MAIPFE:An Efficient Multimodal Approach Integrating Pre-Emptive Analysis,Personalized Feature Selection,and Explainable AI  

作　　者：Moshe Dayan Sirapangi S.Gopikrishnan 

机构地区：[1]School of Computer Science and Engineering,VIT-AP University,Amaravathi,Andhra Pradesh,522241,India

出　　处：《Computers, Materials & Continua》2024年第5期2229-2251,共23页计算机、材料和连续体（英文）

摘　　要：Medical Internet of Things(IoT)devices are becoming more and more common in healthcare.This has created a huge need for advanced predictive health modeling strategies that can make good use of the growing amount of multimodal data to find potential health risks early and help individuals in a personalized way.Existing methods,while useful,have limitations in predictive accuracy,delay,personalization,and user interpretability,requiring a more comprehensive and efficient approach to harness modern medical IoT devices.MAIPFE is a multimodal approach integrating pre-emptive analysis,personalized feature selection,and explainable AI for real-time health monitoring and disease detection.By using AI for early disease detection,personalized health recommendations,and transparency,healthcare will be transformed.The Multimodal Approach Integrating Pre-emptive Analysis,Personalized Feature Selection,and Explainable AI(MAIPFE)framework,which combines Firefly Optimizer,Recurrent Neural Network(RNN),Fuzzy C Means(FCM),and Explainable AI,improves disease detection precision over existing methods.Comprehensive metrics show the model’s superiority in real-time health analysis.The proposed framework outperformed existing models by 8.3%in disease detection classification precision,8.5%in accuracy,5.5%in recall,2.9%in specificity,4.5%in AUC(Area Under the Curve),and 4.9%in delay reduction.Disease prediction precision increased by 4.5%,accuracy by 3.9%,recall by 2.5%,specificity by 3.5%,AUC by 1.9%,and delay levels decreased by 9.4%.MAIPFE can revolutionize healthcare with preemptive analysis,personalized health insights,and actionable recommendations.The research shows that this innovative approach improves patient outcomes and healthcare efficiency in the real world.

关 键 词：Predictive health modeling Medical Internet of Things explainable artificial intelligence personalized feature selection preemptive analysis 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]