone remodeling, leading to excessive alveolar bone resorption and complications such as bone fenestration and dehiscence. Although ferroptosis has emerged as an important regulator of bone metabolism, its role in mechanical force-induced periodontal tissue remodeling during orthodontic tooth movement (OTM) remains incompletely understood. This study investigated whether ferroptosis regulates the osteogenic response of human periodontal ligament stem cells (hPDLSCs) under compressive force (CF), focusing on stearoyl-CoA desaturase 1 (SCD1). hPDLSCs were subjected to static CF, and RNA sequencing, RT–qPCR, Western blotting, and immunofluorescence were performed to analyze SREBP1 and SCD1 expression. Ferroptosis-related alterations were evaluated using biochemical and mitochondrial analyses. Ferrostatin-1 (Fer-1) rescue, SCD1 knockdown/overexpression, and a rat OTM model were used for functional validation. Prolonged CF downregulated SREBP1 and SCD1 expression, accompanied by ferroptosis-associated alterations and reduced RUNX2 and OPN expression. Fer-1 alleviated these changes, whereas SCD1 overexpression partially restored osteogenic markers and SCD1 knockdown further impaired osteogenic responses. In vivo, SREBP1 and SCD1 expression were reduced on the compression side. Collectively, these findings identify SCD1 as an important regulator linking CF-induced ferroptosis-associated alterations with impaired osteogenic differentiation of hPDLSCs, suggesting SCD1 as a potential target for orthodontic bone remodeling.