Engineering Commons^™

A Neural-Genetic Algorithm For Feature Selection And Breast Abnormality Classification In Digital Mammography, Ping Zhang, Brijesh Verma, Kuldeep Kumar

Kuldeep Kumar

Digital mammography is one of the most suitable methods for early detection of breast cancer. It uses digital mammograms to find suspicious areas. However, it is very difficult to distinguish benign and malignant cases, especially for the small size lesions in the early stage of cancer. This is reflected in the high percentage of unnecessary biopsies that are performed and many deaths caused by late detection or misdiagnosis. A computer based feature selection and classification system can provide a second opinion to the radiologists. This work proposes a neural-genetic algorithm for feature selection in conjunction with neural network based classifier. …

Neural Vs. Statistical Classifier In Conjunction With Genetic Algorithm Based Feature Selection, Ping Zhang, Brijesh Verma, Kuldeep Kumar

Kuldeep Kumar

Digital mammography is one of the most suitable methods for early detection of breast cancer. It uses digital mammograms to find suspicious areas containing benign and malignant microcalcifications. However, it is very difficult to distinguish benign and malignant microcalcifications. This is reflected in the high percentage of unnecessary biopsies that are performed and many deaths caused by late detection or misdiagnosis. A computer based feature selection and classification system can provide a second opinion to the radiologists in assessment of microcalcifications. The research in this paper proposes and investigates a neural-genetic algorithm for feature selection in conjunction with neural and …

A Hybrid Classifier For Mass Classification With Different Kinds Of Features In Mammography, Ping Zhang, Kuldeep Kumar, Brijesh Verma

Kuldeep Kumar

This paper proposes a hybrid system which combines computer extracted features and human interpreted features from the mammogram, with the statistical classifier’s output as another kind of feature in conjunction with a genetic neural network classifier. The hybrid system produced better results than the single statistical classifier and neural network. The highest classification rate reached 91.3%. The area value under the ROC curve is 0.962. The results indicated that the mixed features contribute greatly for the classification of mass patterns into benign and malignant.

Neural Vs Statistical Classifier In Conjunction With Genetic Algorithm Feature Selection In Digital Mammography, Ping Zhang, Brijesh Verma, Kuldeep Kumar

Kuldeep Kumar

Digital mammography is one of the most suitable methods for early detection of breast cancer. It uses digital mammograms to find suspicious areas containing benign and malignant microcalcifications. However, it is very difficult to distinguish benign and malignant microcalcifications. This is reflected in the high percentage of unnecessary biopsies that are performed and many deaths caused by late detection or misdiagnosis. A computer based feature selection and classification system can provide a second opinion to the radiologists in assessment of microcalcifications. The research proposes and investigates a neural-genetic algorithm for feature selection in conjunction with neural and statistical classifiers to …

Effect Of Superconsciousness External Energy On Atomic, Crystalline And Powder Characteristics Of Carbon Allotrope Powders, Mahendra Kumar Trivedi, Rama Mohan Tallapragada

Mahendra Kumar Trivedi

Characterization Of Illinois Aggregates For Subgrade Replacement And Subbase, Erol Tutumluer, Debakanta Mishra, Abbas A. Butt

Debakanta Mishra

Construction of a pavement working platform is often needed on soft, unstable soils to provide sufficient stability and adequate immediate support for equipment mobility and paving operations without developing excessive rutting. The aggregate type and quality are important factors for determining the required treatment/ replacement thickness. The overall objective in this project has been to characterize strength, stiffness, and deformation behavior of three crushed and uncrushed aggregate materials, i.e., limestone, dolomite and uncrushed gravel, commonly used in Illinois for subgrade replacement and subbase. The goal has been to develop aggregate thickness correlations with aggregate properties to modify and improve the …

Main Speaker, Kevin Cavicchi

Kevin Cavicchi

No abstract provided.

Mineralization Of Phenol In An Improved Photocatalytic Process Assisted With Ferric Ions: Reaction Network And Kinetic Modeling, Jesus Moreira Del Rio

Jesus Moreira del Rio

No abstract provided.

Polymer , Yu Zhu

Yu Zhu

No abstract provided.

Chemistry Of Materials , Yu Zhu

Yu Zhu

No abstract provided.

Ultrasensitive Polymer Photodetectors, Xiong Gong

Xiong Gong

No abstract provided.

Hysteresis And Reaction Characterization Of Methane Catalytic Partial Oxidation On Rhodium Catalyst, Wei-Hsin Chen

Wei-Hsin Chen

No abstract provided.

An Evaluation On Rice Husks And Pulverized Coal Blends Using A Drop Tube Furnace And A Thermogravimetric Analyzer For Application To A Blast Furnace, Wei-Hsin Chen

Wei-Hsin Chen

No abstract provided.

On The Issue Of Decoupled Decoding Of Codes Derived From Quaternion Orthogonal Designs, Tadeusz Wysocki, Beata Wysocki, Sarah Spence Adams

Sarah Spence Adams

Quaternion orthogonal designs (QODs) have been previously introduced as a basis for orthogonal space-time polarization block codes (OSTPBCs). This note will serve to correct statements concerning the optimality of a decoupled maximum-likelihood (ML) decoding algorithm. It will be shown that when compared to coupled decoding, the decoupled decoding is only optimal in certain cases. This raises several open problems concerning the decoding of OSTPBCs.

Photocatalytic Reactor Efficiency Factor For The Complete Spam Of Reaction Times, Benito Serrano Rosales, Jesus Moreira Del Rio, Hugo De Lasa

Jesus Moreira del Rio

No abstract provided.

Modelling Of Radiation Transfer And Adsorption In Photocatalytic Reactors, Jesus Moreira Del Rio

Jesus Moreira del Rio

No abstract provided.

Polymer Photodetector, Xiong Gong

Xiong Gong

No abstract provided.

Research Themes In Hpt: A Content Analysis Of The Ispi Journals, Linda Huglin

Linda Huglin

No abstract provided.

Solution-Processed Ultrasensitive Polymer Photodetectors, Xiong Gong

Xiong Gong

No abstract provided.

Transient Dynamic Of Hydrogen Permeation Through A Palladium Membrane, Wei-Hsin Chen

Wei-Hsin Chen

No abstract provided.

Learning And Forgetting In Maintenance Outsourcing, Hakan Tarakci, Sunantha Teyarachakul, Kwei Tang

Hakan Tarakci

We study the effects of learning and forgetting on maintenance outsourcing in this paper. This is an extension of the paper by Tarakci et al. (2009), which analyzed the effects of learning on maintenance outsourcing. In our model, a manufacturer offers a short-term outsourcing contract, where payment includes a fixed value along with cost subsidization for each preventive maintenance activity, to an external contractor. The contractor schedules and performs preventive maintenance activities as well as repairs the system if there is a failure. We consider two types of learning: natural learning and learning by costly efforts. We then analyze the …

Comparison Of Three Bacterial Toxicity Assays For Imidazolium-Derived Ionic Liquids, P. Frymier, M. Azimova, Samuel Morton

Samuel A Morton

No abstract provided.

Energy Efficiency In Photocatalytic Reactors For The Full Span Of Reaction Times, Jesus Moreira Del Rio, Benito Serrano Rosales, Aaron Ortiz Gomez, Hugo De Lasa

Jesus Moreira del Rio

No abstract provided.

On The Problem Of Production Deadline And Maintenance Outsourcing, Sharafali Moosa, Hakan Tarakci

Hakan Tarakci

In this paper we consider a production-maintenance problem in which a buyer and a supplier have already entered into a contractual relationship for a firm delivery date. Our focus is on the production-related decisions of the supplier rather than the nature of the contract itself. We assume that production rate is actually constant as long as the system is up and running but randomness arises in production due to downtimes as a result of unpredictable failures (breakdowns) and scheduled preventive maintenance activities. Production maintenance is outsourced. As production output is random due to unreliable production facilities, the supplier needs to …

“Optimizing The Performance And Manufacturability Of Silicon Diffractive Optics”, Co-Investigtor, Nasa Astronomy And Physics Research And Analysis, $670k, 2009-2012., Weisong Wang

Weisong Wang

No abstract provided.

Position Adaptive Direction Finding, Raúl Ordóñez

Raúl Ordóñez

PI, funded by AFRL/RYRR, $88000, year: 2009-2010.

The Convergence In Workers’ Skill Levels Under Learning And Forgetting: The Fixed-Point-Property Approach, Sunantha Teyarachakul, Dogan Comez, Hakan Tarakci

Hakan Tarakci

This paper presents a study on the convergence of workers’ skill levels under learning and forgetting in processing time in a batch-manufacturing environment. The convergence properties are examined under assumptions of an infinite horizon, a constant demand rate, and a fixed lot size. Our work extends the convergence results of Teyarachakul, Chand, and Ward (2008) beyond Globerson and Levin’s (1987) exponential forgetting function and Wright’s (1936) learning curve to more general classes of learning and forgetting functions. We also discuss why early papers other than Teyarachakul, Chand, and Ward (2008) did not find other types of long-term behaviors beyond convergence …

Human Microvasculature Fabrication Using Thermal Inkjet Printing Technology, Xiaofeng Cui, Thomas Boland

Thomas Boland

The current tissue engineering paradigm is that successfully engineered thick tissues must include vasculature. As biological approaches alone, such as VEGF, have fallen short of their promises, one may look for an engineering approach to build microvasculature. Layer-by-layer approaches for customized fabrication of cell/scaffold constructs have shown some potential in building complex 3D structures. With the advent of cell printing, one may be able to build precise human microvasculature with suitable bio-ink. Human microvascular endothelial cells (HMVEC) and fibrin were studied as bio-ink for microvasculature construction. Endothelial cells are the only cells to compose the human capillaries and also form …

Electrophysiological Characterization Of Embryonic Hippocampal Neurons Cultured In A 3d Collagen Hydrogel, Tao Xu, Peter Molnar, Gregory Cassie, Mainak Das, Thomas Boland, Jay Hickman

Thomas Boland

Rat embryonic hippocampal neurons were cultured in (1) 3D collagen hydrogels as 'entrapped' evenly distributed cells, (2) at the interface of two collagen layers (sandwich model), and (3) on the surface of collagen coated coverslips (2D model). In the 'entrapment' model the neuronal processes grew out of the plane of the cell body and extended into the collagen matrix, in contrast to the sandwich model where the cells and their processes rarely left the plane in which they were seeded. Hippocampal neurons 'entrapped' in the 3D collagen gel grew the same number, but shorter, processes and exhibited improved survival compared …

Fabrication And Characterization Of Bio-Engineered Cardiac Pseudo Tissues, Tao Xu, Catalin Baicu, Michael Aho, Michael Zile, Thomas Boland

Thomas Boland

We report to fabricate functional three-dimensional (3D) tissue constructs by using an inkjet based bio-prototyping method. With the use of the modified inkjet printers, contractile cardiac hybrids that exhibit the forms of the 3D rectangular sheet and even the ‘half heart’ (with two connected ventricles) have been fabricated by arranging alternate layers of biocompatible alginate hydrogels and mammalian cardiac cells according to pre-designed 3D patterns. In this study, primary feline adult and H1 cardiomyocytes were used as model cardiac cells. Alginate hydrogels with controlled micro-shell structures were built by spraying cross-linkers in micro-drops onto un-gelled alginic acid. The cells remained …