Search

Text summarization in medical domain is one of the most crucial chores as it deals with the critical human information. Consequently the proper summarization and key point extraction from medical deeds using pre-trained Language models is now the key figure to be focused on for the researchers. But due to the considerable amount of real-world data and enormous amount of memory requirement to train the Large Language Models (LLMs), research on these models become challenging. To overcome these challenges multiple prompting and tuning techniques are being used. In this paper, effectiveness of prompt engineering and parameter efficient fine tuning is being studied to summarize the Hospital Discharge Summary (HDS) papers effectively, so that these models can accurately interprete medical terminologies and contexts, generate brief but compact summaries, and draw out concentrated themes, which opens new approaches for the application of LLMs in healthcare and making HDS more patient-friendly. In this research LLaMA 2 (Large Language Model Meta AI) has been considered as the base model. Also, the model has been fine-tuned using QLoRA (Quantized Low Rank Adapters), which can bring down the memory usage of LLMs without compromising the data quality. This study explores the way to use LLMs on HDS datasets without the hassle of memory usage using QLoRA, into electronic health record systems to further streamline the handling and retrieval of healthcare information. • Presents a method to summarize HDS by parameter-efficient fine-tuning to a LLM. • It utilizes QLoRA fine-tuning on the LLAMA 2 to minimize the memory requirements. • A comparative analysis of prompt engineering and parameter-efficient fine-tuning is performed. • The effectiveness of the approach is evaluated using BLEU and ROGUE-L scores. • Medical professionals have conducted a quality assessment of the summarization. [ABSTRACT FROM AUTHOR]

Due to the hidden nature and complexity of resistance spot welding weld nugget formation, how to avoid the time-consuming and money-consuming problem of traditional defect diagnosis methods and accurately grasp the weld nugget status is still an urgent problem. In this paper, an improved GAN model is proposed to solve the corresponding problem by combining the weld nugget defects with the dynamic resistance curve. Aiming at the problem that traditional GAN algorithms are prone to pattern collapse, this paper utilizes a variational autoencoder integrated with a channel attention mechanism as the generator part of the generative adversarial network, which helps the model pay better attention to the high-weight part of the defective sample data and combines the encoding and decoding processes to highlight defective features, thus reconstructing the defective samples with higher quality. Convolutional neural networks are then utilized to identify the features of the generated samples and diagnose the type of weldment defects. The test results show that the proposed scheme is highly reliable and the model outperforms other schemes in diagnosing welded nugget defects under the same conditions, avoiding undesirable effects such as underfitting. The validation of the actual dataset shows that, compared with other diagnostic methods that generally have an accuracy rate of less than 75%, the accuracy of the weld nugget defects diagnosis of this paper's method reaches more than 94%, which is a positive impetus to the development of auto body welding diagnosis. • Provides guidance for diagnosis in the presence of small samples of weld nugget defects • Proposes a method for diagnosing weld defect types that incorporates the characteristics of the spot-welding process. • An improved generative adversarial network is used to reconstruct weld nugget samples with defect characteristics. • Performance comparison demonstrates the effectiveness of the proposed method in diagnosing weld nugget defect types. [ABSTRACT FROM AUTHOR]

Credit risk prediction can provide essential tools for use in commercial banking credit and credit-related decision-making. This paper proposes a three-way decision method based on prospect theory and evidence theory for predicting credit risk. The first problem in this study is determining the optimal classification boundary, and the second is effectively predicting the sample default status within an uncertain boundary. To address the limitation of the SVDD model, which is that it does not consider the aggregation degree, a new sample-weighted support vector data description (SW-SVDD) model is constructed by ranking samples according to their relative membership degree. The classification boundaries of the default and nondefault samples are determined according to the maximum prediction accuracy of the SW-SVDD model. The samples are divided into definite boundary nondefault, definite boundary default, and uncertain boundary samples. The default status of samples falling into definite boundaries is predicted by the SW-SVDD model. The three-way decision method combining prospect theory and evidence theory predicts the default status of samples falling into uncertain boundaries. This paper also proposes a new interpretability method based on the default probability, nondefault probability, and optimal threshold point obtained by the three-way decision model. The empirical results show that the proposed three-way decision method is a model that balances accuracy and interpretability. It has a higher classification performance than traditional models and can reveal the key features that lead to each customer's default. The proposed three-way decision method can enhance the accuracy and reliability of risk assessments, enabling financial institutions to make more informed lending decisions and more effectively manage credit portfolios. • A model based on the three-way decision is proposed to predict credit risk. • SW-SVDD model is used to determine the optimal classification boundaries. • The three-way decision is used to predict samples within the uncertain boundary. • The proposed model can balance accuracy and interpretability. • Empirical results in five datasets confirm the robustness of the proposed model. [ABSTRACT FROM AUTHOR]

In contemporary healthcare systems, the selection and judicious choice of the right medical device becomes paramount consideration for accurate diagnosis, patient care, mitigate risk and cost incurred on diagnosis of disease. Moreover, medical devices are requisite indispensable tools to help healthcare professionals for effective diagnosis, selection of the right treatment amongst different options, and continuous monitoring of the patient. Thus, to select the best manufacturing company is as equally important to achieve the desired goals. This paper presents a decision making algorithm based on fuzzy theory that helps the decision maker in selecting the most suitable worldwide medical device manufacturer. Fuzzy theory is extensively used in managing databases and analyzing data to summarize information and draw meaningful conclusions. This paper utilizes the aggregation operators within the context of the complex q-rung orthopair fuzzy environment by utilizing Yager t-norm and t-conorm operations. Some aggregation operators are applied, namely, complex q-rung orthopair fuzzy Yager weighted geometric operator, complex q-rung orthopair fuzzy Yager ordered weighted geometric operator, complex q-rung orthopair fuzzy Yager weighted averaging operator and complex q-rung orthopair fuzzy Yager ordered weighted averaging operator. Additionally, the notable properties of these operators are proved. These aggregation operators are generated specifically for multi-criteria decision-making. Decision making algorithm is used to choose the most relevant medical device manufacturer (alternative) to validate its efficacy. The results demonstrate the versatility of parameters, adaptability to multi-criteria decision making, and efficiency of the proposed aggregation operators and algorithm. Test with specific criteria is also conducted to verify the legality and validity of the decision making algorithm. At last, the acquired results underwent a comparative analysis against various established methodologies, confirming the method's capacity to produce precise and accurate outcomes. • Multi criteria decision making algorithm is devised in fuzzy environment. • Aggregation operators are introduced and notable properties are investigated. • Test with specific criteria is conducted for the validity of proposed operators. • The algorithm is utilized in medical industry to choose best device manufacturer. [ABSTRACT FROM AUTHOR]

Rough set is an important method for dealing with incomplete information systems. In incomplete information systems, the most common way to determine the relation between two samples is the tolerance relation. However, the condition for the tolerance relation to determine those samples may belong to the same category is very lenient, which makes the reduction rate low when using the rough set generated by this relation to select features. In response to the above problems, we design the neighborhood equivalence tolerance relation to solve them. Different from other improved tolerance relations, firstly, the relation designed in this paper does not require additional threshold to accomplish the above goals, which will avoid the trouble caused by the given threshold. Secondly, we notice that most of the current improvements for this kind of problems are computationally cumbersome, and the relation designed in this paper is simple and effective. Based on this, we construct a neighborhood rough set model that handles incomplete information by using this relation, introduce its properties, expound the properties that a reduction set should satisfy, quantify the importance of conditional attributes with attribute dependence degree, which provides the basis for the design of feature selection algorithm. Finally, the greedy strategy is used to design a forward feature selection algorithm. Experimental results show that the model is effective in dealing with incomplete information systems. The feature selection algorithm has the smallest size of the average reduced subset on twelve datasets, and maintains the accuracy of the classifier, which verifies that the feature selection algorithm can effectively deal with incomplete information systems. ● As the judgment of tolerance relation is very lenient, the existing improvement methods either rely on the probability distribution of attribute values in the information system, or need to give a additional threshold. To solve the above problems, this paper improves the tolerance relation, and the improved binary relation only relies on the existing information system itself, without giving additional threshold. ● Rough set in incomplete information systems can only select more attributes to refine the knowledge granules when making feature selection due to the lenient judgment of tolerance relation, which is equivalent to making up for the problems caused by tolerance relation when selecting attributes, but this will lead to low reduction rate. In this paper, we use the newly proposed binary relation to construct neighborhood rough set model in incomplete information systems, and use this model to make feature selection, which improves the problem of low reduction rate above, while maintaining the accuracy of the classifier. [ABSTRACT FROM AUTHOR]

Schizophrenia (SCZ) is a severe mental and debilitating neuropsychiatric disorder that disrupts a person's thought processes, emotions, and behavior. Due to misdiagnosis, self-denial, and social stigma, many SCZ cases go untreated. Magnetic resonance imaging (MRI) is an excellent noninvasive tool for soft tissue contrast imaging because it provides crucial data on tissue structure size, position, and shape. The Resnet50 network is a deep residual learning framework used for feature extraction. Random-vector functional link network (RVFL) is an example of a single-hidden-layer feedforward network in which input features and hidden layer features are fed to the output layer. In this paper, we introduced a kernel ridge regression-based random vector functional link (KRR-RVFL) classifier which focuses on addressing the linearity issues in RVFL by designating the kernel function in the input layer for the precise diagnosis of SCZ. The genetic algorithm (GA) seeks to minimize the loss function by optimizing the weights and biases of the KRR-RVFL network. The classification performance is investigated on the SCZ and cognitive normal (CN) subjects, collected from the available open neuro platform, including 99 participants. The results of the suggested network show superior performance to the recent state-of-the-art networks in terms of accuracy 93.66%, sensitivity 92.22%, specificity 95.17%, precision 95.33%, F-measure 93.74%, and G-mean 93.68%. The performance metrics demonstrated the applicability of this framework for assisting clinicians in the automatic, precise evaluation of SCZ. • This paper proposes Schizophrenia diagnosis model using KRR-Opt RVFL classifier. • ResNet50 is deployed for the extraction of complex features from pre-processed data. • GA is incorporated with KRR-RVFL to optimize the parameters of the classifier. • The performance of the proposed KRR-Opt RVFL is compared with the state-of-the-art classifiers. [ABSTRACT FROM AUTHOR]

While sustainability is recognized as a crucial aspect of supply chain (SC) management, its implementation within a SC is often fraught with challenges. These challenges encompass possible causal relationships and trade-offs among sustainable SC (SSC) practices, spanning all three dimensions of sustainability. Furthermore, the increasing importance of consumers' point of view in the rapid and widespread communication era highlights the imperative to integrate this view into the SC's sustainability efforts. This paper proposes a comprehensive model for assessing and managing the implementation of sustainability in the SC while effectively tackling these challenges and requirements. Specifically, across three distinct stages, the model enables (1) identifying the SSC practices from the consumers' point of view and ranking them based on their contribution to the overall sustainability; (2) determining the causal relationships among the practices and assessing the sustainability status of the SC; and (3) prioritizing the practices and developing improvement scenarios to improve the overall sustainability of the SC. To support its objectives, the proposed model employs social media analysis and scenario planning, along with three key techniques, including the Fuzzy Best-Worst Method (FBWM) for ranking the practices, Fuzzy Cognitive Mapping (FCM) for determining the possible causal relationships among the practices and assessing the sustainability status, and Decision-Making Trial and Evaluation Laboratory (DEMATEL) for segmenting the practices based on their cause-and-effect relationships. Finally, the paper presents an empirical study in which the proposed model is applied to a mobile phone and smartphone (MPSP) SC, representing one of the most commonly consumed products in contemporary human life. • Introducing a 3-stage model for sustainable supply chain management, offering an approach for businesses. • The model considers trade-offs and causal relationships among sustainability practices, integrating consumer's opinion. • The model integrates FBWM, FCM, DEMATEL, social media analysis, and scenario planning for sustainability management. • Step-by-step design and multitasking mechanism of the model enhance its adaptability and scalability. • An empirical study in a mobile phone and smartphone supply chain demonstrates model's real-world effectiveness. [ABSTRACT FROM AUTHOR]

The metaverse, constructed through digital technology, serves as a virtual realm intertwining with reality. Within this context, the challenge of evaluating data from diverse sources arises, and the application of large-scale group decision-making (LSGDM) methods emerges as a viable solution. Handling incomplete information and reducing dimensionality for large-scale decision-makers (DMs) is crucial in addressing complex decision-making problems. Moreover, addressing missing data is a fundamental and pivotal concern in tackling real-world decision challenges, given the ubiquitous presence of information gaps that cannot be straightforwardly integrated into decision models. Besides, the intricacies of LSGDM amplify this challenge by introducing a wealth of DMs, thereby augmenting the complexity and diversity of decision-related information. This paper proposes an approach to supplement missing data by double-dimensions. This paper explores various facets of similarity relationships within the data to enhance data completeness. Additionally, this paper categorizes DMs into clusters based on their relevance and establishes a two-stage consensus-reaching process (CRP) that takes into account both group sizes and individual consensus contributions. These CRPs play a crucial role in enhancing the overall consistency and consensus within the decision group. Subsequently, this paper applies a robust decision-making method rooted in MULTIMOORA (Multi-Objective Optimization by Ratio Analysis plus the complete MULTIplicative form) to rank decision objects. Finally, this paper employs this proposed methodology in a practical case study that involves evaluating the operational status of a metaverse's urban construction metro system. Following these considerations, a comprehensive stability analysis of relevant parameters is conducted to guarantee the robustness and reliability of the decision-making process. • A similarity-based double-dimensional patching method for absent values is proposed. • A clustering-based dimensionality reduction for decision-makers is developed. • A two-stage consensus-reaching process based on weight adjustments is constructed. • A robust decision-making method based on MULTIMOORA and TODIM is explored. [ABSTRACT FROM AUTHOR]

Selecting the right educational vendors in the Metaverse, particularly those utilizing extended reality (XR), is crucial for creating an engaging and immersive learning environment. Careful vendor selection ensures the delivery of high-quality XR educational content, considering factors such as seamless integration with different XR devices and addressing associated challenges, which can affect the comfort and well-being of learners.This paper proposes a decision-making model to address the complex task of selecting educational vendors within the Metaverse with a focus on extended reality (XR). Recognizing the challenges inherent in such selections, this work presents a consensus-based measurement of alternatives and ranking according to compromise solution (MARCOS) model, incorporating single-valued neutrosophic (SVN) information and SVN Dombi-Archimedean weighted aggregation operators. Attribute weights crucial for decision-making are determined through an optimization tool employing cross-entropy measures with partial weight information. The evaluation encompasses nine criteria, including cost, security, privacy, pedagogical alignment, configuration, maintenance, level of control, output tools, degree of immersion, availability of XR learning contents, and time. Four leading vendors including Victory XR, ARuVR, zSpace, and Varjo Technologies are assessed, with zSpace emerging as the most favorable alternative with the highest overall utility value of 0.6889. The paper further conducts sensitivity analysis and performance comparisons to establish the robustness of the ranking results, providing a comprehensive methodology for educational stakeholders addressing the evolving dynamics of the Metaverse. • A decision-support model is suggested for ranking XR educational vendors under metaverse. • Some robust Dombi-Archimedean weighted aggregation operators are presented under SVN context. • A consensus based SVN-MARCOS methodology is proposed with partial weights information of attributes. [ABSTRACT FROM AUTHOR]

Due to the rapid growth of imaging modalities in clinical analysis and the indispensable requirement of brain images from various imaging modalities for diagnosing a disease, multi-modal brain image fusion has become an intriguing problem among researchers. Thus, the main motive of this paper is to obtain all the necessary information about the source images in a single fused image of high contrast with clear boundaries and without unnecessary noise. Accordingly, this paper proposes a new approach to eradicate the indeterministic and uncertainty present in brain images with the benefits of the neutrosophic set. Also, a novel neutrosophic entropy is developed to acquire the accurate edge details of the image. In addition, to extract requisite features from the images, Tamura features are implemented. Finally, the fusion is performed by comparing the extracted feature values from the images. Subsequently, the experiment is conducted with three different sets of brain datasets and compared with six other fusion algorithms to prove the efficiency of the proposed method. To support this, qualitative and quantitative assessments for each dataset are executed, and the results are tabulated. The results clearly show that the information is accurately represented, preserving the curves and edges. Moreover, this algorithm consistently produces the highest metric values and remains reasonably efficient in time consumption, thus balancing performance and efficiency. • The indeterministic state of the brain images is addressed effectively by using NS. • A novel neutrosophic entropy is framed to fetch the tiny edge details of the brain. • Tamura features are extracted from the truth and neutrosophic entropy image. • The result of proposed fusion method provides comprehensive detail with clear edges. • Qualitative and quantitative analysis evaluates the efficacy of the proposed method. [ABSTRACT FROM AUTHOR]