Musculoskeletal Symptoms(MSS)often arise from prolonged maintenance of bent postures in the neck and trunk during surgical procedures.To prevent MSS,a passive exoskeleton utilizing carbon fiber beams to offer support ...Musculoskeletal Symptoms(MSS)often arise from prolonged maintenance of bent postures in the neck and trunk during surgical procedures.To prevent MSS,a passive exoskeleton utilizing carbon fiber beams to offer support to the neck and trunk was proposed.The application of support force is intended to reduce muscle forces and joint compression forces.A nonlinear mathematical model for the neck and trunk support beam is presented to estimate the support force.A validation test is subsequently conducted to assess the accuracy of the mathematical model.Finally,a preliminary functional evaluation test is performed to evaluate movement capabilities and support provided by the exoskeleton.The mathematical model demonstrates an accuracy for beam support force within a range of 0.8–1.2 N Root Mean Square Error(RMSE).The exoskeleton was shown to allow sufficient Range of Motion(ROM)for neck and trunk during open surgery training.While the exoskeleton showed potential in reducing musculoskeletal load and task difficulty during simulated surgery tasks,the observed reduction in perceived task difficulty was deemed non-significant.This prompts the recommendation for further optimization in personalized adjustments of beams to facilitate improvements in task difficulty and enhance comfort.展开更多
Background:Trunk lean angle is an underrepre sented biomechanical variable for modulating and redistributing lower extremity joint loading and potentially reducing the risk of running-related overuse injuries.The purp...Background:Trunk lean angle is an underrepre sented biomechanical variable for modulating and redistributing lower extremity joint loading and potentially reducing the risk of running-related overuse injuries.The purpose of this study was to systematically alter the trunk lean angle in distance running using an auditory real-time feedback approach and to derive dose-response relationships between sagittal plane trunk lean angle and lower extremity(cumulative)joint loading to guide overuse load management in clinical practice.Methods:Thirty recreational runners(15 males and 15 females)ran at a constant speed of 2.5 m/s at 5 systematically varied trunk lean conditions on a force-instrumented treadmill while kinematic and kinetic data were captured.Results:A change in trunk lean angle from-2°(extension)to 28°(flexion)resulted in a systematic increase in stance phase angular impulse,cumulative impulse,and peak moment at the hip joint in the sagittal and transversal plane.In contrast,a systematic decrease in these parameters at the knee j oint in the sagittal plane and the hip joint in the frontal plane was found(p<0.001).Linear fitting revealed that with every degree of anterior trunk leaning,the cumulative hip joint extension loading increases by 3.26 Nm·s/kg/1000 m,while simultaneously decreasing knee joint extension loading by 1.08 Nm·s/kg/1000 m.Conclusion:Trunk leaning can reduce knee joint loading and hip joint abduction loading,at the cost of hip joint loading in the sagittal and transversal planes during distance running.Modulating lower extremity joint loading by altering trunk lean angle is an effective strategy to redistribute joint load between/within the knee and hip joints.When implementing anterior trunk leaning in clinical practice,the increased demands on the hip musculature,dynamic stability,and the potential trade-off with running economy should be considered.展开更多
Tree trunk instance segmentation is crucial for under-canopy unmanned aerial vehicles(UAVs)to autonomously extract standing tree stem attributes.Using cameras as sensors makes these UAVs compact and lightweight,facili...Tree trunk instance segmentation is crucial for under-canopy unmanned aerial vehicles(UAVs)to autonomously extract standing tree stem attributes.Using cameras as sensors makes these UAVs compact and lightweight,facilitating safe and flexible navigation in dense forests.However,their limited onboard computational power makes real-time,image-based tree trunk segmentation challenging,emphasizing the urgent need for lightweight and efficient segmentation models.In this study,we present RT-Trunk,a model specifically designed for real-time tree trunk instance segmentation in complex forest environments.To ensure real-time performance,we selected SparseInst as the base framework.We incorporated ConvNeXt-T as the backbone to enhance feature extraction for tree trunks,thereby improving segmentation accuracy.We further integrate the lightweight convolutional block attention module(CBAM),enabling the model to focus on tree trunk features while suppressing irrelevant information,which leads to additional gains in segmentation accuracy.To enable RT-Trunk to operate effectively under diverse complex forest environments,we constructed a comprehensive dataset for training and testing by combining self-collected data with multiple public datasets covering different locations,seasons,weather conditions,tree species,and levels of forest clutter.Com-pared with the other tree trunk segmentation methods,the RT-Trunk method achieved an average precision of 91.4%and the fastest inference speed of 32.9 frames per second.Overall,the proposed RT-Trunk provides superior trunk segmentation performance that balances speed and accu-racy,making it a promising solution for supporting under-canopy UAVs in the autonomous extraction of standing tree stem attributes.The code for this work is available at https://github.com/NEFU CVRG/RT Trunk.展开更多
针对复杂果园环境行间导航树干检测问题,提出一种基于多线激光雷达(Light detection and ranging,Li DAR)的主干形果树树干层级检测方法,使用16线VLP-16型LiDAR采集车辆周围的果园点云数据,通过目标分割和树干检测2个步骤层次化检测树干...针对复杂果园环境行间导航树干检测问题,提出一种基于多线激光雷达(Light detection and ranging,Li DAR)的主干形果树树干层级检测方法,使用16线VLP-16型LiDAR采集车辆周围的果园点云数据,通过目标分割和树干检测2个步骤层次化检测树干,去除非树干目标,提高树干检测精度。首先,设置环形感兴趣区域(Region of interest,ROI),采用地面拟合算法移除地面点云,消除果园目标点云之间的连通性;其次,设置矩形ROI,采用基于密度的带噪声空间聚类(Density-based spatial clustering of applications with noise,DBSCAN)算法对非地面点云进行x Oy平面聚类,根据Li DAR测量分辨率和果园目标参数设置DBSCAN算法超参数,将非地面点云分割为若干目标簇;然后,从全局和局部2个尺度提取目标簇的几何和强度特征,用这些特征描述树干与其他果园目标间的差异;最后,采用训练好的树干检测器融合特征,将目标簇划分为树干与非树干2个类别,输出树干簇。树干检测步骤采用随机森林(Random forest,RF)算法进行离线特征选择与融合,使用树干和非树干训练样本,基于基尼指数(Gini index,GI)改变量评价特征重要性,从初始特征中选择22个鉴别力较强的特征,再融合这些特征生成树干检测器。实验场景为标准化种植核桃园,共采集1317帧点云数据,从中分割12213个目标簇,其中,树冠、杂草、支撑杆、围栏、土坡、农具、行人等非树干目标占比58.04%。按照帧比例1∶4将目标簇随机划分为训练集和测试集,测试集树干检测精确率为99.16%、召回率为99.21%、F1分数为99.19%,树干层级检测平均帧耗时85.25 ms。本文方法能对复杂果园场景快速、精准地检测出树干,满足果园行间导航对树干检测的准确性和实时性要求。展开更多
基金funded by China Scholarship Council,Grant Number 201906840121department of rehabilitation medicine,University Medical Center Groningen,University of Groningen,grant number:O/085350.
文摘Musculoskeletal Symptoms(MSS)often arise from prolonged maintenance of bent postures in the neck and trunk during surgical procedures.To prevent MSS,a passive exoskeleton utilizing carbon fiber beams to offer support to the neck and trunk was proposed.The application of support force is intended to reduce muscle forces and joint compression forces.A nonlinear mathematical model for the neck and trunk support beam is presented to estimate the support force.A validation test is subsequently conducted to assess the accuracy of the mathematical model.Finally,a preliminary functional evaluation test is performed to evaluate movement capabilities and support provided by the exoskeleton.The mathematical model demonstrates an accuracy for beam support force within a range of 0.8–1.2 N Root Mean Square Error(RMSE).The exoskeleton was shown to allow sufficient Range of Motion(ROM)for neck and trunk during open surgery training.While the exoskeleton showed potential in reducing musculoskeletal load and task difficulty during simulated surgery tasks,the observed reduction in perceived task difficulty was deemed non-significant.This prompts the recommendation for further optimization in personalized adjustments of beams to facilitate improvements in task difficulty and enhance comfort.
文摘Background:Trunk lean angle is an underrepre sented biomechanical variable for modulating and redistributing lower extremity joint loading and potentially reducing the risk of running-related overuse injuries.The purpose of this study was to systematically alter the trunk lean angle in distance running using an auditory real-time feedback approach and to derive dose-response relationships between sagittal plane trunk lean angle and lower extremity(cumulative)joint loading to guide overuse load management in clinical practice.Methods:Thirty recreational runners(15 males and 15 females)ran at a constant speed of 2.5 m/s at 5 systematically varied trunk lean conditions on a force-instrumented treadmill while kinematic and kinetic data were captured.Results:A change in trunk lean angle from-2°(extension)to 28°(flexion)resulted in a systematic increase in stance phase angular impulse,cumulative impulse,and peak moment at the hip joint in the sagittal and transversal plane.In contrast,a systematic decrease in these parameters at the knee j oint in the sagittal plane and the hip joint in the frontal plane was found(p<0.001).Linear fitting revealed that with every degree of anterior trunk leaning,the cumulative hip joint extension loading increases by 3.26 Nm·s/kg/1000 m,while simultaneously decreasing knee joint extension loading by 1.08 Nm·s/kg/1000 m.Conclusion:Trunk leaning can reduce knee joint loading and hip joint abduction loading,at the cost of hip joint loading in the sagittal and transversal planes during distance running.Modulating lower extremity joint loading by altering trunk lean angle is an effective strategy to redistribute joint load between/within the knee and hip joints.When implementing anterior trunk leaning in clinical practice,the increased demands on the hip musculature,dynamic stability,and the potential trade-off with running economy should be considered.
基金supported in part by the National Natural Science Foundation of China(No.31470714 and 61701105).
文摘Tree trunk instance segmentation is crucial for under-canopy unmanned aerial vehicles(UAVs)to autonomously extract standing tree stem attributes.Using cameras as sensors makes these UAVs compact and lightweight,facilitating safe and flexible navigation in dense forests.However,their limited onboard computational power makes real-time,image-based tree trunk segmentation challenging,emphasizing the urgent need for lightweight and efficient segmentation models.In this study,we present RT-Trunk,a model specifically designed for real-time tree trunk instance segmentation in complex forest environments.To ensure real-time performance,we selected SparseInst as the base framework.We incorporated ConvNeXt-T as the backbone to enhance feature extraction for tree trunks,thereby improving segmentation accuracy.We further integrate the lightweight convolutional block attention module(CBAM),enabling the model to focus on tree trunk features while suppressing irrelevant information,which leads to additional gains in segmentation accuracy.To enable RT-Trunk to operate effectively under diverse complex forest environments,we constructed a comprehensive dataset for training and testing by combining self-collected data with multiple public datasets covering different locations,seasons,weather conditions,tree species,and levels of forest clutter.Com-pared with the other tree trunk segmentation methods,the RT-Trunk method achieved an average precision of 91.4%and the fastest inference speed of 32.9 frames per second.Overall,the proposed RT-Trunk provides superior trunk segmentation performance that balances speed and accu-racy,making it a promising solution for supporting under-canopy UAVs in the autonomous extraction of standing tree stem attributes.The code for this work is available at https://github.com/NEFU CVRG/RT Trunk.
文摘针对复杂果园环境行间导航树干检测问题,提出一种基于多线激光雷达(Light detection and ranging,Li DAR)的主干形果树树干层级检测方法,使用16线VLP-16型LiDAR采集车辆周围的果园点云数据,通过目标分割和树干检测2个步骤层次化检测树干,去除非树干目标,提高树干检测精度。首先,设置环形感兴趣区域(Region of interest,ROI),采用地面拟合算法移除地面点云,消除果园目标点云之间的连通性;其次,设置矩形ROI,采用基于密度的带噪声空间聚类(Density-based spatial clustering of applications with noise,DBSCAN)算法对非地面点云进行x Oy平面聚类,根据Li DAR测量分辨率和果园目标参数设置DBSCAN算法超参数,将非地面点云分割为若干目标簇;然后,从全局和局部2个尺度提取目标簇的几何和强度特征,用这些特征描述树干与其他果园目标间的差异;最后,采用训练好的树干检测器融合特征,将目标簇划分为树干与非树干2个类别,输出树干簇。树干检测步骤采用随机森林(Random forest,RF)算法进行离线特征选择与融合,使用树干和非树干训练样本,基于基尼指数(Gini index,GI)改变量评价特征重要性,从初始特征中选择22个鉴别力较强的特征,再融合这些特征生成树干检测器。实验场景为标准化种植核桃园,共采集1317帧点云数据,从中分割12213个目标簇,其中,树冠、杂草、支撑杆、围栏、土坡、农具、行人等非树干目标占比58.04%。按照帧比例1∶4将目标簇随机划分为训练集和测试集,测试集树干检测精确率为99.16%、召回率为99.21%、F1分数为99.19%,树干层级检测平均帧耗时85.25 ms。本文方法能对复杂果园场景快速、精准地检测出树干,满足果园行间导航对树干检测的准确性和实时性要求。
