scholarly journals Towards Automated Three-Dimensional Tracking of Nephrons through Stacked Histological Image Sets

2015 ◽  
Vol 2015 ◽  
pp. 1-13 ◽  
Author(s):  
Charita Bhikha ◽  
Arne Andreasen ◽  
Erik I. Christensen ◽  
Robyn F. R. Letts ◽  
Adam Pantanowitz ◽  
...  
Keyword(s):  
Intelligent System ◽  
Learning Algorithm ◽  
Three Dimensional ◽  
Rule Base ◽  
Interstitial Tissue ◽  
Tracking Accuracy ◽  
Histological Image ◽  
Manual Intervention ◽  
Lower Accuracy ◽  
Standard Tool

An automated approach for tracking individual nephrons through three-dimensional histological image sets of mouse and rat kidneys is presented. In a previous study, the available images were tracked manually through the image sets in order to explore renal microarchitecture. The purpose of the current research is to reduce the time and effort required to manually trace nephrons by creating an automated, intelligent system as a standard tool for such datasets. The algorithm is robust enough to isolate closely packed nephrons and track their convoluted paths despite a number of nonideal, interfering conditions such as local image distortions, artefacts, and interstitial tissue interference. The system comprises image preprocessing, feature extraction, and a custom graph-based tracking algorithm, which is validated by a rule base and a machine learning algorithm. A study of a selection of automatically tracked nephrons, when compared with manual tracking, yields a 95% tracking accuracy for structures in the cortex, while those in the medulla have lower accuracy due to narrower diameter and higher density. Limited manual intervention is introduced to improve tracking, enabling full nephron paths to be obtained with an average of 17 manual corrections per mouse nephron and 58 manual corrections per rat nephron.

scholarly journals Penerapan Estimasi Posisi dan Tracking Wajah Pada Sistem Presensi Mahasiswa

2020 ◽  
Vol 6 (2) ◽  
Author(s):  
Afrillebar Putra Pratama ◽  
Agi Prasetiadi ◽  
Elisa Usada
Keyword(s):  
Learning Algorithm ◽  
Three Dimensional ◽  
Position Estimation ◽  
Three Dimensions ◽  
Tracking Accuracy ◽  
Object Based ◽  
Face Location ◽  
The Face ◽  
Selection Operators ◽  
Computerized System

The current presence system can be done with a computerized system, one of which is the face biometric system. This study focuses on the application of position estimation and tracking based on clustering on people's faces to determine the position in three dimensions. Position estimation can be obtained by making a kernel that is ready to be used to predict three-dimensional coordinates of faces based on two-dimensional coordinates of two images. Position estimation can be done by utilizing the Machine Learning algorithm family. In this study, Least Absolute Shrinkage and Selection Operators (LASSO) is used to perform the position estimation. Meanwhile, clustering in this study uses the K-Means algorithm. Based on the test results, the kernel error obtained in estimating the face location is 9.23 cm. The tracking accuracy of an object based on clustering is 100%.

Intelligent system of English composition scoring model based on improved machine learning algorithm

2020 ◽  
pp. 1-11
Author(s):  
Jie Liu ◽  
Lin Lin ◽  
Xiufang Liang
Keyword(s):  
Machine Learning ◽  
Evaluation System ◽  
Intelligent System ◽  
Learning Algorithm ◽  
Learning Algorithms ◽  
Machine Learning Algorithms ◽  
Assessment System ◽  
English Composition ◽  
Region Extraction ◽  
Constraint Model

The online English teaching system has certain requirements for the intelligent scoring system, and the most difficult stage of intelligent scoring in the English test is to score the English composition through the intelligent model. In order to improve the intelligence of English composition scoring, based on machine learning algorithms, this study combines intelligent image recognition technology to improve machine learning algorithms, and proposes an improved MSER-based character candidate region extraction algorithm and a convolutional neural network-based pseudo-character region filtering algorithm. In addition, in order to verify whether the algorithm model proposed in this paper meets the requirements of the group text, that is, to verify the feasibility of the algorithm, the performance of the model proposed in this study is analyzed through design experiments. Moreover, the basic conditions for composition scoring are input into the model as a constraint model. The research results show that the algorithm proposed in this paper has a certain practical effect, and it can be applied to the English assessment system and the online assessment system of the homework evaluation system algorithm system.

scholarly journals Estimating intrafraction tumor motion during fiducial-based liver stereotactic radiotherapy via an iterative closest point (ICP) algorithm

2019 ◽  
Vol 14 (1) ◽  
Author(s):  
Wu-zhou Li ◽  
Zhi-wen Liang ◽  
Yi Cao ◽  
Ting-ting Cao ◽  
Hong Quan ◽  
...  
Keyword(s):  
Stereotactic Radiotherapy ◽  
Imaging System ◽  
Liver Tumors ◽  
Three Dimensional ◽  
Rigid Motion ◽  
Iterative Closest Point ◽  
Tumor Motion ◽  
Tracking Accuracy ◽  
Icp Algorithm ◽  
Anterior Posterior

Abstract Background Tumor motion may compromise the accuracy of liver stereotactic radiotherapy. In order to carry out a precise planning, estimating liver tumor motion during radiotherapy has received a lot of attention. Previous approach may have difficult to deal with image data corrupted by noise. The iterative closest point (ICP) algorithm is widely used for estimating the rigid registration of three-dimensional point sets when these data were dense or corrupted. In the light of this, our study estimated the three-dimensional (3D) rigid motion of liver tumors during stereotactic liver radiotherapy using reconstructed 3D coordinates of fiducials based on the ICP algorithm. Methods Four hundred ninety-five pairs of orthogonal kilovoltage (KV) images from the CyberKnife stereo imaging system for 12 patients were used in this study. For each pair of images, the 3D coordinates of fiducial markers inside the liver were calculated via geometric derivations. The 3D coordinates were used to calculate the real-time translational and rotational motion of liver tumors around three axes via an ICP algorithm. The residual error was also investigated both with and without rotational correction. Results The translational shifts of liver tumors in left-right (LR), anterior-posterior (AP),and superior-inferior (SI) directions were 2.92 ± 1.98 mm, 5.54 ± 3.12 mm, and 16.22 ± 5.86 mm, respectively; the rotational angles in left-right (LR), anterior-posterior (AP), and superior-inferior (SI) directions were 3.95° ± 3.08°, 4.93° ± 2.90°, and 4.09° ± 1.99°, respectively. Rotational correction decreased 3D fiducial displacement from 1.19 ± 0.35 mm to 0.65 ± 0.24 mm (P<0.001). Conclusions The maximum translational movement occurred in the SI direction. Rotational correction decreased fiducial displacements and increased tumor tracking accuracy.

AN FPGA HARDWARE/SOFTWARE CO-DESIGN TOWARDS EVOLVABLE SPIKING NEURAL NETWORKS FOR ROBOTICS APPLICATION

2010 ◽  
Vol 20 (06) ◽  
pp. 447-461 ◽  
Author(s):  
S. P. JOHNSTON ◽  
G. PRASAD ◽  
L. MAGUIRE ◽  
T. M. MCGINNITY
Keyword(s):  
Autonomous Navigation ◽  
Intelligent System ◽  
Learning Algorithm ◽  
Spike Timing ◽  
Design Solution ◽  
Gate Arrays ◽  
Hardware Realization ◽  
Field Programmable ◽  
Programmable Gate Arrays ◽  
Avoidance Problem

This paper presents an approach that permits the effective hardware realization of a novel Evolvable Spiking Neural Network (ESNN) paradigm on Field Programmable Gate Arrays (FPGAs). The ESNN possesses a hybrid learning algorithm that consists of a Spike Timing Dependent Plasticity (STDP) mechanism fused with a Genetic Algorithm (GA). The design and implementation direction utilizes the latest advancements in FPGA technology to provide a partitioned hardware/software co-design solution. The approach achieves the maximum FPGA flexibility obtainable for the ESNN paradigm. The algorithm was applied as an embedded intelligent system robotic controller to solve an autonomous navigation and obstacle avoidance problem.

A Study of the Continuous Optimization Problem Using a Wood Robot Controlled by a Biologically Motivated System

2015 ◽  
Vol 137 (7) ◽  
Author(s):  
Jong-Chen Chen
Keyword(s):  
Physical Environment ◽  
Intelligent System ◽  
Optimization Problems ◽  
Learning Algorithm ◽  
Continuous Optimization ◽  
Physical Structure ◽  
Experimental Results ◽  
Self Organized ◽  
Selection Operators ◽  
Continuous Optimization Problems

Continuous optimization plays an increasingly significant role in everyday decision-making situations. Our group had previously developed a multilevel system called the artificial neuromolecular system (ANM) that possessed structure richness allowing variation and/or selection operators to act on it in order to generate a broad range of dynamic behaviors. In this paper, we used the ANM system to control the motions of a wooden walking robot named Miky. The robot was used to investigate the ANM system's capability to deal with continuous optimization problems through self-organized learning. Evolutionary learning algorithm was used to train the system and generate appropriate control. The experimental results showed that Miky was capable of learning in a continued manner in a physical environment. A further experiment was conducted by making some changes to Miky's physical structure in order to observe the system's capability to deal with the change. Detailed analysis of the experimental results showed that Miky responded to the change by appropriately adjusting its leg movements in space and time. The results showed that the ANM system possessed continuous optimization capability in coping with the change. Our findings from the empirical experiments might provide us another dimension of information of how to design an intelligent system comparatively friendlier than the traditional systems in assisting humans to walk.

scholarly journals Enabling Technology for Remote Prosthetic Alignment Tuning

2021 ◽  
Vol 186 (Supplement_1) ◽  
pp. 659-664
Author(s):  
David A Boone ◽  
Sarah R Chang
Keyword(s):  
Machine Learning ◽  
Learning Algorithm ◽  
Three Dimensional ◽  
Coordinate Measuring Machine ◽  
Software Application ◽  
Prosthetic Socket ◽  
Prosthetic Alignment ◽  
Prosthesis Alignment ◽  
Alignment Tool ◽  
Measuring Machine

ABSTRACT Introduction This research has resulted in a system of sensors and software for effectively adjusting prosthetic alignment with digital numeric control. We called this suite of technologies the Prosthesis Smart Alignment Tool (ProSAT) system. Materials and Methods The ProSAT system has three components: a prosthesis-embedded sensor, an alignment tool, and an Internet-connected alignment expert system application that utilizes machine learning to analyze prosthetic alignment. All components communicate via Bluetooth. Together, they provide for numerically controlled prosthesis alignment adjustment. The ProSAT components help diagnose and guide the correction of very subtle, difficult-to-see imbalances in dynamic gait. The sensor has been cross-validated against kinetic measurement in a gait laboratory, and bench testing was performed to validate the performance of the tool while adjusting a prosthetic socket based on machine learning analyses from the software application. Results The three-dimensional alignment of the prosthetic socket was measured pre- and postadjustment from two fiducial points marked on the anterior surface of the prosthetic socket. A coordinate measuring machine was used to derive an alignment angular offset from vertical for both conditions: pre- and postalignment conditions. Of interest is the difference in the angles between conditions. The ProSAT tool is only controlling the relative change made to the alignment, not an absolute position or orientation. Target alignments were calculated by the machine learning algorithm in the ProSAT software, based on input of kinetic data samples representing the precondition and where a real prosthetic misalignment condition was known a priori. Detected misalignments were converted by the software to a corrective adjustment in the prosthesis alignment being tested. We demonstrated that a user could successfully and quickly achieve target postalignment change within an average of 0.1°. Conclusions The accuracy of a prototype ProSAT system has been validated for controlled alignment changes by a prosthetist. Refinement of the ergonomic form and technical function of the hardware and clinical usability of the mobile software application are currently being completed with benchtop experiments in advance of further human subject testing of alignment efficiency, accuracy, and user experience.

scholarly journals IMPROVING UAV TELEMETRY POSITIONING FOR DIRECT PHOTOGRAMMETRY

2021 ◽  
Vol XLIII-B2-2021 ◽  
pp. 61-68
Author(s):  
L. Rossi ◽  
F. Ioli ◽  
E. Capizzi ◽  
L. Pinto ◽  
M. Reguzzoni
Keyword(s):  
Low Cost ◽  
Three Dimensional ◽  
Control Points ◽  
Three Dimensional Model ◽  
Commercial Software ◽  
Ground Control Points ◽  
Lower Accuracy ◽  
Least Squares Adjustment ◽  
Gnss Receivers

Abstract. A fundamental step of UAV photogrammetric processes is to collect Ground Control Points (GCPs) by means of geodetic-quality GNSS receivers or total stations, thus obtaining an absolutely oriented model with a centimetric accuracy. This procedure is usually time-consuming, expensive and potentially dangerous for operators who sometimes need to reach inaccessible areas. UAVs equipped with low-cost GNSS/IMU sensors can provide information about position and attitude of the images. This telemetry information is not enough for a photogrammetric restitution with a centimetric accuracy, but it can be usefully exploited when a lower accuracy is required. The algorithm proposed in this paper aims at improving the quality of this information, in order to introduce it into a direct-photogrammetric process, without collecting GCPs. In particular, the estimation of an optimal trajectory is obtained by combining the camera positions derived from UAV telemetry and from the relative orientation of the acquired images, by means of a least squares adjustment. Then, the resulting trajectory is used as a direct observation of the camera positions into a commercial software, thus replacing the information of GCPs. The algorithm has been tested on different datasets, comparing the classical photogrammetric solution (with GCPs) with the proposed one. These case-studies showed that using the improved trajectory as input to the commercial software (without GCPs) the reconstruction of the three-dimensional model can be improved with respect to the solution computed by using the UAV raw telemetry only.

Three-dimensional (3D) velocity map imaging: from technique to application

2022 ◽  
Author(s):  
Gihan Basnayake ◽  
Yasashri Ranathunga ◽  
Suk Kyoung Lee ◽  
Wen Li
Keyword(s):  
Three Dimensional ◽  
Cost Effective ◽  
Velocity Map Imaging ◽  
Imaging Method ◽  
Translational Energy ◽  
Ion Imaging ◽  
Standard Tool ◽  
Recent Advancement ◽  
Acquisition Speed ◽  
Energy And Angular Distributions

Abstract The velocity map imaging (VMI) technique was first introduced by Eppink and Parker in 1997, as an improvement to the original ion imaging method by Houston and Chandler in 1987. The method has gained huge popularity over the past two decades and has become a standard tool for measuring high-resolution translational energy and angular distributions of ions and electrons. VMI has evolved gradually from 2D momentum measurements to 3D measurements with various implementations and configurations. The most recent advancement has brought unprecedented 3D performance to the technique in terms of resolutions (both spatial and temporal), multi-hit capability as well as acquisition speed while maintaining many attractive attributes afforded by conventional VMI such as being simple, cost-effective, visually appealing and versatile. In this tutorial we will discuss many technical aspects of the recent advancement and its application in probing correlated chemical dynamics.

scholarly journals Q-learning trajectory planning based on Takagi–Sugeno fuzzy parallel distributed compensation structure of humanoid manipulator

2019 ◽  
Vol 16 (1) ◽  
pp. 172988141983020 ◽  
Author(s):  
Shuhuan Wen ◽  
Xueheng Hu ◽  
Xiaohan Lv ◽  
Zongtao Wang ◽  
Yong Peng
Keyword(s):  
Reinforcement Learning ◽  
Trajectory Planning ◽  
Trajectory Tracking ◽  
Fuzzy Controller ◽  
Learning Algorithm ◽  
Fuzzy Model ◽  
Large Error ◽  
Learning Trajectory ◽  
Tracking Accuracy ◽  
Takagi Sugeno

NAO is the first robot created by SoftBank Robotics. Famous around the world, NAO is a tremendous programming tool and he has especially become a standard in education and research. Aiming at the large error and poor stability of the humanoid robot NAO manipulator during trajectory tracking, a novel framework based on fuzzy controller reinforcement learning trajectory planning strategy is proposed. Firstly, the Takagi–Sugeno fuzzy model based on the dynamic equation of the NAO right arm is established. Secondly, the design and the gain solution of the state feedback controller based on the parallel feedback compensation strategy are studied. Finally, the ideal trajectory of the motion is planned by reinforcement learning algorithm so that the end of the manipulator can track the desired trajectory and realize the valid obstacle avoidance. Simulation and experiment shows that the end of the manipulator based on this scheme has good controllability and stability and can meet the accuracy requirements of trajectory tracking accuracy, which verifies the effectiveness of the proposed framework.

Sign in / Sign up

Export Citation Format

Share Document