• Korean Journal of Radiology
• /
• v.24 no.2
• /
• pp.155-165
• /
• 2023

Objective: Little is known about the effects of using different expert-determined reference standards when evaluating the performance of deep learning-based automatic detection (DLAD) models and their added value to radiologists. We assessed the concordance of expert-determined standards with a clinical gold standard (herein, pathological confirmation) and the effects of different expert-determined reference standards on the estimates of radiologists' diagnostic performance to detect malignant pulmonary nodules on chest radiographs with and without the assistance of a DLAD model. Materials and Methods: This study included chest radiographs from 50 patients with pathologically proven lung cancer and 50 controls. Five expert-determined standards were constructed using the interpretations of 10 experts: individual judgment by the most experienced expert, majority vote, consensus judgments of two and three experts, and a latent class analysis (LCA) model. In separate reader tests, additional 10 radiologists independently interpreted the radiographs and then assisted with the DLAD model. Their diagnostic performance was estimated using the clinical gold standard and various expert-determined standards as the reference standard, and the results were compared using the t test with Bonferroni correction. Results: The LCA model (sensitivity, 72.6%; specificity, 100%) was most similar to the clinical gold standard. When expert-determined standards were used, the sensitivities of radiologists and DLAD model alone were overestimated, and their specificities were underestimated (all p-values < 0.05). DLAD assistance diminished the overestimation of sensitivity but exaggerated the underestimation of specificity (all p-values < 0.001). The DLAD model improved sensitivity and specificity to a greater extent when using the clinical gold standard than when using the expert-determined standards (all p-values < 0.001), except for sensitivity with the LCA model (p = 0.094). Conclusion: The LCA model was most similar to the clinical gold standard for malignant pulmonary nodule detection on chest radiographs. Expert-determined standards caused bias in measuring the diagnostic performance of the artificial intelligence model.

• Proceedings of the KSME Conference
• /
• 2003.04a
• /
• pp.195-200
• /
• 2003

An expert system for estimation of fatigue properties from simple tensile data of material is developed, considering nearly all important estimation methods proposed so far, i.e., 7 estimation methods. The expert system is developed using an expert system shell, UNIK, and the knowledge base is constructed with production rules and frames. Forward chaining is employed as a reasoning method. The expert system has three major functions including the function to update the knowledge base. The performance of the expert system is tested using the 54 ${\sigma}-N$ curves consisting of 381 ${\sigma}-N$ data points obtained for 22 materials. It is found that the expert system developed has excellent performance especially for steel materials, and reasonably good for titanium alloys.

• Proceedings of the KSME Conference
• /
• 2001.06a
• /
• pp.114-119
• /
• 2001

An expert system for estimation of fatigue properties from simple tensile data of material is developed, considering nearly all important estimation methods proposed so far, i.e., 7 estimation methods. The expert system is developed to utilize for the case of only hardness data available. The knowledge base is constructed with production rules and frames using an expert system shell, UNIK. Forward chaining is employed as a reasoning method. The expert system has three major functions including the function to update the knowledge base. The performance of the expert system is tested using the 54 $\varepsilon$-N curves consisting of 381 $\varepsilon$-N data points obtained for 22 materials. It is found that the expert system developed has excellent performance especially for steel materials, and reasonably good for aluminum alloys.

• International Journal of Precision Engineering and Manufacturing
• /
• v.2 no.1
• /
• pp.62-75
• /
• 2001

The design and manufacture of injection molded polymeric parts with desired properties is a costly process dominated by empiricism including repeated modification of actual tooling. This paper presents and expert design evaluation system which can predict the mechanical performance of a molded product and diagnose the design before the actual mold is machined. The knowledge-based system synergistically combines a rule-based expert system with CAE programs. An iterative boundary pressure reflection method(IBPR) is developed to automate the cavity filling simulation program and to predict thermo-mechanical properties of a molded part precisely. Mathematical models of weldline and frozen-in molecular orientation are established to determine the spatial variation of microstructural anisotropies of a molded part from the result of cavity filling simulation. The strength ellipse is devised as and index which represents th spatial distribution of the microstructural anisotropies of a molded part, Heuristic knowledge of injection molding, flow simulation, and mechanical performance prediction is formalized as rules of an expert consultation system. The expert system interprets the analytical results of the process simulation, predicts the performance, evaluates the design and generates recommendations for optimal design alternative.

• Journal of the Microelectronics and Packaging Society
• /
• v.8 no.3
• /
• pp.11-17
• /
• 2001

The design and manufacture of injection molded polymeric parts with desired properties is a costly process dominated by empiricism, including the repeated modification of actual tooling. This paper presents an expert design evaluation system which can predict the mechanical performance of a molded product and diagnose the design before the actual mold is machined. The knowledge-based system synergistically combines a rule-based expert system with CAE programs. Heuristic knowledge of injection molding is formalized as rules of an expert consultation system. The expert system interprets the analytical results of the process simulation, predicts the performance, evaluates the design and generates recommendations for optimal design alternatives.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.4 no.6
• /
• pp.153-163
• /
• 1996

An expert system to design compressor valve systems has been developed. Design process is viewed as a chain of devisions based on the results of necessary analyses. Actual design is implemented by the interaction between the expert system and the user. In this work, it is demonstrated how a final design is achieved by utilizing the rule bases and analysis capability of the system. The structure of the rule bases and related parameter studies are also explained. Advantages of using an expert system approach for valve designs are explained using a practical example.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.16 no.2
• /
• pp.54-58
• /
• 2002

PID Controller is widely used as automatic equipment for industry. However, when a system has various characters, parameter decision and tuning for accurate control is a hard task. In this paper, expert auto-tuning PID controller using PLC is presented as away of solving this problem. Expert auto tuning algorithm is based on Ziegler-Nichols step response and expert knowledge. The test of control performance is carried out in practical speed control of Induction Motor in variable load, the experimental results suggest its superior performance.

• The Journal of Information Systems
• /
• v.1
• /
• pp.95-109
• /
• 1992

A number of success story about various application areas including manufacturing, accounting, finance, education, and engineering are reported. MIS professionals predicted that expert systems would improve the productivity enormously. However, the expert system revolution has not happened yet. Although not reported in the open society widely, there are failure stories of expert systems. Most of problems concerning expert system failure stem from the non-technical issues such as cognitive and psychological problems rather than the technical issues. We hypothesize that human factor principle enables designers to handle most of these non-technical problems elegantly and to improve the performance and acceptance of the expert systems. Major reasons for expert system failure and needs of human factors are discussed. Human factor guidelines to expert system make the prospects of the expert systems with human factors clear and understandable.

• Journal of Information Technology Services
• /
• v.12 no.1
• /
• pp.173-187
• /
• 2013

The advent of Web2.0 has believed to be the solution against many barriers in information sharing, especially wiki. Information sharing and collaboration are realized voluntarily and unselfishly with wiki. The purpose of this paper is to analyze the benefits and challenges of using wiki as a project management method in an IT/IS project. Wiki-based project management could provide project managers and members with expert knowledge collaboration for better project results. In the research model, we used TMS(Transactive Memory System) theory to define the relation of collaboration and performance of project members. Based on a survey among project members, the interactions between wiki characteristics and performance are examined in an IS project environment. Using Smart PLS 2.0, the data was analyzed to define the interactions by the structural equation modeling. From the empirical data, the mediated effect of expert knowledge collaboration is supported. We also derive the implications of wiki-based method. It is expected to bring new possibilities of Project management performance.

• Journal of Korean Society for Quality Management
• /
• v.20 no.1
• /
• pp.147-157
• /
• 1992

This article describes a prototype expert system for yarn spinning process planning in textile industry. This expert system is intended as a consultant to give the technicians interactive assistance for the appropriate process planning in accordance with used materials, required count, and other factors affected yarn spinning by means of many types of machine. Also, this system has the other function that can be compared the standard values with the measured ones for quality characteristics control. VP-EXPERT-a rule-based microcomputer expert system development tool-provides the expert system components for this development. The details of knowledge organization, rule representation, inference reasoning process, and performance of this expert system are demonstrated with the practical yarn spinning operations.