Interactive Modules Research Articles

Abstract 4137441: Online Cognitive Behavioral Therapy Targeting Psychological Distress Following Open Heart Surgery: An Uncontrolled Pilot Study

Background: Cardiac surgery is a significant procedure that reduces mortality and alleviates symptom burden in individuals with cardiovascular diseases. However, many patients develop depression and cardiac-related anxiety postoperatively, which negatively affects long-term prognosis and rehabilitation. There is a clinical need to develop scalable psychological treatments to mitigate these negative health effects. Objective: To investigate the feasibility and potential efficacy of a brief internet-delivered cognitive behavioral therapy (internet-CBT) intervention to reduce psychological distress in patients following cardiac surgery. Methods: The pilot study included 32 cardiac surgery patients (i.e., CABG, heart valve repair/replacement, aortic repair, or combined CABG and valve repair/replacement) with endorsed postoperative psychological distress and/or interference with daily life who underwent surgery from eight weeks to nine months prior to enrollment. Exclusion criteria included prior cardiac surgery with a ventricular assist device or heart transplant and severe medical or psychiatric illness. The internet-CBT targeted cardiac anxiety and depressive inactivity, lasted for five weeks, and was guided by clinical psychologists via text-based interactive online treatment modules. It included interoceptive exposure, in-vivo exposure, and behavioral activation. Self-assessments were completed at pre-treatment, post-treatment, and at 6-month follow-up. Results: Preliminary analyses post-treatment showed substantial pre-post improvements across multiple domains. Notable findings included significant reductions in depressive symptoms (Cohen’s d = 0.70; p = 0.04), cardiac anxiety (Cohen’s d = 1.53; p < 0.001), and perceived severity of post-operative symptoms (Cohen’s d = 1.06; p = 0.006). Participants demonstrated high adherence to the treatment, with the large majority of participants completing all five treatment modules (83.3%). Satisfaction with the treatment was also high (25.5 points out of 32), as measured by the Client Satisfaction Questionnaire, and no adverse events were reported. Conclusion: This novel internet-CBT intervention post-cardiac surgery appears to be feasible, acceptable, and clinically promising in reducing postoperative psychological distress. It could be used as a viable adjunct treatment to enhance recovery post-cardiac surgery. These preliminary findings warrant further testing in a randomized controlled trial.

Integration of genomic and transcriptomic layers in RNA‐Seq data leads to protein interaction modules with improved Alzheimer's disease associations

AbstractAlzheimer's disease (AD) is the most common neurodegenerative disease, and it is currently untreatable. RNA sequencing (RNA‐Seq) is commonly used in the literature to identify AD‐associated molecular mechanisms by analysing changes in gene expression. RNA‐Seq data can also be used to detect genomic variants, enabling the identification of the genes with a higher load of deleterious variants in patients compared with controls. Here, we analysed AD RNA‐Seq datasets to obtain differentially expressed genes and genes with a higher load of pathogenic variants in AD, and we combined them in a single list. We mapped these genes on a human protein–protein interaction network to discover subnetworks perturbed by AD. Our results show that utilizing gene pathogenicity information from RNA‐Seq data positively contributes to the disclosure of AD‐related mechanisms. Moreover, dividing the discovered subnetworks into highly connected modules reveals a clearer picture of altered molecular pathways that, otherwise, would not be captured. Repeating the whole pipeline with human metabolic network genes led to results confirming the positive contribution of gene pathogenicity information and enabled a more detailed identification of altered metabolic pathways in AD.

Innovative Digital Learning

This poster highlights the effective use of learning technologies like animation, generative AI tools, and immersive reality to create digital learning and assessment artifacts for university nursing students. The purpose is to deliver a high-quality digital learning experience aligned with the university's pedagogy and education plan. It also showcases accessible, inclusive digital artifacts and early evaluations of their impact on learning. Animation technology is at the forefront of creating engaging educational content. By using animation, complex concepts can be shown in a visually appealing and easily digestible manner, making learning more engaging for students (Liu & Elms, 2019). It also allows for dynamic storytelling and the illustration of abstract ideas, which can be particularly beneficial in concepts that are challenging to visualize (Rasmusson & Bourne, 2017). Generative AI plays a pivotal role in developing scripts for these educational animations. Leveraging AI, we generate content tailored to learners' needs and optimized for clarity and engagement. Though AI-generated, our academic team thoroughly edits and revises the scripts to be clinically representative of the topic covered. This AI-driven script development ensures efficient collaboration with academics, resulting in accurate, relevant, and pedagogically sound material. (ProQuest LLC, 2024). Using cutting-edge animation and generative AI technologies can create impactful nursing education materials that boost student engagement. (White, Huang, & Campanale, 2023) and facilitate learning outcomes (Huang et al., 2020). These activities are anchored in the university’s signature pedagogy—active, applied, and authentic—immersing students in educational content and inspiring them to apply their knowledge and skills in real-world scenarios presented through animated learning activities. In addition, immersive reality technologies, including Virtual Reality (VR) and Augmented Reality (AR), further enhance the educational experience. By incorporating AR and VR as additional resources to complement the animations, we create immersive environments where students can interact with the material in a hands-on manner. This not only can make learning more engaging but can also help students better understand and retain complex concepts through experiential learning (Tene et al., 2024). A digital learning developer collaborated with the academic team to create an augmented reality-based artifact for nursing students. Integrated into Canvas our institution’s learning management system this artifact presented difficult topics in three dimensions with the aim of effectively explaining concepts and enhancing student engagement. While animation and immersive technologies can boost engagement, they can also present inclusivity and accessibility challenges for some students. To facilitateinclusivity and accessibility, we followed digital accessibility guidelines and universal design principles (Ismailov & Chiu, 2022). We evaluated tools like animation, AR/VR, and interactive modules for accessibility, offering alternatives when needed. This approach accommodated diverse learning preferences, making education accessible and inclusive for all students. The environmental impact of adopting these technologies is a significant consideration. Digital tools can reduce the need for physical materials and decrease travel-related emissions, making education more sustainable. The early assessment of the digital learning materials, incorporating animation, AI, and immersive learning technologies, reveals promising results in enhancing student engagement. These materials cater to various learning preferences reflecting insights from the academic team.

RS-DETR: An Improved Remote Sensing Object Detection Model Based on RT-DETR

Object detection is a fundamental task in computer vision. Recently, deep-learning-based object detection has made significant progress. However, due to large variations in target scale, the predominance of small targets, and complex backgrounds in remote sensing imagery, remote sensing object detection still faces challenges, including low detection accuracy, poor real-time performance, high missed detection rates, and high false detection rates in practical applications. To enhance remote sensing target detection performance, this study proposes a new model, the remote sensing detection transformer (RS-DETR). First, we incorporate cascaded group attention (CGA) into the attention-driven feature interaction module. By capturing features at different levels, it enhances the interaction between features through cascading and improves computational efficiency. Additionally, we propose an enhanced bidirectional feature pyramid network (EBiFPN) to facilitate multi-scale feature fusion. By integrating features across multiple scales, it improves object detection accuracy and robustness. Finally, we propose a novel bounding box regression loss function, Focaler-GIoU, which makes the model focus more on difficult samples, improving detection performance for small and overlapping targets. Experimental results on the satellite imagery multi-vehicles dataset (SIMD) and the high-resolution remote sensing object detection (TGRS-HRRSD) dataset show that the improved algorithm achieved mean average precision (mAP) of 78.2% and 91.6% at an intersection over union threshold of 0.5, respectively, which is an improvement of 2.0% and 1.5% over the baseline model. This result demonstrates the effectiveness and robustness of our proposed method for remote sensing image object detection.

Simulation-Based Training in Clinical Event Debriefing Improves Leadership Performance.

Clinical event debriefing (CED) improves healthcare team performance and patient outcomes. Most pediatric emergency medicine (PEM) physicians do not receive formal training in leading CED. Our objectives were to develop a CED curriculum and evaluate its effect on performance, knowledge, comfort, and clinical practice. This was a single group pre-post-retention study. We developed a hybrid curriculum with simulation, an interactive module, and individual feedback. We invited faculty and fellows from the PEM division of our hospital to participate. During an in-person training day, participants led standardized clinical simulation scenarios followed by simulated CED with immediate feedback on their leadership performance. They watched an interactive module between scenarios. Participants returned for a retention assessment 2-6 months later with a third simulation and debrief. Participants completed surveys measuring attitudes, experiences, and knowledge. Participants also evaluated the curriculum.The primary outcome was CED leadership performance using a novel 21-item tool that we developed, the Debrief Leadership Tool for Assessment (DELTA). A blinded, trained rater measured performance with DELTA. Secondary outcomes included changes in knowledge and comfort and changes in clinical practice. Twenty-seven participants enrolled and completed all parts of the curriculum and assessments. Debrief leadership performance improved by a mean of 3.7 points on DELTA pre-training to post-training (95% confidence interval = 2.7, 4.6, P < 0.01) and by 1.4 points from pre-training to retention (95% confidence interval = 0.1, 2.8, P = 0.03). Knowledge and comfort also significantly improved from pre-training to post-training and were sustained at retention. Most (67%) participants changed their clinical practice of CED after completing the curriculum. All participants would recommend the training to other PEM physicians. A hybrid simulation-based curriculum in leading CED for PEM physicians was associated with improvement in CED leadership performance, knowledge, and comfort. PEM physicians incorporated training into their clinical practice.

SAMCT: Segment Any CT Allowing Labor-Free Task-Indicator Prompts.

Segment anything model (SAM), a foundation model with superior versatility and generalization across diverse segmentation tasks, has attracted widespread attention in medical imaging. However, it has been proved that SAM would encounter severe performance degradation due to the lack of medical knowledge in training and local feature encoding. Though several SAM-based models have been proposed for tuning SAM in medical imaging, they still suffer from insufficient feature extraction and highly rely on high-quality prompts. In this paper, we propose a powerful foundation model SAMCT allowing labor-free prompts and train it on a collected large CT dataset consisting of 1.1M CT images and 5M masks from public datasets. Specifically, based on SAM, SAMCT is further equipped with a U-shaped CNN image encoder, a cross-branch interaction module, and a task-indicator prompt encoder. The U-shaped CNN image encoder works in parallel with the ViT image encoder in SAM to supplement local features. Cross-branch interaction enhances the feature expression capability of the CNN image encoder and the ViT image encoder by exchanging global perception and local features from one to the other. The task-indicator prompt encoder is a plug-and-play component to effortlessly encode task-related indicators into prompt embeddings. In this way, SAMCT can work in an automatic manner in addition to the semi-automatic interactive strategy in SAM. Extensive experiments demonstrate the superiority of SAMCT against the state-of-the-art task-specific and SAM-based medical foundation models on various tasks. The code, data, and model checkpoints are available at https://github.com/xianlin7/SAMCT.

Evaluating a novel clinical reasoning tool for physical therapy students: A pilot study.

This paper explores the use of the Anticipate-Plan-Pause Clinical Reasoning (APP CR) Tool to enhance clinical reasoning (CR) in Doctor of Physical Therapy (DPT) students. Current pedagogical methods often inadequately integrate CR abilities in a manner that is both engaging and effective. Drawing on several theoretical frameworks, the APP CR Tool was integrated into an interactive module for DPT students, aiming to improve CR abilities while reducing cognitive load and stress during high-stakes simulations. A randomised controlled trial was conducted with nine DPT students, divided into a control group (n = 5) and an intervention group (n = 4). Both groups used an interactive module with case-based scenarios, but only the intervention group used the APP CR Tool. Improvements in CR abilities and reduced stress and cognitive load were observed in both groups. However, statistical analysis revealed no significant differences between the groups. Despite this, a trend toward greater improvements in CR abilities was noted in the intervention group (p = 0.084). Further research is needed to validate these findings and refine the tool. The APP CR Tool shows potential for enhancing CR in DPT students, although its current application may require refinement to achieve statistically significant results. Educators should consider iterative improvements to the tool and its curriculum integration to better support CR development.

Smartphone-Based Cognitive Behavioral Therapy and Customized Sound Therapy for Tinnitus: A Randomized Controlled Trial.

To evaluate the efficacy of a smartphone-based application providing tinnitus-specific cognitive behavioral therapy and customized sound therapy for tinnitus. In a prospective randomized controlled trial, the treatment group participated in daily sound therapy and weekly interactive CBT modules, and the control group did not receive the program (waitlisted). Outcome measures after 8 weeks included the Tinnitus Functional Index (TFI), Generalized Anxiety Disorder (GAD-7), Patient Health Questionnaire (PHQ-9), Perceived Stress Scale (PSS), and Pittsburgh Sleep Quality Index (PSQI). Ninety-two patients (mean age = 57.2 ± 11.9 years) were included. The treatment (n = 47) and control (n = 45) cohorts had similar TFI, PHQ-9, GAD-7, PSS, and PSQI scores at presentation (all P > .05). Treatment-group patients had significantly higher improvements than controlled patients in their TFI (16.7 ± 14.9 vs 1.9 ± 10.8, P < .001), PHQ-9 (1.9 ± 4.2 vs -0.7 ± 3.4, P = .002), GAD-7 (1.1 ± 3.6 vs -0.9 ± 3.3, P = .009), and PSQI scores (2.5 ± 3.2 vs -1.1 ± 2.1, P < .001). Treatment-group subjects also significantly improved in 7 of the 8 TFI domains (all P < .05). TFI improvements of ≥20 occurred in 18 (38.3%) treatment subjects. Of the 9 treatment-group patients presenting with moderate-severe depression, 4 (44.4%) improved to minimal depression. Of the 4 treatment-group patients with moderate-severe anxiety, 2 (50.0%) improved to minimal anxiety. Of the 17 treatment-group patients with moderate-severe sleep difficulties, 10 (55.6%) had considerable sleep improvement. This novel smartphone application providing cognitive behavioral therapy and customized sound therapy was effective in reducing symptom severity and improving anxiety, sleep, and mood for tinnitus patients.

IFGAN: Pre- to Post-Contrast Medical Image Synthesis Based on Interactive Frequency GAN

Medical images provide a visual representation of the internal structure of the human body. Injecting a contrast agent can increase the contrast of diseased tissues and assist in the accurate identification and assessment of conditions. Considering the adverse reactions and side effects caused by contrast agents, previous methods synthesized post-contrast images with pre-contrast images to bypass the administration process. However, existing methods pay inadequate attention to reasonable mapping of the lesion area and ignore gaps between post-contrast and real images in the frequency domain. Thus, in this paper, we propose an interactive frequency generative adversarial network (IFGAN) to solve the above problems and synthesize post-contrast images from pre-contrast images. We first designed an enhanced interaction module that is embedded in the generator to focus on the contrast enhancement region. Within it, target and reconstruction branch features interact to control the local contrast enhancement region feature and maintain the anatomical structure. We propose focal frequency loss to ensure the consistency of post-contrast and real images in the frequency domain. The experimental results demonstrated that IFGAN outperforms other sophisticated approaches in terms of preserving the accurate contrast enhancement of lesion regions and anatomical structures. Specifically, our method produces substantial improvements of 7.9% in structural similarity (SSIM), 36.3% in the peak signal-to-noise ratio (PSNR), and 8.5% in multiscale structural similarity (MSIM) compared with recent state-of-the-art methods.

Diagnosis of Custard Apple Disease Based on Adaptive Information Entropy Data Augmentation and Multiscale Region Aggregation Interactive Visual Transformers

Accurate diagnosis of plant diseases is crucial for crop health. This study introduces the EDA–ViT model, a Vision Transformer (ViT)-based approach that integrates adaptive entropy-based data augmentation for diagnosing custard apple (Annona squamosa) diseases. Traditional models like convolutional neural network and ViT face challenges with local feature extraction and large dataset requirements. EDA–ViT overcomes these by using a multi-scale weighted feature aggregation and a feature interaction module, enhancing both local and global feature extraction. The adaptive data augmentation method refines the training process, boosting accuracy and robustness. With a dataset of 8226 images, EDA–ViT achieved a classification accuracy of 96.58%, an F1 score of 96.10%, and a Matthews Correlation Coefficient (MCC) of 92.24%, outperforming other models. The inclusion of the Deformable Multi-head Self-Attention (DMSA) mechanism further enhanced feature capture. Ablation studies revealed that the adaptive augmentation contributed to a 0.56% accuracy improvement and a 0.34% increase in MCC. In summary, EDA–ViT presents an innovative solution for custard apple disease diagnosis, with potential applications in broader agricultural disease detection, ultimately aiding precision agriculture and crop health management.

AdaGIN: Adaptive Graph Interaction Network for Click-Through Rate Prediction

The goal of click-through rate (CTR) prediction in recommender systems is to effectively work with input features. However, existing CTR prediction models face three main issues. First, many models use a basic approach for feature combinations, leading to noise and reduced accuracy. Second, there is no consideration for the varying importance of features in different interaction orders, affecting model performance. Third, current model architectures struggle to capture different interaction signals from various semantic spaces, leading to sub-optimal performance. To address these issues, we propose the Adaptive Graph Interaction Network (AdaGIN) with the Graph Neural Networks-based Feature Interaction Module (GFIM), the Multi-semantic Feature Interaction Module (MFIM), and the Negative Feedback-based Search (NFS) algorithm. GFIM explicitly aggregates information between features and assesses their importance, while MFIM captures information from different semantic spaces. NFS uses negative feedback to optimize model complexity. Experimental results show AdaGIN outperforms existing models on large-scale public benchmark datasets.

Pengembangan Modul Interaktif Mahasiswa PPG PAUD Terintegrasi Pembelajaran Berbasis Proyek sebagai Optimasi Pembelajaran Mahasiswa PPG di Sekolah Mitra

This article discusses the development of an interactive module for early childhood education (PPG PAUD) students integrated with project-based learning as an effort to optimize learning in partner schools. The program aims to enhance the skills and understanding of PPG PAUD students through the use of interactive modules designed to support project-based learning. The methodology includes needs analysis, module design, implementation, and effectiveness evaluation. The results indicate that the interactive module significantly increases students' learning motivation, engagement, and material comprehension. Additionally, the integration of project-based learning allows students to develop practical skills relevant to the teaching context in partner schools. In conclusion, the development of interactive modules integrated with project-based learning provides a positive contribution to optimizing learning for PPG PAUD students.

AIFormer: Adaptive Interaction Transformer for 3D Point Cloud Understanding

Recently, significant advancements have been made in 3D point cloud analysis by leveraging transformer architecture in 3D space. However, it remains challenging to effectively implement local and global learning within irregular and sparse structures of 3D point clouds. This paper presents the Adaptive Interaction Transformer (AIFormer), a novel hierarchical transformer architecture designed to enhance 3D point cloud analysis by fusing local and global features through the adaptive interaction of features. Specifically, AIFormer mainly consists of several stacked AIFormer Blocks. Each AIFormer module employs the Local Relation Aggregation Module and the Global Context Aggregation Module, respectively, to extract local details of relationships within the reference point and long-range dependencies between reference points. Then, the local and global features are fused using the Adaptive Interaction Module for adaptive interaction to optimize the point representation. Additionally, the AIFormer Block further designs geometric relation functions and contextual relative semantic encoding to enhance local and global feature extraction capabilities, respectively. Extensive experiments on three popular 3D point cloud datasets verify that AIFormer achieves state-of-the-art or comparable performances. Our comprehensive ablation study further validates the effectiveness and soundness of the AIFormer design.

Retrieve water quality parameters of urban rivers from hyperspectral images through feature interaction based two-stage method fused with spectral unmixing and spatial relatedness

Monitoring water quality through efficient quantification of water quality parameters (WQPs) is of paramount importance to environmental management of urban rivers. Unmanned aerial vehicle (UAV) remote sensing techniques posed a great opportunity for visualizing spatial distributions of WQPs concentrations with higher flexibility and monitoring frequency compared to satellite remote sensing techniques, assisting to trace potential contamination sources and prevent water quality from degradation. However, current methods of water quality monitoring usually involved large masses of water samples as training data to keep calculation accuracy every time their study area changed, increasing financial cost and incurring time delay in monitoring and evaluating water quality. This study proposed a UAV-based two-stage multidirectional fusion with probabilistic matrix factorization (TSMF-PMF) method in unified framework, to effectively quantify WQPs including chemical oxygen demand (COD), biochemical oxygen demand (BOD), and turbidity from UAV hyperspectral images. TSMF-PMF established feature interaction and outlier feedback module, ensuring relatively high calculation stability with less training samples. The experimental results showed that TSMF-PMF achieved good performance on predicting concentrations of WQPs with coefficient of determination (R2) and mean absolute percentage error (MAPE) ranging from 0.82 to 0.91 and from 5.35 % to 10.23 %. This study established theoretical and technical foundation to optimize environmental management scheme of urban rivers and provided an application demonstration.

ChatCAD+: Toward a Universal and Reliable Interactive CAD Using LLMs.

The integration of Computer-Aided Diagnosis (CAD) with Large Language Models (LLMs) presents a promising frontier in clinical applications, notably in automating diagnostic processes akin to those performed by radiologists and providing consultations similar to a virtual family doctor. Despite the promising potential of this integration, current works face at least two limitations: (1) From the perspective of a radiologist, existing studies typically have a restricted scope of applicable imaging domains, failing to meet the diagnostic needs of different patients. Also, the insufficient diagnostic capability of LLMs further undermine the quality and reliability of the generated medical reports. (2) Current LLMs lack the requisite depth in medical expertise, rendering them less effective as virtual family doctors due to the potential unreliability of the advice provided during patient consultations. To address these limitations, we introduce ChatCAD+, to be universal and reliable. Specifically, it is featured by two main modules: (1) Reliable Report Generation and (2) Reliable Interaction. The Reliable Report Generation module is capable of interpreting medical images from diverse domains and generate high-quality medical reports via our proposed hierarchical in-context learning. Concurrently, the interaction module leverages up-to-date information from reputable medical websites to provide reliable medical advice. Together, these designed modules synergize to closely align with the expertise of human medical professionals, offering enhanced consistency and reliability for interpretation and advice. The source code is available at GitHub.

A temporal–spectral fusion transformer with subject-specific adapter for enhancing RSVP-BCI decoding

The Rapid Serial Visual Presentation (RSVP)-based Brain–Computer Interface (BCI) is an efficient technology for target retrieval using electroencephalography (EEG) signals. The performance improvement of traditional decoding methods relies on a substantial amount of training data from new test subjects, which increases preparation time for BCI systems. Several studies introduce data from existing subjects to reduce the dependence of performance improvement on data from new subjects, but their optimization strategy based on adversarial learning with extensive data increases training time during the preparation procedure. Moreover, most previous methods only focus on the single-view information of EEG signals, but ignore the information from other views which may further improve performance. To enhance decoding performance while reducing preparation time, we propose a Temporal-Spectral fusion transformer with Subject-specific Adapter (TSformer-SA). Specifically, a cross-view interaction module is proposed to facilitate information transfer and extract common representations across two-view features extracted from EEG temporal signals and spectrogram images. Then, an attention-based fusion module fuses the features of two views to obtain comprehensive discriminative features for classification. Furthermore, a multi-view consistency loss is proposed to maximize the feature similarity between two views of the same EEG signal. Finally, we propose a subject-specific adapter to rapidly transfer the knowledge of the model trained on data from existing subjects to decode data from new subjects. Experimental results show that TSformer-SA significantly outperforms comparison methods and achieves outstanding performance with limited training data from new subjects. This facilitates efficient decoding and rapid deployment of BCI systems in practical use.

IGGCN: Individual-Guided Graph Convolution Network for Pedestrian Trajectory Prediction

Accurately predicting the future trajectory of pedestrians is crucial for applications such as autonomous driving and robot navigation. Graph convolution is widely used in trajectory prediction tasks due to its scalability and adaptive feature-learning capabilities. However, there are two problems with pedestrian trajectory prediction methods based on graph convolution: 1. Previous methods struggled to adjust social interactions according to the attributes of different pedestrians, making it difficult to accurately model the relative importance between different pedestrians and others; 2. Previous methods lacked dynamic processing of pedestrian spatial-temporal interaction features to capture high-level spatial-temporal interaction features effectively. Therefore, we propose an Individual-Guided Graph Convolution Network (IGGCN) for pedestrian trajectory prediction. To tackle problem 1, we design an individual-guided interaction module that can adjust pedestrian social interaction modeling according to the pedestrian's attributes, thereby achieving an accurate description of the relative importance of pedestrians. We extend the module to temporal interaction modeling to further achieve an accurate description of the relative importance of time frames. To address problem 2, we design a deformable convolution module to dynamically process spatial-temporal interaction features through deformable convolution kernels, facilitating the capture of high-level spatial-temporal interaction features. We evaluate our method on the ETH, UCY, and SDD datasets. Quantitative analysis shows that our method has lower prediction errors than the current state-of-the-art methods. Qualitative analysis further reveals that our method effectively eliminates the influence of irrelevant pedestrians and accurately models the spatial-temporal interaction relationship of pedestrians.

Bi-directional information interaction for multi-modal 3D object detection in real-world traffic scenes

Multimodal 3D object detection methods are poorly adapted to real-world traffic scenes due to sparse distribution of point clouds and misalignment multimodal data during actual collection. Among the existing methods, they focus on high-quality open-source datasets, with performance relying on the accurate structural representation of point clouds and the precise mapping relationship between point clouds and images. To solve the above challenges, this paper proposes a multimodal feature-level fusion method based on the bi-directional interaction between image and point cloud. To overcome the sparsity issue in asynchronous multi-modal data, a point cloud densification scheme based on visual guidance and point cloud density guidance is proposed. This scheme can generate object-level virtual point clouds even when the point cloud and image are misaligned. To deal with the unalignment issue between point cloud and image, a bi-directional interaction module based on image-guided interaction with key points of point clouds and point cloud-guided interaction with image context information is proposed. It achieves effective feature fusion even when the point cloud and image are misaligned. The experiments on the VANJEE and KITTI datasets demonstrated the effectiveness of the proposed method, with average precision improvements of 6.20% and 1.54% compared to the baseline.

Choreographing multi-degree of freedom behaviors in large-scale crowd simulations

This study introduces a novel framework for choreographing multi-degree of freedom (MDoF) behaviors in large-scale crowd simulations. The framework integrates multi-objective optimization with spatio-temporal ordering to effectively generate and control diverse MDoF crowd behavior states. We propose a set of evaluation criteria for assessing the aesthetic quality of crowd states and employ multi-objective optimization to produce crowd states that meet these criteria. Additionally, we introduce time offset functions and interpolation progress functions to perform complex and diversified behavior state interpolations. Furthermore, we designed a user-centric interaction module that allows for intuitive and flexible adjustments of crowd behavior states through sketching, spline curves, and other interactive means. Qualitative tests and quantitative experiments on the evaluation criteria demonstrate the effectiveness of this method in generating and controlling MDoF behaviors in crowds. Finally, case studies, including real-world applications in the Opening Ceremony of the 2022 Beijing Winter Olympics, validate the practicality and adaptability of this approach.

GREMI: An Explainable Multi-Omics Integration Framework for Enhanced Disease Prediction and Module Identification.

Multi-omics integration has demonstrated promising performance in complex disease prediction. However, existing research typically focuses on maximizing prediction accuracy, while often neglecting the essential task of discovering meaningful biomarkers. This issue is particularly important in biomedicine, as molecules often interact rather than function individually to influence disease outcomes. To this end, we propose a two-phase framework named GREMI to assist multi-omics classification and explanation. In the prediction phase, we propose to improve prediction performance by employing a graph attention architecture on sample-wise co-functional networks to incorporate biomolecular interaction information for enhanced feature representation, followed by the integration of a joint-late mixed strategy and the true-class-probability block to adaptively evaluate classification confidence at both feature and omics levels. In the interpretation phase, we propose a multi-view approach to explain disease outcomes from the interaction module perspective, providing a more intuitive understanding and biomedical rationale. We incorporate Monte Carlo tree search (MCTS) to explore local-view subgraphs and pinpoint modules that highly contribute to disease characterization from the global-view. Extensive experiments demonstrate that the proposed framework outperforms state-of-the-art methods in seven different classification tasks, and our model effectively addresses data mutual interference when the number of omics types increases. We further illustrate the functional- and disease-relevance of the identified modules, as well as validate the classification performance of discovered modules using an independent cohort.