Method of Assessing the State of Monuments based on Fuzzy Logic

Telenyk, Sergii; Czajkowski, Krzysztof; Бідюк, Петро Іванович; Zharikov, Eduard

doi:10.1109/idaacs.2019.8924315

Cited by 4 publications

(3 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In order to address the methods, techniques and algorithms used for making decisions in biomedicine, let us take into account the following aspects of medical data processing: missing data imputation, diagnostics (classification and prediction), clustering and personalizing the treatment. A previous study predicted missing data, analyzed the nature of data gaps, and filled these gaps using decision tree-based computation techniques and regression approach [10]. Similar outcomes for associative rules mining in medical data were found by another study [11].…”

Section: Related Worksupporting

confidence: 57%

Liver Disease Classification - An XAI Approach to Biomedical AI

Agbozo,

Balungu

2024

IJCAI

View full text Add to dashboard Cite

Explosive amounts of biological and physiological data, including medical images, electroencephalograms, genomic information, and protein sequences, have been made available to us thanks to advances in biological and medical technologies. Understanding human health and disease is made easier by using this data for learning. Deep learning-based algorithms, which were developed from artificial neural networks, have significant potential for identifying patterns and extracting features from large amounts of complex data. However, these recent advancements involve blackbox models: algorithms that do not provide human-understandable explanations in support of their decisions. This limitation hampers the fairness, accountability and transparency of these models; the field of XAI tries to solve this problem providing human-understandable explanations for black-box models. This paper focuses on the requirement for XAI to be able to explain in detail the decisions made by an AI in a biomedical setting to the expert in the domain, e.g., the physician in the case of AI-based clinical decisions related to diagnosis, treatment, or prognosis of a disease. In this paper, we made use of the Indian Patient Liver Dataset (IPLD) collected from Andhra Pradesh region. The deep learning model with a 0.81 accuracy score (0.82 for the hyperparameter-tuned model) is built on Keras-Tensorflow and due to the imbalance in the target values, we integrated GANs as a means of oversampling the dataset. This study integrated the XAI concept of Shapley Values to shed light on the predictive results obtained by the liver disease detection model. Povzetek: Študija obravnava klasifikacijo jetrnih bolezni z uporabo razložljive umetne inteligence (XAI), ki omogoča razumevanje odločitev modelov globokega učenja z integracijo Shapley vrednosti za razlago prediktivnih rezultatov.

show abstract

Section: Related Worksupporting

confidence: 57%

Liver Disease Classification - An XAI Approach to Biomedical AI

Agbozo,

Balungu

2024

IJCAI

View full text Add to dashboard Cite

show abstract

“…• Clustering and personalizing the treatment. P. Bidiuk [11] uses decision tree-based computation procedures and regression approach to predict missing data, analyze the nature of data gaps and fill out these gaps. Dangare [12] obtained similar results for associative rules mining in medical data.…”

Section: Related Sourcesmentioning

confidence: 99%

Data Driven Analytics for Personalized Medical Decision Making

Melnykova¹,

Shakhovska²,

ml.³

et al. 2020

Preprint

View full text Add to dashboard Cite

The study was conducted on applying machine learning and data mining methods to personalizing the treatment. This allows investigating individual patient characteristics. Personalization is built on the clustering method and associative rules. It was suggested to determine the average distance between instances for optimal performance metrics finding. The formalization of the medical data pre-processing stage for finding personalized solutions based on current standards and pharmaceutical protocols is proposed. The model of patient data is built. The paper presents the novel approach to clustering built on ensemble of cluster algorithm with better than k-means algorithm Hopkins metrics. The personalized treatment usually is based on decision tree. Such approach requires a lot of computation time and cannot be paralyzed. Therefore, it is proposed to classify persons by conditions, to determine deviations of parameters from the normative parameters of the group, as well as the average parameters. This made it possible to create a personalized approach to treatment for each patient based on long-term monitoring. According to the results of the analysis, it becomes possible to predict the optimal conditions for a particular patient and to find the medicaments treatment according to personal characteristics.

show abstract

“…Bidiuk [11] used decision-tree-based computation procedures and regression approaches to predict missing data and for PM decisions. Dangare [12] obtained similar results for mining of associative rules in medical data.…”

mentioning

confidence: 99%

Data-Driven Analytics for Personalized Medical Decision Making

et al. 2020

View full text Add to dashboard Cite

The study was conducted by applying machine learning and data mining methods to treatment personalization. This allows individual patient characteristics to be investigated. The personalization method was built on the clustering method and associative rules. It was suggested to determine the average distance between instances in order to find the optimal performance metrics. The formalization of the medical data preprocessing stage was proposed in order to find personalized solutions based on current standards and pharmaceutical protocols. The patient data model was built using time-dependent and time-independent parameters. Personalized treatment is usually based on the decision tree method. This approach requires significant computation time and cannot be parallelized. Therefore, it was proposed to group people by conditions and to determine deviations of parameters from the normative parameters of the group, as well as the average parameters. The novelty of the paper is the new clustering method, which was built from an ensemble of cluster algorithms, and the usage of the new distance measure with Hopkins metrics, which were 0.13 less than for the k-means method. The Dunn index was 0.03 higher than for the BIRCH (balanced iterative reducing and clustering using hierarchies) algorithm. The next stage was the mining of associative rules provided separately for each cluster. This allows a personalized approach to treatment to be created for each patient based on long-term monitoring. The correctness level of the proposed medical decisions is 86%, which was approved by experts.

show abstract

Method of Assessing the State of Monuments based on Fuzzy Logic

Cited by 4 publications

References 7 publications

Liver Disease Classification - An XAI Approach to Biomedical AI

Liver Disease Classification - An XAI Approach to Biomedical AI

Data Driven Analytics for Personalized Medical Decision Making

Data-Driven Analytics for Personalized Medical Decision Making

Contact Info

Product

Resources

About