Table 1.

A summary of the literature referred in this study.

Author	Work Description	Data Traces
Shah et al. [42]	Addressed the problems with service quality for medical applications.	Historical patient data with real-time patient data.
Nandyala et al. [43]	Enhanced communication between hospitals and smart homes.	Designed u-health care monitoring system using cloud to fog (C2F) and IoT computing.
Costanzo et al. [44]	The main goal was to use mobile devices to monitor patients who are stationed far away. For quick patient rescue in an emergency, the suggested monitoring technique was used for interfacing by means of the first-aid software.	Wearable technology and embedded technologies-based system was devised. The overall goal of the suggested monitoring system is to suggest the appropriate course of action in cases of serious medical disorders.
Oluwagbemi et al. [45]	The suggested approach was made to diagnose and suggest treatment for the Ebola virus disease. In a survey conducted, 61% of respondents agreed that the suggested approach might suggest a course of treatment for the Ebola virus disease.	Constructed Ebola fuzzy informatics system using fuzzy logic and expert systems.
Sood et al. [46]	The proposed system was devised to track and distinguish between the numerous diseases spread by mosquitoes. The suggested system’s goal is to regulate diseases at their earliest stages. The suggested structure calculates similarity factors to distinguish between diseases.	IoT sensors, fog, and cloud computing make up the main components of the proposed health care system. The infected users are classified using the J48 decision tree classifier.
Thota et al. [47]	A security-based architecture for geographically distant health care systems was created. The tracking, identification, and security of authorization and authentication for all devices are the primary goals of the proposed design.	The suggested architecture enables asynchronous communication between cloud-based health applications and data servers.
Venckauskas et al. [48]	Datagram Transport Layer Security (DTLS) and User Datagram Protocol are replaced with the suggested protocol as a secure transport for Constrained Application Protocol (CoAP) (UDP). The experimental findings demonstrated that the suggested protocol performs better than DTLS and UDP in lossy networks and with CoAP block transfer mode.	For a fog-based eHealth architecture, a protected self-authenticable transfer protocol was proposed.
Saxena et al. [49]	A health care system was designed to manage mosquito infections at an initial stage.	The suggested solution uses wearables and IoT devices, fog computing, fuzzy k-nearest neighbor technique, and social network analysis concepts.
Ginier et al. [50]	Zika fever might be mistaken for dengue fever, though Zika infection seldom causes fever. It It has been observed that the only symptoms of a Zika infection are skin rashes and slight edema in the patient.	A discussion on the potential treatments for ZIKV infection was carried out.
Pabbaraju et al. [52]	To identify and distinguish between these viruses for the purpose of proper therapy, the RT-PCR assay was used for testing the blood of a patient. According to the findings, the RT-PCR assay is completely precise and did not exaggerate any of the several viruses examined.	Reverse Transcription—Polymerase Chain Reaction (RT-PCR) assay was used to identify the Zika, dengue, and chikungunya viruses.
Campion et al. [53]	Using data on trap counts from 2005 to 2015 and historical weather data, the authors suggested a prediction technique using the partial least squares regression technique to forecast the mosquito trap counts.	A web interface based on Google Maps was created to display information regarding the frequency of the West Nile virus, the density of mosquitoes, and the weather.
Lambert et al. [54]	An age tracking tool was developed to accurately predict the age of the mosquito. The author also argued that a crucial parameter for killing adult mosquitoes is the mosquito’s age. This objective age assessment generates a precise mosquito population.	The suggested method made use of boosted regression trees, random forests, main components regression, and neural networks with near-infrared spectroscopy, among other machine learning approaches.
Kirk et al. [55]	The system’s primary objectives are to identify environmental changes, make risk as-assessments, and provide real-time advice for mitigating mosquito illness outbreaks.	The DEAR (Detect, Evaluate, Assess and Recommend action) decision-making system was created.
Devarajan et al. [56]	A health care system was put into place dealing with the Parkinson’s disease. The suggested system examined patient voice samples to suggest best course of action.	In the suggested architecture, fog computing served as a midway layer in the end user and the cloud server. Further, the classification of Parkinson and non-Parkinson subjects was performed using the fuzzy k-nearest neighbor (FKN) classifier, case-based reasoning (CBR) classifier.
Kaur et al. [57]	The recommendations for diagnostics are provided based on the past data stored in the cloud. The judgments of how to hide the numerous patterns in the database were also aided by the suggested method.	A health monitoring system utilizing cloud concept, multiple machine learning methods, and IoT structure was described.
Parthasarathy et al. [58]	The proposed LMM system for joint inflammatory disease made use of wearable sensor devices and uric acid sensors as a component of IoT infrastructure. The suggested technique is also utilized to transform health information and identify foot motion in order to diagnose GOUT arthritis.	A leg movement monitoring (LMM) system was designed to identify the onset of disease or joint pain.
Tuli et al. [59]	The suggested model offered fog services via IoT devices and maintained medical data in accordance with user requests. Using FogBus, the implementation time, latency, power consumed, accuracy, bandwidth of network, and jitter of HealthFog are evaluated. The findings demonstrated that HealthFog offers the highest level of service quality and forecast accuracy.	A novel model called HealthFog was created for the automatic analysis of cardiac disorders. The HealthFog integrated edge computing (EC) hardware with deep learning (DL).
Priyadarshini et al. [60]	A DeepFog health care model to forecast overall wellness was developed. It used fog computing to gather patient data and deep neural networks to forecast three aspects of well-being, including stress level, hypertension attacks, and diabetes.	Fog computing and deep learning was used for constructing the model.
Jabeen et al. [61]	Recommender system was established to diagnose heart illness. The primary purpose of the suggested system is providing consumers nutrition and exercise advice.	Biosensors, IoT, prediction classifiers RF, NB, MLP, and SVM used for designing the system.
Sood et al. [62]	A cyber–physical localization (CPL) system was proposed with the fundamental goal of assessing the jeopardy of coronary heart disease, for tracking patients’ ECG readings, to inform users and specialists when readings are aberrant, and to suggest medications and preventative measures in accordance with risk category.	The proposed system is based on the concepts of cloud computation and neuro-fuzzy implication.
Gu et al. [63]	A diagnostic knowledge model (DKM) established for classifying the clinical conditions. The suggested system’s main goals are to discharge the health staff of the hefty weight of hospital duties and to offer appropriate decision-support.	The proposed system incorporated medical devices and made use of knowledge systems using the Component-Based Medical Cyber–Physical System framework (CBMCPS).
Sood et al. [64]	A health care system was introduced to identify early-stage hypertension individuals based on user health data. The suggested method continuously evaluates and keeps track of the patients’ blood pleasure.	The system uses IoT sensors, artificial neural networks, mobile devices, and cloud storage.
Lakshmanaprabu et al. [65]	A health care system was developed to categorize the various diseases according to chosen criteria. Using a precision parameter, the suggested system was assessed using several real-time hospital datasets.	The system employed an IoT structure, MapReduce, the enhanced dragonfly algorithm, and RF classifier.
Anand et al. [66]	A hybrid framework was suggested to categorize the hepatic syndrome. The suggested system’s performance was assessed, and the findings proved that it outperforms as compared to existing systems classification accuracy.	The techniques used are updated particle swarm optimization, updated artificial neural network, the SPARK tool.
Sood et al. [67]	A diagnostic system suggested that incorporates social network analysis (SAS) in cloud subsystem to offer a GPS-based worldwide risk assessment of dengue infection on Google Maps for preventing the spread of the infection. The effectiveness of the suggested system’s diagnosis, warning production, and risk assessment based on GPS capability was acknowledged using various statistical measurements and experimental methodologies.	A system with NB network and fog computing suggested and used Google Maps, GPS, SAS.
Sood et al. [68]	An IoT-based fog-cloud diagnosed system for controlling and detecting dengue infection in 2021. To analyze the influence of the proposed system, the investigational findings were assessed using a numeral of analytical constraints.	The proposed system uses SVM, Google Maps, and temporal network analysis (TNA).
Suggala et al. [69]	A novel dengue prediction method using fog computing introduced. The dengue infected was detected by checking the similarity factors between the disease and the users. Finally, at the cloud layer, an innovative Temporal Social Network Analysis (TSNA) was designed to evaluate the risk of disease outbreak, analyze sick users, and direct an awareness text to initiate preventive steps.	The proposed method uses cloud concept and temporal social network analysis (TSNA).