Landmark Localization Research Articles

Three-dimensional mapping of the greater palatine artery location and physiology.

To develop a novel technique for localizing and reconstructing the greater palatine artery (GPA) using three-dimensional (3D) technology. A miniaturized intraoral ultrasound transducer was used to imaging landmarks including the GPA, gingival margin (GM), and palatal masticatory mucosa (PMM). A 5-mm-thick solid hydrogel couplant was integrated to replace traditional ultrasound gel and avoid bubbles when moving the transducer. A panorama image provided the relative localization of landmarks including the GPA, PMM, and hard palate. Short- and long-axis imaging of GPA was performed in five subjects including 3D mapping of GPA branches and surrounding tissues in a volume of 10 mm × 8 mm × 10 mm. Full-mouth Doppler imaging was also demonstrated on both the dorsal and ventral tongue as well as buccal mucosa and sublingual region on two subjects. This study can measure the vertical distance from the GM to the GPA and depth from PMM to GPA and visualize the GPA localization in a 3D manner, which is critical to evaluate the available volume of palatal donor tissues and avoid sectioning of GPA during surgical harvesting of the tissues. Finally, the transducer's small size facilitates full-mouth Doppler imaging with the potential to improve the assessment, diagnosis, and management of oral mucosa.

FetusMapV2: Enhanced fetal pose estimation in 3D ultrasound

Fetal pose estimation in 3D ultrasound (US) involves identifying a set of associated fetal anatomical landmarks. Its primary objective is to provide comprehensive information about the fetus through landmark connections, thus benefiting various critical applications, such as biometric measurements, plane localization, and fetal movement monitoring. However, accurately estimating the 3D fetal pose in US volume has several challenges, including poor image quality, limited GPU memory for tackling high dimensional data, symmetrical or ambiguous anatomical structures, and considerable variations in fetal poses. In this study, we propose a novel 3D fetal pose estimation framework (called FetusMapV2) to overcome the above challenges. Our contribution is three-fold. First, we propose a heuristic scheme that explores the complementary network structure-unconstrained and activation-unreserved GPU memory management approaches, which can enlarge the input image resolution for better results under limited GPU memory. Second, we design a novel Pair Loss to mitigate confusion caused by symmetrical and similar anatomical structures. It separates the hidden classification task from the landmark localization task and thus progressively eases model learning. Last, we propose a shape priors-based self-supervised learning by selecting the relatively stable landmarks to refine the pose online. Extensive experiments and diverse applications on a large-scale fetal US dataset including 1000 volumes with 22 landmarks per volume demonstrate that our method outperforms other strong competitors.

Surface anatomical landmarks for spine surgery: A CT-based study of the sternal notch and sternal angle in 1,035 patients

BackgroundUnderstanding the location of surface anatomical landmarks in relation to the cervical and thoracic spine is important for a wide array of clinical applications. The objective of this study was to investigate the influence of patient demographic and anthropometric characteristics on the locations of the sternal notch and sternal angle in relation to the spine using computed tomography (CT) of a large cohort of young adult patients without spinal pathology. MethodsVertebral levels corresponding to the sternal notch and sternal angle were analyzed using CT of 1,035 patients. Influences of patient height, weight, body mass index (BMI), sex, and ethnicity were assessed. Results567 male and 468 female patients were included in this study. Mean patient height, weight, BMI, and age were 1.68 ± 0.11 m, 81.94 ± 24.39 kg, 27.79 ± 7.9 kg/m2, and 25.9 ± 5.9 years. Of the 1,035 patients, 495 were Hispanic or Latino, 321 were Caucasian, 130 were African American, 68 were Asian, 5 were identified as “other,” and 16 did not have racial or ethnic data available. The location of the sternal notch in relation to the thoracic spine demonstrated a bimodal distribution pattern clustered at the T2 and T3 vertebral bodies. The location of the sternal angle in relation to the thoracic spine demonstrated a bimodal distribution pattern clustered at the T4 and T5 vertebral bodies. Moderate, negative correlations were observed between patient weight and location of the sternal notch (r = −0.447; p <.001) and sternal angle (r = −0.499; p <.001), respectively. Zero significant correlations were observed between patient height and location of the sternal notch (r = −0.045; p =.377) or sternal angle (r = −0.080; p =.229). A weak, negative correlation was observed between patient BMI and location of the sternal notch (r = −0.378; p <.001). A moderate, negative correlation was observed between patient BMI and location of the sternal angle (r = −0.445; p <.001). The locations of the sternal landmarks did not differ based on patient sex, race or ethnicity. ConclusionsThe location of the sternal notch most frequently corresponds to the T2 or T3 vertebral body, while the sternal angle is most frequently located at the T4 or T5 vertebral body. Increased patient weight is associated with relatively cephalad sternal landmarks. Patient height, sex, and ethnicity are not associated with either sternal landmark.

A novel artificial intelligence model for fetal facial profile marker measurement during the first trimester

BackgroundTo study the validity of an artificial intelligence (AI) model for measuring fetal facial profile markers, and to evaluate the clinical value of the AI model for identifying fetal abnormalities during the first trimester.MethodsThis retrospective study used two-dimensional mid-sagittal fetal profile images taken during singleton pregnancies at 11–13+ 6 weeks of gestation. We measured the facial profile markers, including inferior facial angle (IFA), maxilla-nasion-mandible (MNM) angle, facial-maxillary angle (FMA), frontal space (FS) distance, and profile line (PL) distance using AI and manual measurements. Semantic segmentation and landmark localization were used to develop an AI model to measure the selected markers and evaluate the diagnostic value for fetal abnormalities. The consistency between AI and manual measurements was compared using intraclass correlation coefficients (ICC). The diagnostic value of facial markers measured using the AI model during fetal abnormality screening was evaluated using receiver operating characteristic (ROC) curves.ResultsA total of 2372 normal fetuses and 37 with abnormalities were observed, including 18 with trisomy 21, 7 with trisomy 18, and 12 with CLP. Among them, 1872 normal fetuses were used for AI model training and validation, and the remaining 500 normal fetuses and all fetuses with abnormalities were used for clinical testing. The ICCs (95%CI) of the IFA, MNM angle, FMA, FS distance, and PL distance between the AI and manual measurement for the 500 normal fetuses were 0.812 (0.780–0.840), 0.760 (0.720–0.795), 0.766 (0.727-0.800), 0.807 (0.775–0.836), and 0.798 (0.764–0.828), respectively. IFA clinically significantly identified trisomy 21 and trisomy 18, with areas under the ROC curve (AUC) of 0.686 (95%CI, 0.585–0.788) and 0.729 (95%CI, 0.621–0.837), respectively. FMA effectively predicted trisomy 18, with an AUC of 0.904 (95%CI, 0.842–0.966). MNM angle and FS distance exhibited good predictive value in CLP, with AUCs of 0.738 (95%CI, 0.573–0.902) and 0.677 (95%CI, 0.494–0.859), respectively.ConclusionsThe consistency of fetal facial profile marker measurements between the AI and manual measurement was good during the first trimester. The AI model is a convenient and effective tool for the early screen for fetal trisomy 21, trisomy 18, and CLP, which can be generalized to first-trimester scanning (FTS).

Towards Automated Dosimetric Analysis of the Bladder Trigone: Deep-Learning-Based Joint Segmentation and Landmark Localization

The bladder trigone dosimetry is hypothesized to have a stronger correlation with post-SBRT urinary toxicity than that of the entire bladder. However, the trigone tends to move significantly between simulation and daily treatment. Its small size, large daily motion, and proximity to the target lead to potentially consequential but unaccounted-for dosimetric uncertainties. Manual segmentation of the structure can be inconsistent and time-consuming, even with MR-guided RT. Here, we propose and demonstrate a deep-learning-based framework for joint segmentation and landmark localization to support deformable registration and comprehensive dosimetric analysis. A total of 30 patients were randomly selected for training, and 20 were held out for testing. Each patient had 1 simulation and 5 daily pre-treatment images obtained from a clinical 0.35T MR Linac. The trigone is defined as the triangular bladder section among three landmarks (2 ureteral orifices and the internal urethral orifice). In the manual contouring process, the 3 landmarks were identified first, followed by trigone segmentation. The proposed joint method uses a modified 3D nnU-Net with 2 decoders, one for segmentation and the other for landmark localization. The shared encoder is expected to extract features useful for both tasks. The joint framework was compared with a baseline method using two separate 3D nnU-Nets for landmark localization and trigone segmentation, respectively. Since the trigone is small (∼2% of the bladder volume), we further experimented with a second-stage prediction mimicking the human contouring process. The predicted landmarks from the first stage were used to crop the trigone region, and a second network was trained on cropped images. Evaluation metrics included the Dice score, 95% Hausdorff distance (HD95), and average surface distance (ASD) for segmentation, and Euclidean distance (ED) between the predicted and ground truth landmarks for localization. The quantification metrics are summarized in the table below. The joint approach shows similar Dice performance to the baseline method but markedly better HD95 by 13%. For landmark localization, the proposed method is similar to the baseline, but the integration of the segmentation task stabilizes the training process. The two-stage approach further improves HD95, ASD, and ED by 27%, 24%, and 19%. Combining segmentation and landmark localization exhibits a synergistic effect. The proposed two-stage approach provided additional improvement. Future studies will explore the deformable registration of the trigone based on the segmentation and landmark detection, as well as analyze cumulated dose distribution.

Repeatability and reproducibility of landmark localization on panoramic images for PA (Posteroanterior) cephalometric analysis

Repeatability and reproducibility of landmark localization on panoramic images for PA (Posteroanterior) cephalometric analysis

Open-Source 3D Morphing Software for Facial Plastic Surgery and Facial Landmark Detection Research and Open Access Face Data Set Based on Deep Learning (Artificial Intelligence) Generated Synthetic 3D Models.

Background: The scarcity of 3D facial models presents a significant hurdle for researchers and educators. Gathering such data demands substantial resources. Objective: To introduce an open-source 3D morphing software to generate 3D facial data sets for research and to provide a large sample data set that is based on synthetically generated 3D models. Methods: Software is developed to morph 3D facial models in bulk by altering landmark locations. Twenty synthetic 3D facial models are generated utilizing deep learning tools and 28 landmarks located on each. The measurements of synthetic models are confirmed to be realistic by comparing them with facial statistics. Several facial deformities and types are simulated at various magnitudes on 3D models to generate a large data set. Results: An open-source software and an open-access data set of 980 3D facial models, each with 28 landmark locations, are provided. Since the data set is based on synthetically generated 3D models, no institutional review board approval is required. Conclusion: The 3D morphing software and the large 3D data set are expected to benefit researchers and educators in the field of facial surgery and facial landmark detection.

Learning Spatial Configuration Feature for Landmark Localization in Hand X-rays

Medical landmark localization is crucial for treatment planning. Although FCN-based heatmap regression methods have made significant progress, there is a lack of FCN-based research focused on features that can learn spatial configuration between medical landmarks, notwithstanding the well-structured patterns of these landmarks. In this paper, we propose a novel spatial-configuration-feature-based network that effectively learns the anatomical correlation between the landmarks. Specifically, we focus on a regularization method and a spatial configuration loss that capture the spatial relationship between the landmarks. Each heatmap, generated using U-Net, is transformed into an embedded spatial feature vector using the soft-argmax method and spatial feature maps, here, Cartesian and Polar coordinates. A correlation map between landmarks based on the spatial feature vector is generated and used to calculate the loss, along with the heatmap output. This approach adopts an end-to-end learning approach, requiring only a single feedforward execution during the test phase to localize all landmarks. The proposed regularization method is computationally efficient, differentiable, and highly parallelizable. The experimental results show that our method can learn global contextual features between landmarks and achieve state-of-the-art performance. Our method is expected to significantly improve localization accuracy when applied to healthcare systems that require accurate medical landmark localization.

Motivation moderates gender differences in navigation performance

Gender differences in navigation performance are a recurrent and controversial topic. Previous research suggests that men outperform women in navigation tasks and that men and women exhibit different navigation strategies. Here, we investigate whether motivation to complete the task moderates the relationship between navigation performance and gender. Participants learned the locations of landmarks in a novel virtual city. During learning, participants could trigger a top-down map that depicted their current position and the locations of the landmarks. During testing, participants were divided into control and treatment groups and were not allowed to consult the map. All participants were given 16 minutes to navigate to the landmarks, but those in the treatment group were monetarily penalized for every second they spent completing the task. Results revealed a negative relationship between physiological arousal and the time required to locate the landmarks. In addition, gender differences in strategy were found during learning, with women spending more time with the map and taking 40% longer than men to locate the landmarks. Interestingly, an interaction between gender and treatment group revealed that women in the control group required more time than men and women in the treatment group to retrieve the landmarks. During testing, women in the control group also took more circuitous routes compared to men in the control group and women in the treatment group. These results suggest that a concurrent and relevant stressor can motivate women to perform similarly to men, helping to diminish pervasive gender differences found in the navigation literature.

Region-Aware Deep Feature-Fused Network for Robust Facial Landmark Localization

In facial landmark localization, facial region initialization usually plays an important role in guiding the model to learn critical face features. Most facial landmark detectors assume a well-cropped face as input and may underperform in real applications if the input is unexpected. To alleviate this problem, we present a region-aware deep feature-fused network (RDFN). The RDFN consists of a region detection subnetwork and a region-wise landmark localization subnetwork to explicitly solve the input initialization problem and derive the landmark score maps, respectively. To exploit the association between tasks, we develop a cross-task feature fusion scheme to extract multi-semantic region features while trading off their importance in different dimensions via global channel attention and global spatial attention. Furthermore, we design a within-task feature fusion scheme to capture the multi-scale context and improve the gradient flow for the landmark localization subnetwork. At the inference stage, a location reweighting strategy is employed to transform the score maps into 2D landmark coordinates. Extensive experimental results demonstrate that our method has competitive performance compared to recent state-of-the-art methods, achieving NMEs of 3.28%, 1.48%, and 3.43% on the 300W, AFLW, and COFW datasets, respectively.

Three-dimensional analysis of mandibular and mental canals corroborating with teeth and mental foramen by cone beam computed tomography.

This study aimed to analyse the 3D patterns of the mandibular and mental canals (MDC and MC) referring to the surrounding prominent surgical landmarks such as teeth and mental foramen by cone beam computed tomography (CBCT). CBCT scans of 354 patients aged 18-67 years with mandibular first premolar to second molar were included and reconstructed 3-dimensionally (3D) by mimics. The parameters of MDC and MC were measured referring to teeth and mental foramen. From the first premolars to the second molars, the mandibular canals showed a trend of gradually closer to the cementoenamel junction (CEJ) of the adjacent teeth and farther away from the buccal cortical plate. The distance of the MDC with the root apexes (RA) was relatively constant from the first premolar to the first molar, but became much closer to the second molar. About 10.8% of the second molars had MDC-RA distances of shorter than 2 mm, and 1.34% even had the MDC superior to the RA. Moreover, the Type III of MC presented in 66.0% of the subjects and had a relatively longer length. Besides, the existence of Type I MC may be related to the MDC featuring with close distances to the RA and CEJ of the adjacent teeth. Dentists and surgeons should know the patterns of mandibular and mental canals. A better understanding of the MDC and MC and their relationship to local anatomical landmarks may facilitate the planning of surgeries and alert potential nerve injuries in the operative procedures.

Angular Photogrammetric Analysis of Facial Soft Tissue by Image Processing Algorithms.

The main aim of this study was to present an automatic method based on image processing algorithms for facial anatomical landmark localization and angular photogrammetric analysis applicable for rhinoplasty surgery. We studied and measured color profile photographs of 100 patients before and after rhinoplasty surgery. In facial anthropometry analysis, anatomical landmarks are often defined by specialists, manually. This process is time-consuming and requires training and skill. The Cascade Regression Method (CRM) was utilized for facial landmark detection to overcome the mentioned problem. In this study, 11 anatomical landmarks were used to measure 9 facial angular metrics. Finally, a t-test (with the significance level set at a p-value of 0.05) was applied to analyze before surgery versus after surgery comparisons. Experimental results dedicated that there is a significance difference (p < 0.001) in nasofrontal, nasolabial, mentolabial, nasomental, facial convexity including nose, facial convexity excluding nose, projection of the upper lip to chin, and H angles before and after surgery. Also, results showed that there is not a significance difference in nose tip angle. We believe that the presented system can aim to reduce the personal errors made by manual measurement and to facilitate facial anthropometry analysis before and after surgery with high accuracy. Also, the normative data for Iranian women can be used as a guide for the diagnosis and planning of oral and maxillofacial, ENT, and plastic surgeries. This journal requires that authors assign a level of evidence to each article. For a full description of these Evidence-Based Medicine ratings, please refer to the Table of Contents or the online Instructions to Authors www.springer.com/00266 .

Orthodontic craniofacial pattern diagnosis: cephalometric geometry and machine learning.

Efficient and reliable diagnosis of craniofacial patterns is critical to orthodontic treatment. Although machine learning (ML) is time-saving and high-precision, prior knowledge should validate its reliability. This study proposed a craniofacial ML diagnostic workflow base on a cephalometric geometric model through clinical verification. A cephalometric geometric model was established to determine the landmark location by analyzing 408 X-ray lateral cephalograms. Through geometric information and feature engineering, nine supervised ML algorithms were conducted for sagittal and vertical skeleton patterns. After dimension reduction, plane decision boundary and landmark contribution contours were depicted to demonstrate the diagnostic consistency and the consistency with clinical norms. As a result, multi-layer perceptron achieved 97.56% accuracy for sagittal, while linear support vector machine reached 90.24% for the vertical. Sagittal diagnoses showed average superiority (91.60 ± 5.43)% over the vertical (82.25 ± 6.37)%, where discriminative algorithms exhibited more steady performance (93.20 ± 3.29)% than the generative (85.98 ± 9.48)%. Further, the Kruskal-Wallis H test was carried out to explore statistical differences in diagnoses. Though sagittal patterns had no statistical difference in diagnostic accuracy, the vertical showed significance. All aspects of the tests indicated that the proposed craniofacial ML workflow was highly consistent with clinical norms and could supplement practical diagnosis.

Ultrasound-guided stenting in the common femoral vein for accurate stent distal landing.

Venous stenting has been reported with excellent clinical results; however, inadequate inflow can increase the risk of stent occlusion. When extending the stent into the common femoral vein (CFV), it is essential to ensure adequate inflow from the femoral vein, deep femoral vein (DFV), and great saphenous vein. Accurate identification of the distal landing zone (DLZ) of the stent is crucial to ensure adequate inflow. The DLZ is usually determined by venography or intravascular ultrasound (IVUS) with reference to bony landmarks. However, the uncertainty can lead to misidentification of the DLZ and inadequate stent placement, resulting in stent occlusion. From December 2016 to December 2022, 42 venous stent placements were performed in 40 patients with post-thrombotic iliofemoral vein obstruction and/or stenosis. Three patients had developed early stent occlusion owing to a misidentified DLZ. To improve accuracy in identifying the DLZ during venous stenting, ultrasound-guided placement was performed in the CFV of five patients (four on the left and one on the right) with post-thrombotic changes in the CFV and occlusion of the common iliac vein and external iliac vein. The distal end of the stent was adjusted just above the saphenofemoral junction in two cases and just proximal to the DFV confluence in three cases. Stent placement was verified using both long-axis ultrasound and fluoroscopy. Ultrasound images of the CFV region provided clear visualization of the stent deployment site and accurate landmark locations, such as the saphenofemoral junction and DFV confluence, allowing for precise adjustments during stent deployment. This technique enabled easier and more definitive identification of other branches of the CFV than previously provided by IVUS and venography. No complications were observed in any of the 42 cases, and long-term patency was achieved at the final follow-up after stenting (average, 10months; range, 3-14months). Ultrasound-guided stenting in the CFV allows for real-time and accurate stent deployment with precise adjustment to the optimal DLZ. Using this technique, combined with venography and IVUS, missed distal lesions and subsequent stent occlusion can be prevented, potentially contributing to better treatment outcomes.

Anatomical landmarks localization for capsule endoscopy studies.

Wireless Capsule Endoscopy is a medical procedure that uses a small, wireless camera to capture images of the inside of the digestive tract. The identification of the entrance and exit of the small bowel and of the large intestine is one of the first tasks that need to be accomplished to read a video. This paper addresses the design of a clinical decision support tool to detect these anatomical landmarks. We have developed a system based on deep learning that combines images, timestamps, and motion data to achieve state-of-the-art results. Our method does not only classify the images as being inside or outside the studied organs, but it is also able to identify the entrance and exit frames. The experiments performed with three different datasets (one public and two private) show that our system is able to approximate the landmarks while achieving high accuracy on the classification problem (inside/outside of the organ). When comparing the entrance and exit of the studied organs, the distance between predicted and real landmarks is reduced from 1.5 to 10 times with respect to previous state-of-the-art methods.

Automatic Localization of the Pons and Vermis on Fetal Brain MR Imaging Using a U-Net Deep Learning Model.

An MRI of the fetus can enhance the identification of perinatal developmental disorders, which improves the accuracy of ultrasound. Manual MRI measurements require training, time, and intra-variability concerns. Pediatric neuroradiologists are also in short supply. Our purpose was developing a deep learning model and pipeline for automatically identifying anatomic landmarks on the pons and vermis in fetal brain MR imaging and suggesting suitable images for measuring the pons and vermis. We retrospectively used 55 pregnant patients who underwent fetal brain MR imaging with a HASTE protocol. Pediatric neuroradiologists selected them for landmark annotation on sagittal single-shot T2-weighted images, and the clinically reliable method was used as the criterion standard for the measurement of the pons and vermis. A U-Net-based deep learning model was developed to automatically identify fetal brain anatomic landmarks, including the 2 anterior-posterior landmarks of the pons and 2 anterior-posterior and 2 superior-inferior landmarks of the vermis. Four-fold cross-validation was performed to test the accuracy of the model using randomly divided and sorted gestational age-divided data sets. A confidence score of model prediction was generated for each testing case. Overall, 85% of the testing results showed a ≥90% confidence, with a mean error of <2.22 mm, providing overall better estimation results with fewer errors and higher confidence scores. The anterior and posterior pons and anterior vermis showed better estimation (which means fewer errors in landmark localization) and accuracy and a higher confidence level than other landmarks. We also developed a graphic user interface for clinical use. This deep learning-facilitated pipeline practically shortens the time spent on selecting good-quality fetal brain images and performing anatomic measurements for radiologists.

Application of Artificial Intelligence (AI) in a Cephalometric Analysis: A Narrative Review.

In recent years, the application of artificial intelligence (AI) has become more and more widespread in medicine and dentistry. It may contribute to improved quality of health care as diagnostic methods are getting more accurate and diagnostic errors are rarer in daily medical practice. The aim of this paper was to present data from the literature on the effectiveness of AI in orthodontic diagnostics based on the analysis of lateral cephalometric radiographs. A review of the literature from 2009 to 2023 has been performed using PubMed, Medline, Scopus and Dentistry & Oral Sciences Source databases. The accuracy of determining cephalometric landmarks using widely available commercial AI-based software and advanced AI algorithms was presented and discussed. Most AI algorithms used for the automated positioning of landmarks on cephalometric radiographs had relatively high accuracy. At the same time, the effectiveness of using AI in cephalometry varies depending on the algorithm or the application type, which has to be accounted for during the interpretation of the results. In conclusion, artificial intelligence is a promising tool that facilitates the identification of cephalometric landmarks in everyday clinical practice, may support orthodontic treatment planning for less experienced clinicians and shorten radiological examination in orthodontics. In the future, AI algorithms used for the automated localisation of cephalometric landmarks may be more accurate than manual analysis.

Mask-FPAN: Semi-supervised face parsing in the wild with de-occlusion and UV GAN

The field of fine-grained semantic segmentation for a person’s face and head, which includes identifying facial parts and head components, has made significant progress in recent years. However, this task remains challenging due to the difficulty of considering ambiguous occlusions and large pose variations. To address these difficulties, we propose a new framework called Mask-FPAN. Our framework includes a de-occlusion module that learns to parse occluded faces in a semi-supervised manner, taking into account face landmark localization, face occlusion estimations, and detected head poses. Additionally, we improve the robustness of 2D face parsing by combining a 3D morphable face model with the UV GAN. We also introduce two new datasets, named FaceOccMask-HQ and CelebAMaskOcc-HQ, to aid in face parsing work. Our proposed Mask-FPAN framework successfully addresses the challenge of face parsing in the wild and achieves significant performance improvements, with a mIoU increase from 0.7353 to 0.9013 compared to the current state-of-the-art on challenging face datasets.

Investigating the different domains of environmental knowledge acquired from virtual navigation and their relationship to cognitive factors and wayfinding inclinations

When learning an environment from virtual navigation people gain knowledge about landmarks, their locations, and the paths that connect them. The present study newly aimed to investigate all these domains of knowledge and how cognitive factors such as visuospatial abilities and wayfinding inclinations might support virtual passive navigation. A total of 270 participants (145 women) were tested online. They: (i) completed visuospatial tasks and answered questionnaires on their wayfinding inclinations; and (ii) learnt a virtual path. The environmental knowledge they gained was assessed on their free recall of landmarks, their egocentric and allocentric pointing accuracy (location knowledge), and their performance in route direction and landmark location tasks (path knowledge). Visuospatial abilities and wayfinding inclinations emerged as two separate factors, and environmental knowledge as a single factor. The SEM model showed that both visuospatial abilities and wayfinding inclinations support the environmental knowledge factor, with similar pattern of relationships in men and women. Overall, factors related to the individual are relevant to the environmental knowledge gained from an online virtual passive navigation.

SDMT: Spatial Dependence Multi-Task Transformer Network for 3D Knee MRI Segmentation and Landmark Localization

Knee segmentation and landmark localization from 3D MRI are two significant tasks for diagnosis and treatment of knee diseases. With the development of deep learning, Convolutional Neural Network (CNN) based methods have become the mainstream. However, the existing CNN methods are mostly single-task methods. Due to the complex structure of bone, cartilage and ligament in the knee, it is challenging to complete the segmentation or landmark localization alone. And establishing independent models for all tasks will bring difficulties for surgeon's clinical using. In this paper, a Spatial Dependence Multi-task Transformer (SDMT) network is proposed for 3D knee MRI segmentation and landmark localization. We use a shared encoder for feature extraction, then SDMT utilizes the spatial dependence of segmentation results and landmark position to mutually promote the two tasks. Specifically, SDMT adds spatial encoding to the features, and a task hybrided multi-head attention mechanism is designed, in which the attention heads are divided into the inter-task attention head and the intra-task attention head. The two attention head deal with the spatial dependence between two tasks and correlation within the single task, respectively. Finally, we design a dynamic weight multi-task loss function to balance the training process of two task. The proposed method is validated on our 3D knee MRI multi-task datasets. Dice can reach 83.91% in the segmentation task, and MRE can reach 2.12 mm in the landmark localization task, it is competitive and superior over other state-of-the-art single-task methods.