| Gupta et al. [125]
|
Deep learning using edge computing |
Deep neural network forecasting for health surveillance with cutting-edge computing |
The combination of edge computing and IoT concepts are used in a CNN-based forecasting framework. |
Result is shown on specific data. |
| Kumar et al. [39]
|
Blockchain, Deep learning, IoMT |
Deeper Sparse AutoEncoder (DSAE) combined with Bidirectional Long Short-Term Memory (BiLSTM) is a blockchain-orchestrated deep neural network technique for safe transmission of information in healthcare applications. |
A neural networks technique controlled by cryptocurrency for safe data transfer in an IoT-enabled medical facility |
Tested for specific dataset only. |
| Ahmed et al. [126]
|
IoT, Cloud Computing and Artificial Intelligence |
The combination of derived attributes from neural network topologies is accomplished using the concurrent greatest covariance methods, and feature selection is accomplished using a multi-logistic regression controlled entropy variance approach. |
To improve the diversity of the knowledge set, data enlargement is performed, while neural network algorithms such as VGG-19 and Inception-V3 are used in conjunction with transfer learning approaches to obtain attributes. |
Result is for specific diseases only. |
| Ravi et al. [127]
|
Deep learning, Malware detection |
Utilizing the portable executable (PE)-Header, PE-Image, PE-Imports, or application programming interface (API) calls, artificial intelligence (AI) is utilized to detect infections. |
Hyper-spectral attention-based deep learning framework for malware detection in smart healthcare systems |
In intelligent medical facilities, there is a lack of multi-view attention-based deep learning frameworks and powerful feature fusion approaches for recognizing malware. |
| Jiao et al. [128]
|
Capsule network, Deep learning, Convolutional neural network |
The vital feature extraction ability of neural networks can extract ECG features to solve many problems |
The spatial and temporal components of the ECG are extracted using a 1D convolutional neural network (CNN) and a long short-term memory (LSTM) network as an integrated extraction of the features layer. |
Accuracy not measured |
| Ahmed et al. [129]
|
H-IoT, Deep learning |
The technology use the model known as YOLOv3 to detect motion and irritation. |
For obtaining the spatial and temporal properties of the ECG, a parallel feature extraction layer comprising long short-term memory (LSTM) network and 1D convolutional neural network (CNN) is implemented. |
No measurement of accuracy. |
| Parida et al. [130]
|
DL-based detection/classification of diseases |
Deep learning based diseases detection and patient monitoring |
Deep learning (DL), defeats the limitations of the visual evaluation and the conventional ML |
No measurement of accuracy. |
| Hammad et al. [131]
|
Deep learning, H-Iot, convolutional neural network (CNN) |
Deep Learning Models for Arrhythmia Detection in IoT Healthcare |
After representing the input ECG signals in 2D format, the generated pictures are fed into the suggested DLMs for classification. |
Efficiency calculated based on noisy data only. |
| Sujith et al. [132]
|
Deep Learning, Artificial intelligence |
Smart health monitoring using deep learning and Artificial intelligence |
By utilizing various IoT, GSM, and SHM modules, DL is beneficial for gathering in-depth information on many important patients, notably those who are coma sufferers. |
No measurement of accuracy. |
| Refaee et al. [133]
|
Deep learning, Artificial intelligence, IoT |
Data are collected from various IoT wearable devices; these data are prepossessed and applied iForest for outlier detection with linear time complexity and high precision |
It integrates deep learning, and the internet of things for effective disease diagnosis |
Result is outstanding for specific diseases only. |
| Moqurrab et al. [134]
|
IoT, cloud computing, Fog computing and Artificial Intelligence |
Propose a new model using IoT, cloud computing, Fog computing and Artificial Intelligence |
Deep learning is used in a fog-enabled privacy-preserving model to enhance the healthcare system. |
Result is for specific diseases only. |
| Awotunde et al. [135]
|
Deep learning, IoMT, ArtificialIntelligence |
A real-time NF-ToN-IoT dataset for IoT applications that gathered telemetry, operating system, and network data was used to assess the system's performance. |
Because IoMT-based devices have a limited capacity for storage and computation, patient health data must be sent to cloud database storage and external computer equipment for processing. |
Model attains 89% accuracy over the ToN-IoT dataset only. |
| Sahu et al. [136]
|
Deep learning, IoMT, Artificial Intelligence |
Constant Verification based on deep learning algorithms for an IoT-enabled medical facility |
The system collects data from clients and verifies them using a Deep Learning-based short-term and long-term memory algorithm for categorization that has only been tested on a single dataset. |
Tested for specific dataset only |
| Munnangi et al. [137]
|
Deep learning, IoMT |
Enhanced Deep Learning-Based Approach for Medical Data Analysis in IoT Systems |
As a result, the inference of IoT knowledge necessitates the use of effective, lightweight methodologies that are appropriate for this compromise and to validate with limited resources in IoT devices such as wearables. |
Result is shown on specific data. |
