Healthcare Revolution
Neeraj Agnihotri, Karthik Kulkarni, Sahil Sharma, Md Miran Asif , Divya HN
This paper presents an efficient system for heart disease diagnosis using machine learning. It uses various classification algorithms and feature selection methods, including the new Fast Conditional Mutual Information (FCMIM) feature selection algorithm. The goal of the system is to increase accuracy and reduce execution time. The results indicate that FCMIM coupled with Support Vector Machine outperforms existing methods and offers a feasible and accurate solution for healthcare implementation in heart disease identification. This paper addresses the problem of unbalanced datasets in the classification of dermatological diseases, specifically for skin cancer using machine learning (ML) methods. The proposed approach combines expansion with a category swing to compensate for imbalances, thereby increasing the penalty for misclassified cases. The method focuses on three key contributions: tailored feature extraction using different backbone models, optimization of loss features and training parameters, and handling unbalanced samples by optimizing weights between asymmetric classes. Evaluation on the ISIC2018 benchmark and chest X-ray dataset with popular backbone networks such as EfficientNets, MobileNets, and DenseNets demonstrates the effectiveness and stability of the proposed approach compared to existing methods, achieving higher accuracy and stable performance without the need for dataset expansion. Experimental results on the ISIC2018 dataset reveal significant improvements over other methods in specific evaluation criteria. For example, with the EfficientNet backbone, it outperforms the focal loss method by 2.73% in recall, 2.63% in precision, 2.81% in specificity, and 3.09% in F1, demonstrating its effectiveness in dealing with unbalanced datasets. This study addresses the increasing incidence of chronic kidney disease (CKD) worldwide by using artificial intelligence (AI) and machine learning (ML) for early diagnosis. The research recognizes the importance of explainability for clinician acceptance and introduces a CKD predictive model with explainable AI (XAI). Optimized for accuracy and explainability, the model uses extreme gradient amplification and identifies three key properties (hemoglobin, s
Article Details
Unique Paper ID: 162191

Publication Volume & Issue: Volume 10, Issue 8

Page(s): 220 - 225
Article Preview & Download

Share This Article

Join our RMS

Conference Alert

NCSEM 2024

National Conference on Sustainable Engineering and Management - 2024

Last Date: 15th March 2024

Call For Paper

Volume 11 Issue 1

Last Date for paper submitting for Latest Issue is 25 June 2024

About Us enables door in research by providing high quality research articles in open access market.

Send us any query related to your research on

Social Media

Google Verified Reviews