Enhanced Case-Based Reasoning Framework with Weighted Jaccard Similarity for Malnutrition Diagnosis in Toddlers

Limited information makes people have little knowledge of the early symptoms of Malnutrition in Toddlers. This disease must be treated quickly from an early age otherwise it will not get worse. This study aims to accurately diagnose the symptoms to provide fast, precise and accurate information in classifying the types of Malnutrition in Toddlers. This research is an expert system using Case Based Reasoning (CBR) method. The CBR method makes decisions from new cases based on solutions from previous cases. The data processed were 22 symptoms and 8 types of disease for 22 cases. The accuracy results are very good by being able to identify all types of malnutrition. So that this research can be used as a recommendation in speed to identify malnutrition in toddlers quickly.

Operational optimization is essential in modern industry and unsuitable operations will deteriorate the performance of industrial processes. Since measuring error and multiple working conditions are inevitable in practice, it is necessary to reduce their negative impacts on operational optimization under the case-based reasoning (CBR) framework. In this paper, a local density-based abnormal case removal method is proposed to remove the abnormal cases in a case retrieval step, so as to prevent performance deterioration in industrial operational optimization. More specifically, the reasons as to why classic CBR would retrieve abnormal cases are analyzed from the perspective of case retrieval in industry. Then, a local density-based abnormal case removal algorithm is designed based on the Local Outlier Factor (LOF), and properly integrated into the traditional case retrieval step. Finally, the effectiveness and the superiority of the local density-based abnormal case removal method was tested by a numerical simulation and an industrial case study of the cut-made process of cigarette production. The results show that the proposed method improved the operational optimization performance of an industrial cut-made process by 23.5% compared with classic CBR, and by 13.3% compared with case-based fuzzy reasoning.

An intelligent model for liver disease diagnosis

Fixtures are widely used in almost any modern manufacturing. They add directly to the cost base, impact manufacturing firms’ responsiveness and contribute to the overall product quality. Computer-aided intelligent fixture design was developed over the years to give a competitive edge to the manufacturing firms who are facing unprecedented competition and challenges. Among the techniques, case-based reasoning (CBR) method leverages previous design experience and emerged as one of the most popular methods. However, existing CBR methods are more focused on frame work building and less on detailed techniques on case retrieving, which is the central part of any CBR methods. This would inevitably impose negative impact on the overall efficiency of any CBR-based methods. In light of this, this paper presents a new case retrieving method based on a constrained common subgraph technique. This technique tracks down similar cases from a case library through comparing the maximal common subgraphs constrained by meeting fixturing functional requirement. Efficient and robust algorithms have been developed subsequently to implement this technique. The developed method can be highly effective for retrieving cases related to some manufacturing parts with complex geometry. An illustrative example, combined with other key fixture design factors, demonstrates the effectiveness of the proposed method. The presented method is intuitive and can be used in combination with existing CBR methods and well positioned for the upcoming “big data” manufacturing. Note to Practitioners —Providing capability for rapid responsiveness, enhancement of product quality, and production at low cost are the three main objectives for the wide manufacturing firms. Fixtures are widely used across manufacturing and assembly processes, and they are closely linked to all these three objectives. Today’s manufacturing enterprises face unprecedented challenges to control costs and to deal with an ever increasing number of product variants and smaller lot sizes. All these facts raise high demands on computer-aided intelligent fixture design. Case-based reasoning (CBR) method, tackling new problem by using the solution of similar past problems or through revising the previous solution, gains popularity among researchers and practitioners. However, existing CBR methods put more focus on CBR framework design while the case retrieving, the key process of any CBR methods, still relies on some basic feature attributes comparison. This would inevitably reduce the overall effectiveness of the CBR method. In view of that, this paper proposed a graph method for case retrieval based on comparing maximal common subgraphs (MCSs). Dictated by fixture functional requirement, a constraint (to match machining features and/or other user defined requirement) has been imposed during the MCS search process. This constraint turns out to be quite useful in term of reducing search space and increasing computation efficiency. Robust algorithms have been subsequently developed to implement this technique. Graph is a powerful tool to analyze structured object, the proposed method can handle some bespoke and complex fixture design cases. The presented method is intuitive and flexible and can be integrated into existing CBR frameworks to improve the allover effectiveness of current intelligent fixture design.

Medical applications have special features (interpretation of results in medical metrics, experiment reproducibility and dealing with complex data) that require the development of particular tools. The eXiT*CBR framework is proposed to support the development of and experimentation with new case-based reasoning (CBR) systems for medical diagnosis. Our framework offers a modular, heterogeneous environment that combines different CBR techniques for different application requirements. The graphical user interface allows easy navigation through a set of experiments that are pre-visualized as plots (receiver operator characteristics (ROC) and accuracy curves). This user-friendly navigation allows easy analysis and replication of experiments. Used as a plug-in on the same interface, eXiT*CBR can work with any data mining technique such as determining feature relevance. The results show that eXiT*CBR is a user-friendly tool that facilitates medical users to utilize CBR methods to determine diagnoses in the field of breast cancer, dealing with different patterns implicit in the data. Although several tools have been developed to facilitate the rapid construction of prototypes, none of them has taken into account the particularities of medical applications as an appropriate interface to medical users. eXiT*CBR aims to fill this gap. It uses CBR methods and common medical visualization tools, such as ROC plots, that facilitate the interpretation of the results. The navigation capabilities of this tool allow the tuning of different CBR parameters using experimental results. In addition, the tool allows experiment reproducibility.

Although the case-based reasoning (CBR) process is domain dependent, certain aspects of it can readily be captured into a generic framework which in turn can be applied to various engineering domains. One such exercise that has been carried out is described here. In this paper, we present the notion that CBR can be formalized and applied in a specialized framework in an integrated knowledge-based environment. We first analyze the CBR process to abstract the steps involved in the development of a CBR system. We then propose a framework in which most of these steps are formalized so that they can be applied in a domain-independent manner. The salient features of this framework, called CASETOOL (CASE-based reasoning TOOL-kit), are then described. The highlight of this approach is the use of a concept called design criticism in the CBR process. The versatility of the tool is demonstrated through an application from the bridge engineering domain.

Multiple retrieval case-based reasoning for incomplete datasets.

Health is one of the most valuable parts of human life. So healthy is the goal of every human being. Many things cause the decline in human health, such as hereditary genes, sensitive immunity and exposure to viruses or bacteria. One of the diseases caused by bacteria is tuberculosis (TB). Tuberculosis is a disease caused by exposure to a bacterium called Mycobacterium tuberculosis. There are two types of tuberculosis, namely pulmonary tuberculosis and extra pulmonary tuberculosis. Pulmonary tuberculosis can be defined as a disease that attacks the lungs and affects the lung parenchyma [4]. It's just that this type of disease does not attack other organs. While extrapulmonary tuberculosis is a tuberculosis disease in which this type of disease can attack other related organs such as the hilum, pleura and various other organs. the lack of funds for health checks makes it too late for many people to get treatment. Therefore, the development of technology should be utilized in handling this problem. One of the technologies that can be used in dealing with these problems is to use an expert system. An expert system is a system that is developed using the development of the knowledge that is owned by many experts and is used as a reference in developing the technology. In using an expert system, a method is needed that can help solve existing problems, therefore in this study the method used is the Case-Based Reasoning (CBR) method. The Case-Based Reasoning (CBR) method is the most suitable method for use in this study because the main function of this method is to diagnose disease. Based on the results of the calculation process using the Case-Based Reasoning method for each type of TB disease, the results obtained are for pulmonary tuberculosis to obtain a value of 85%, while for tuberculosis with extra pulmonary tuberculosis it is 62%. So based on the results obtained in this study it was determined that the sample was diagnosed with pulmonary tuberculosis. With a similarity of 85%.

Diseases in pregnant women are diseases that many people need to pay attention to, because diseases that occur in pregnant women will not only endanger one life, but more than that. Hypertension currently occupies the 2nd position as a type of disease that threatens the lives of many pregnant women. The application of Case Based Reasoning (CBR) in diagnosing diseases that occur during pregnancy is motivated by the difficulty of consulting an obstetrician due to costs, time or even the limited number of doctors in a hospital. The use of CBR aims to solve new problems by adapting solutions to problems that occurred before. The expert system itself is one of the solutions to solve problems faced by users in the health sector, this system can minimize costs incurred to consult about diseases in pregnancy to specialist doctors. "Use of Case Base Reasoning (CBR) to Diagnose Diseases in Pregnancy" is expected to help the general public, especially pregnant women, make a simple diagnosis of symptoms and diseases in pregnancy.

Our aim is to build an integrated learning framework of neural network and case-based reasoning. The main idea is that feature weights for case-based reasoning can be evaluated by neural networks. In this paper, we propose MBNR (Memory-Based Neural Reasoning), case-based reasoning with local feature weighting by neural network. In our method, the neural network guides the case-based reasoning by providing case-specific weights to the learning process. We developed a learning algorithm to train the neural network to learn the case-specific local weighting patterns for case-based reasoning. We showed the performance of our learning system using four datasets.

Domain knowledge ontology supports the implementation of intelligent Case Based Reasoning (CBR) systems. Standardized terminologies support efficient indexing and processing of patient data. It is an essential element for the implementation of knowledge-based clinical decision support by exploiting pre-defined semantic relationships, both hierarchical and non-hierarchical in nature. Systemized Nomenclature of Medicine-Clinical Terms (SNOMED CT) is the most comprehensive and complete terminology. This paper proposes an encoding methodology for clinical data using SNOMED CT. A case study for a diabetes diagnosis data set will be tested where SNOMED CT provides a concept coverage of ∼75% for its clinical terms. Custom codes will be provided for uncovered terms. The encoded data set is derived from electronic health record database, and it represents a case base knowledge. The collected concept IDs will be used to build a domain ontology for diabetes diagnosis CBR. This ontology contains 550 concept IDs. The encoded case base and the domain ontology can be used to build a knowledge intensive CBR.

A fuzzy-ontology-oriented case-based reasoning framework for semantic diabetes diagnosis.

Stroke is a type of cerebrovascular disease that occurs because blood flow to the brain is disrupted. Examination of stroke accurately using CT scan, but the tool is not always available, so it can be done by the Siriraj Score. Each type of stroke has similar symptoms so doctors should re-examine similar cases prior to diagnosis. The hypothesis of the Case-based reasoning (CBR) method is a similar problems having similar solution.This research implements CBR concept using Siriraj score, dense index and Jaccard Coeficient method to perform similarity calculation between cases.The test is using k-fold cross validation with 4 fold and set values of threshold (0.65), (0.7), (0.75), (0.8), (0.85), (0.9), and (0.95). Using 45 cases of data test and 135 cases of case base. The test showed that threshold of 0.7 is suitable to be applied in sensitivity (89.88%) and accuracy (84.44% for CBR using indexing and 87.78% for CBR without indexing). Threshold of 0.65 resulted high sensitivity and accuracy but showed many cases of irrelevant retrieval results. Threshold (0.75), (0.8), (0.85), (0.9) and (0.95) resulted in sensitivity (65.48%, 59.52%, 5.95%, 3,57% and 0%) and accuracy of CBR using indexing (61.67%, 55.56%, 5.56%, 3.33%, and 0%) and accuracy of CBR without indexing (62.78% 56.67%, 55.56%, 5.56%, 3.33%, and 0%).

Knowledge management and retrieval are key issues to be addressed in the medical domain, where a large amount of information is generally available, and where the expert's skills need to be properly shared across the organisation, with the aim of improving the quality of care. Case Based Reasoning (CBR) is a very well suited methodology for the Knowledge Management task, when knowledge is in the operative form. Nevertheless, also well-assessed, formalised medical knowledge, such as clinical guidelines, should be made available to physicians in order to optimise their reasoning process. To take advantage of both knowledge types, we have defined a Multi Modal Reasoning methodology, that integrates CBR and Rule Based Reasoning, for supporting context detection, information retrieval and therapy revision in diabetes care.

Case-based reasoning (CBR) is a technique which solves a problem using past experiences, where a case base stores these past experiences called cases. CBR is used to solve different kinds of problems where past information is available. With the advent of modern and efficient digital tools, huge amount of image data is being captured. Hence, huge amount of image data is stored in different application fields. Different organizations use CBR to find valuable and useful information from this image data. Therefore, in the recent times, there is a huge use of CBR in image data applications. This paper discusses the framework of CBR in image processing and reports different application models in different applications. From literature, it is observed that the performance of CBR in image processing is significant and acceptable.