A5076782497- Handwritten Text Recognition Techniques
- Medical Image Segmentation Techniques
- Image and Object Detection Techniques
- Enzyme Structure and Function
- Image Retrieval and Classification Techniques
- Advanced Software Engineering Methodologies
- Protein Structure and Dynamics
- Multimodal Machine Learning Applications
- Bioinformatics and Genomic Networks
- Spectroscopy Techniques in Biomedical and Chemical Research
- Mass Spectrometry Techniques and Applications
- Advanced Neural Network Applications
- Video Analysis and Summarization
- RNA and protein synthesis mechanisms
- MRI in cancer diagnosis
- Software Engineering Research
- Usability and User Interface Design
- Image Processing and 3D Reconstruction
- Multi-Agent Systems and Negotiation
- Advanced Image and Video Retrieval Techniques
- Digital Accessibility for Disabilities
- Atomic and Subatomic Physics Research
- Evolution and Genetic Dynamics
- Photochemistry and Electron Transfer Studies
- Gaze Tracking and Assistive Technology
Technische Informationsbibliothek (TIB)
2019-2021
George Mason University
2017-2019
Stevens Institute of Technology
2002-2003
University of Michigan
2002
In recent years, the problem of scene text extraction from images has received extensive attention and significant progress. However, scholarly figures such as plots charts remains an open problem, in part due to difficulty locating irregularly placed lines. To best our knowledge, literature not described implementation a system for that adapts deep convolutional neural networks used detection. this paper, we propose approach forgoes preprocessing favor using network line localization. Our...
B1 inhomogeneity and non-ideal slice profiles introduce contributions from stimulated indirect echoes into the multi-echo spin-echo (MESE) T 2 mapping sequence, leading to variations in quantitative values across scanners vendors despite use of a harmonised scan protocol. This study used an EPG-based fitting method include corrections for stimulated-echo effects, applied phantom ‘travelling kidney’ data collected on three different MR (GE, Philips, Siemens). Compared with...
The ability of the Ada programming language to support a software-engineering approach building fault-tolerant applications systems by distributed redundant abstract data types (ADTs) is discussed. It shown that package, separation specification from implementation parts, typing and exception handling mechanisms, nondeterministic rendezvous contribute fault-tolerance constructed in this fashion.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">></ETX>
Manual segmentation of the kidneys is very time consuming and reader dependent, this renders measurements total kidney volume (TKV) in large multi-site studies impractical. Here we use a convolutional neural network (CNN), trained on data from single MRI vendor, to segment volunteers scanned with harmonised FSE image protocol MR scanners three different vendors (GE, Philips Siemens). The were manually segmented by two readers, both which demonstrated significant difference TKV across...
This paper investigates thinning as applied to feature-based, off-line handwritten cursive and printed characters. In this report, the sources of errors introduced by conventional processes are considered. A novel method reduce these using gray level Gaussian filtering etching is presented justified. Experimental results given.
With biomolecular structure recognized as central to understanding mechanisms in the cell, computational chemists and biophysicists have spent significant efforts on modeling dynamics. While advances been made, particularly design of sophisticated energetic models molecular representations, such are experiencing diminishing returns. One culprits is low exploration capability. The impasse has attracted AI researchers offer adaptations robot motion planning algorithms for structures motions....
Reconstructing the energy landscape of a protein holds key to characterizing its structural dynamics and function [1]. While disparate spatio-temporal scales spanned by slow challenge reconstruction in wet dry laboratories, computational efforts have had recent success on proteins where wealth experimentally-known structures can be exploited extract modes motion. In [2], authors propose SoPriM method that extracts principle components (PCs) utilizes them as variables structure space...
Nearly all cellular processes involve proteins structurally rearranging to accommodate molecular partners. The energy landscape underscores the inherent nature of as dynamic molecules interconverting between structures with varying energies. In principle, reconstructing a protein's holds key characterizing structural dynamics and its regulation protein function. practice, disparate spatio-temporal scales spanned by slow challenge both wet dry laboratories. However, growing number deposited...
As software system have become more complex, the need for better abstraction mechanisms has led to widespread use of abstract data types (ADT's). A measure ADT design complexity is an important factor in describing, explaining, evaluating, and predicting key aspects development maintenance systems that them. This paper discusses a novel method measuring through process proving its correctness. Then, relative ADT's expressed terms abstract-to-concrete representation mappings. Comparisons...