Advanced searches left 3/3

Medical Diagnosis - DOAJ

Summarized by Plex Scholar
Last Updated: 19 June 2022

* If you want to update the article please login/register

A Two-Step Data Normalization Approach for Improving Classification Accuracy in the Medical Diagnosis Domain

"Data normalization is a data preprocessing step and one of the first tasks to be performed during scientific investigation, especially in the case of tabular data. " The sensitivity of the artificial intelligence model to the values of the data's values is determined by the need to reduce the sensitivity of the predictive model to the values of the features in the dataset, which would raise the investigated model's adequacy. In the case of performing the multiclass classification task, it was experimentally shown that the computerized classification method improves the accuracy of the Decision Tree and Extra Trees Classifier by 11%. Increasing the accuracy of these classifications can only be achieved by using the latest data normalization technique, which meets all of the requirements for its use in practice when doing various medical data mining tasks. ".

Source link: https://doi.org/10.3390/math10111942


A Novel Sequential Three-Way Decision Model for Medical Diagnosis

"Forte, reliability, and the decision threshold pair are two key variables that influence the classification results, and in the sequential three-way decision model. " The conditional probability is estimated based on a tight equival relationship, which limits its use in reality. Firstly, we present two methods of calculating conditional probability based on the property of symmetry. We construct an S3WD model for a medical records management system and use three distinct classes of cost functions as the basis for changing the threshold pair at each level.

Source link: https://doi.org/10.3390/sym14051004


Micro/Nanorobots for Medical Diagnosis and Disease Treatment

"Compared to conventional robots, micro/nanorobots can perform a variety of tasks on the micro/nanoscale, which has the benefit of high precision, high durability, strong flexibility, and wide adaptability. " In addition, such robots can also do tasks in a cluster fashion. This paper seeks to introduce current driving methods for micro/nanorobots preparation in detail, summarizes the growth of science in medical applications, discusses the challenges it faces in clinical research, and the future direction of development.

Source link: https://doi.org/10.3390/mi13050648


Research on Application of Data Mining Algorithm in Cardiac Medical Diagnosis System

"Due to the wide variety of heart disease pathology, how to enhance heart disease diagnosis and prevention, as well as early intervention and treatment, is a problem that must be addressed urgently. " It is aimed at finding which of these four algorithms is more appropriate for heart disease diagnosis problems and optimizing them based on the development of a heart disease diagnosis classifier based on data mining algorithms. The introduction of ultrasound diagnostic tools for heart disease at the same time is discussed, and the critical role of ultrasound diagnosis in heart disease diagnosis is discussed. This thesis employs the heart disease clinical records of the patients to determine a heart disease diagnosis classifier based on the decision tree algorithm, neural network algorithm, support vector machine algorithm, and Co-SVM algorithm.

Source link: https://doi.org/10.1155/2022/7262010

* Please keep in mind that all text is summarized by machine, we do not bear any responsibility, and you should always check original source before taking any actions

* Please keep in mind that all text is summarized by machine, we do not bear any responsibility, and you should always check original source before taking any actions