Most of the medical datasets suffer from missing data, due to the expense of some tests or human faults while recording these tests. This issue affects the performance of the machine learning models because the values of some features will be missing. Therefore, there is a need for a specific type of methods for imputing these missing data. In this research, the salp swarm algorithm (SSA) is used for generating and imputing the missing values in the pain in my ass (also known Pima) Indian diabetes disease (PIDD) dataset, the proposed algorithm is called (ISSA). The obtained results showed that the classification performance of three different classifiers which are support vector machine (SVM), K-nearest neighbour (KNN), and Naïve Bayesian classifier (NBC) have been enhanced as compared to the dataset before applying the proposed method. Moreover, the results indicated that issa was performed better than the statistical imputation techniques such as deleting the samples with missing values, replacing the missing values with zeros, mean, or random values.

The concept and growth of superior individualized healthcare technologies are influenced in significant ways by the rising areas of "Artificial Intelligence (AI) and the Internet of Things (IoT)". Most people use wearable devices for mHealth, hence there are many potential applications for the "Internet of Medical Things (IoMT)". Only 5G can provide the necessary support for smart medical devices to perform many different types of demanding computing activities. Today, heart disease was the major mortality on a global scale. For patients who need a greater accurate diagnosis and treatment, the advancement of medical innovation has created new obstacles. Although many studies have focused on diagnosing cardiac disease, the findings are often inaccurate and fail to fulfill patients' expectations of quality of service (QoS). So, this paper introduces a novel "feed-forward Bi-directional long-short term memory (FF-Bi-LSTM) algorithm to predict heart disease more accurately with enhanced QoS in IoMT based on 5G". Linear discriminant analysis (LDA) and min-max normalization are employed, respectively, for preprocessing and feature extraction. Several measures, including precision, recall, accuracy, and f1-score, are used to the assess effectiveness of the suggested strategy. The proposed method also compared to certain existing techniques. These results show that the suggested strategy outperforms existing strategies in terms of improving QoS.