Computer-aided Quality Control Research Articles

EP04.01-08 Users’ Satisfaction and Usefulness of Cloud-based Computer Aided Quality Control System using in Korean National Lung Cancer Screening Program

EP04.01-08 Users’ Satisfaction and Usefulness of Cloud-based Computer Aided Quality Control System using in Korean National Lung Cancer Screening Program

EP01.04-003 Effectiveness of Cloud-based Computer Aided Quality Control System in Korean National Lung Cancer Screening

Korean national lung cancer screening program (KNLCS) targeting high-risk smoking population using low-dose CT (LDCT) was implemented in 2019. A cloud-based quality control system (CQCS) using computer aided detection program (CAD) was used to assist radiologists in detection, measurement and categorization of lung nodules in LDCT. From 2021, Artificial Intelligence (AI)-based CAD has been launched as a developed version of CQCS. This study aimed to evaluate the effectiveness of CQCS on positive rate and inter-observer variability in KNLCS on national level.

A cell-level quality control workflow for high-throughput image analysis

BackgroundImage-based high throughput (HT) screening provides a rich source of information on dynamic cellular response to external perturbations. The large quantity of data generated necessitates computer-aided quality control (QC) methodologies to flag imaging and staining artifacts. Existing image- or patch-level QC methods require separate thresholds to be simultaneously tuned for each image quality metric used, and also struggle to distinguish between artifacts and valid cellular phenotypes. As a result, extensive time and effort must be spent on per-assay QC feature thresholding, and valid images and phenotypes may be discarded while image- and cell-level artifacts go undetected.ResultsWe present a novel cell-level QC workflow built on machine learning approaches for classifying artifacts from HT image data. First, a phenotype sampler based on unlabeled clustering collects a comprehensive subset of cellular phenotypes, requiring only the inspection of a handful of images per phenotype for validity. A set of one-class support vector machines are then trained on each biologically valid image phenotype, and used to classify individual objects in each image as valid cells or artifacts. We apply this workflow to two real-world large-scale HT image datasets and observe that the ratio of artifact to total object area (ARcell) provides a single robust assessment of image quality regardless of the underlying causes of quality issues. Gating on this single intuitive metric, partially contaminated images can be salvaged and highly contaminated images can be excluded before image-level phenotype summary, enabling a more reliable characterization of cellular response dynamics.ConclusionsOur cell-level QC workflow enables identification of artificial cells created not only by staining or imaging artifacts but also by the limitations of image segmentation algorithms. The single readout ARcell that summaries the ratio of artifacts contained in each image can be used to reliably rank images by quality and more accurately determine QC cutoff thresholds. Machine learning-based cellular phenotype clustering and sampling reduces the amount of manual work required for training example collection. Our QC workflow automatically handles assay-specific phenotypic variations and generalizes to different HT image assays.

Defect detection of seals in multilayer aseptic packages using deep learning

Sealing in aseptic packages, one of the healthiest and cheapest technologies to protect food from parasites in the liquid food industry, requires a detailed and careful control process. Since the controls are made manually and visually by expert machine operators, the human factor can lead to the failure to detect defects, resulting in high cost and food safety risks. Therefore, this study aims to perform a leak test in aseptic package seals by a system that makes decisions using independent deep learning methods. The proposed Faster R-CNN and the Updated Faster R-CNN deep learning models were subjected to training and testing with a total of 400 images taken from a real production environment, resulting in a correct classification rate of 99.25 %. As a result, it can be said that the study is the second study that performs a computer-aided quality control process with promising results, having distinctive features such as being the first study that conducts analysis using the deep learning method.

Automated dimensional inspection planning using the combination of laser scanner and tactile probe

Combining multiple sensors on CMMs (Coordinate Measuring Machines) is useful to fulfil the increasing requirements on both complexity and accuracy in dimensional metrology. Yet, the methodology to plan measurement strategies for systems combining different types of sensors is still a major challenge. Such planning is commonly done in an interactive way. This paper presents a methodology which can create inspection plans automatically for CMM inspection combining a touch trigger probe and a laser scanner. The inspection features are specified based on the extracted geometry features and the associated PMI (Product and Manufacturing Information) items from a CAD model. A knowledge based sensor selection method is applied to choose the suited sensor for each inspection feature. For touch trigger measurements, the sampling strategy considers the measurement uncertainty calculated by simulation. A geometry-guide method is developed for collision-free probing path generation. For laser scan measurements, the required view angles and positions of the laser scanner are determined iteratively, based on which the scan path is generated automatically. The proposed methodology is tested for several cases and validated by measurement experiments. The methodology provides suited planning results and can be used for automated dimensional inspection, i.e. Computer Aided Quality Control (CAQC).

Computer-Aided Quality Control for Colonoscopy: Automatic Documentation of Cecal Intubation

Computer-Aided Quality Control for Colonoscopy: Automatic Documentation of Cecal Intubation

Precipitation and phase transformation modelling to predict surface cracks and slab quality

Improved numerical simulation tools provide valuable information for equipment designers and operators of continuous casting plants. In comparison to the collection and evaluation of process data, simulation models based on physical principles are able to provide a more detailed understanding of how individual process parameters can influence the production process and final product quality. One major goal of the numerical modelling of metallurgical processes is to predict the material behaviour using a proper description of production processes and their correlation to product properties and quality. Thus, the amount of expensive experimental trials can be reduced in the process of development. Later for standard production a numerical quality assessment is appropriate to reduce inspection efforts. Utilising commercial and in house made simulation software, the continuous casting process is simulated regarding thermal, mechanical, and phase transformation aspects. In a practical example, metallographic observations of strand surface defects, such as transverse corner cracks, are correlated to material history during the production process. As a result, the initiation of material failure can be localised more precisely and countermeasures can be taken with a higher chance of success compared to models that use only linear combinations of steel analysis and process parameters. The applicability of these simulation tools for computer aided quality control (CAQC) systems is discussed.

A contemplative stance on the automation of continuous casting in steel processing

Abstract This paper gives a contemplative stance on the automation of continuous casting in steel processing. It gives a brief overview of the control techniques used in this industry, typical applications of such techniques, and discusses supportive technologies that form a critical part of the implementation of such techniques. The casting process and operational issues are described. Particular attention is paid to production planning and scheduling, computer-aided quality control, and the crucial role that instrumentation plays in the automation of a caster. The importance of integrating casting automation systems into plant wide production systems, and the development of new instrumentation, are highlighted.

An approach to computer-aided quality control based on 3D coordinate metrology

The quality of manufactured products usually needs to be verified. This paper presents an advanced CAQ approach to compare manufactured objects with reference data from underlying CAD models. First, an overview about the current state-of-the-art in optical 3D measurement techniques is given. After that the research method adopted in this paper is discussed. Furthermore, a software prototype of the presented approach in which a stripe projection system with combined gray-code and phase shifting is described. With this equipment, 3D shapes of objects or manufactured products can be measured. In order to compare the 3D data (represented in sensor coordinate system) a registration to the CAD coordinate system is needed. At first, the selection of a starting point for the orientation parameters is described. For the registration process different numerical algorithms are used to minimize a distance function. To achieve a better performance, an optimization process based on 3D voxel arrays is introduced. After the registration process, several parameters for the kind of geometric displacement can be calculated and visualized. For objects that cannot be measured from one direction, a pair-wise registration as well as a global registration have been developed. Furthermore, some rapid prototyping examples to which our CAQ approach has been applied are presented. Those examples show that our method works well in practice. Finally, some application fields for the CAQ approach presented here are outlined.

A contemplative stance on the automation of continuous casting in steel processing

This paper gives a contemplative stance on the automation of continuous casting in steel processing. It gives a brief overview of the control techniques used in this industry, typical applications of such techniques, and discusses supportive technologies that form a critical part of the implementation of such techniques. The casting process and operational issues are described. Particular attention is paid to production planning and scheduling, computer-aided quality control, and the crucial role that instrumentation plays in the automation of a caster. The importance of integrating casting automation systems into plant wide production systems, and the development of new instrumentation, is highlighted.

A proposed control system/CAQC methodology and prediction system for the improvement of surface defects in the continuous casting of slabs

A control system or computer aided quality control system methodology is presented to determine surface defects in the continuous casting of steel slabs. When no defects are present, slabs can be either hot charged or direct rolled, thereby increasing the throughput of the system and decreasing energy consumption.

Data modelling with IDEF1x

The idea of integrating business, product development, and manufacturing activities has gained momentum in manufacturing industry. To remain competitive, companies must optimize the utilization of their resources. This integration of various functional areas, e.g. computer-aided design, computer-aided process planning, computer-aided production planning and control, computer-aided quality control, and computer-aided manufacturing requires a large volume of data to be stored by the users contributing to those areas. IDEF1x provides a formal framework for consistent modelling of the data necessary for this integration. This paper provides insights into the IDEF1x modelling technique. The paper concludes with a discussion of several relevant research issues.

Design for automating the inspection of manufacturing parts

The integration of quality into management of advanced manufacturing systems has not typically included automated approaches. With the advent of automated inspection and quality tools such as coordinate measuring machines and other dimensional measuring equipment, the linkage with integrated manufacturing systems seems natural. This paper shows the functional relationship of computer aided process planning to automated inspection process planning. The series of functional activities within each of these frameworks are mapped in this paper. The operational environment to include computer aided quality control in an integrated manufacturing system is also defined.

Selection of acceptance sampling plans with multi - attribute defects in computer-aided quality control

Abstract A multicriteria quality control model for acceptance sampling is developed which defines several conflicting and non-commensurate objectives. The following objectives are proposed and quantified in this model: (1) to minimize the cost of inspection, (2) to minimize the number of critical defects, (3) to minimize the number of major defects, and (4) to minimize the number of minor defects. The interactive paired comparison method identifies the most desirable sample sizes and acceptance levels for critical, major, and minor defects for a given lot size. At each iteration a small number of alternatives is presented to a decision maker. The decision maker chooses the most favourable alternative by expressing his/her strength of preference for pairs of alternatives. An unknown non-linear utility function of objectives is assumed. The optimal sampling plan is determined using a minimal number of interactions. A computer package is developed and used to determine the optimal sampling plan for an applic...

Growth, crisis and future of the factory

The process of industrialization in the 19th century was coupled with the existence of the factory. The organization of the work within factories through the combination of manual and mechanical activities led to a substantial productivity increase and a significant improvement in living standards. The satisfaction of basic needs and consumer demands led initially to the refinement of products and production processes; while at a later stage they contributed to the crisis of the factory. Presumably a new period of growth for the factory starts due to the thrust of innovation in the information technology. The key element of the future production renewal is the computer integrated and flexible automated factory of the future, which provides a continuous production flow and ties together energy, materials and information. Substantial parts of the computer integrated production are communication networks, intelligent generation systems, computer-aided design (CAD), computer-aided planning (CAP), computer-aided manufacturing (CAM), computer-aided quality control (CAQ) and stimulation systems. These combine intelligent machine-tools, intelligent robots, intelligent assembly machines, intelligent inspection machines and intelligent means of transportation within the factory of the future. The artificial intelligence and the possibility of remote control due to the developments in the information technology are the prerequisites for the setting-up of new production schemes. Next to the central organized factories, new decentralized production structures can be developed which would make possible drastic changes in the social life.