Description:Field of Invention
With the advent of technological devices, which plays a big role in the medical field, that can recognize such tiny viruses undetectable to the naked eye. And now that these kinds of innovations are available, we can advance the process and triumph over COVID-19. The innovation is based on IoT (Internet of Things) and AI (Artificial Intelligence) since it uses a biological sensor to identify the virus using a trained AI model (based on the shape, size, and texture of the virus) that is associated with a warning mechanism that sends out alarm messages to alert the user.
The Objectives of this Invention
The primary goal of this invention is to locate the infection using a variety of AI tools, with a few additional techniques for deep learning (DL) taken into account to meet the necessary objectives. These DL techniques include recurrent neural networks (RNNs) and long short-term memory (LSTM). It also features an integrated method for developing bioinformatics approaches, in which different data from various data sources are brought together for user-friendly platforms. The prevalence of platforms with AI capabilities will aid in accelerating the detection and treatment of COVID-19. Additionally, it offers platform-specific inputs, such as various readily available data, such as healthcare data and photographs, that can support the objective of the current approaches to get the best results in actual serious situations.
Background of the Invention
According to (AU2020/101728A4), Smart COVID Scanner: A mobile, cost-effective scanner for finding the COVID-19 virus Abstract The most recent coronavirus strain of, known as COVID-19 is currently causing an outbreak affecting humanity. Therefore, investigators and researchers from various fields constantly collaborate to combat this brand-new type of virus. Therefore, there is a pressing requirement for preventative tools like sanitizers and masks to stop people from being sick, as prevention is currently the only viable treatment. In addition to this, technology can aid in the fight against COVID-19 because in these kinds of situations, advanced technologies are now a crucial element. The suggested remedy is a portable scanner that can identify the COVID-19 virus on any surface or person and alert the user before it spreads. Similar to how an electron microscope emits rays to detect microorganisms, the scanner uses a microbial scanner to capture and detect viruses. This bio scanner being used. Another type of application invented in (US2020/0364157A1), An approach for identifying the COVID-19 virus in a living being a sample or an environmental specimen containing one or more viruses and bacteria that cause illness, can be found here. Total nucleic acids are obtained after processing samples. To produce fluorescently labeled COVID-19 virus-specific amplicons, a combination of asymmetric DNA amplification (PCR) procedures is carried out. Microarray hybridization detects the amplicons close to the lowest limit of detection. The approach described above can also detect other respiratory disease-causing organisms, such as viruses, bacteria, and fungi, simultaneously with the COVID-19 virus. Another method was invented in (US2020/10902955B1), A system for diagnosing illnesses that uses sensor data from a user device (such as a mobile or activity tracker) to determine whether a user is likely to have contracted a disease. Each symptom is detected by comparing sensor data to a baseline and the change to a specified symptom threshold. The triage system additionally employs substitutes to identify some symptoms since direct measurement of sensations using the sensors that are accessible to the user may not be possible or sufficiently precise. For instance, a fever can be detected using heart data, a cough or shortness of breath can be detected by listening to documented sounds, and the device's circulation can detect fatigue. Loss of senses such as taste and smell can be detected by preserving sound and then employing speech recognition algorithms to find phrases that indicate the condition.
According to Chinnasamy et al.'s research ((2022), Materials Today: Proceedings, Vol.64, Part-I, pp-448-451), For the study, we provide sentiment analysis utilizing data from Twitter. The program we use first retrieves tweets and hashtags about various types of covid vaccines posted on Twitter using the Twitter API. The downloaded Tweets are then dynamically set up to produce a set of unknown variables and inexperienced rules. We're using Tweepy, a wrapper for the Twitter API, to build our model. The software then generates donut graphs as part of the sentiment evaluation of new Posts. According to Chinnasamy et al. ((2022), 022 International Conference on Computer Communication and Informatics (ICCCI), Coimbatore, India, 2022, pp. 1-5), In this study, machine learning techniques, notably the Convolutional Neural Networks (CNN) VGG16 approach, have been built. The trained model is utilised to predict using the trained dataset. Four deep CNN architectures are evaluated for COVID-19 treatment on chest X-ray images. Data sets of covid 19 X-ray imageries and non-covid 19 X-ray imageries are collected to train the model and assess the model's precision. It was found that CNN-based designs can diagnose COVID-19 disease.
Summary of the Invention
While AI is accelerating tactics to defeat COVID-19, real-world trials must be planned because it is still being determined precisely what advantages and disadvantages AI-based strategies have. Become proficient in handling problems of this difficulty. To attack COVID-19 and ensure its complete eradication, it is essential to develop a toolkit of platforms, strategies, tactics, and innovations that work together to produce desired results and save more lives.
Detailed Description of the Invention
In order to cope with COVID in medical facilities around the world, the proposed system aims to incorporate specific, realistic AI-based methodologies that complement the current conventional manufacturing methodology. To illustrate the efficacy of these strategies and technologies was based on the most recently published medical updates related to artificial intelligence or even the most recent revision on COVID-19. As a result, these varied difficulties are connected to how ANN-based procedures may be improved to accomplish a process that increases healthcare identification and diagnosis, as well as preventive and restorative ways. However, how much humans participate and work together in various roles will determine how much AI will be used to avoid the COVID-19 epidemic.

The layer that receives input is the initial layer that interacts with and communicates with the database. High-speed connections connect this layer to the input (front-end) computer devices. The user might supply Medical Diagnosis. Taking Images into Account A CT scan image and the model would then utilize those images to determine whether or not the patient was likely to contract the virus. LSTM networks are better suited for learning long-range pattern repetitions of uncertain length. Since an ML solution, unlike humans, can manage large datasets far greater than the range of individuals can handle or even monitor regularly, human contributions are of the utmost importance at this point. When processing large or complex amounts of data makes machine learning (ML) or more conventional data processing approaches ineffective, deep learning (DL) strategies might be used. DL techniques don't necessitate human involvement, as seen in Figure 1. A branch of machine learning known as deep learning (DL) uses numerous layers of techniques to provide a unique perspective on the data it is fed. The manner that DL represents the data in the system, in contrast, is how it differs from ML. While DL networks work using stages of artificial neural networks (ANNs), ML algorithms frequently rely on structured data. Unsupervised learning, like supervised learning, is typified by minimal human supervision. They can be considered a particular kind of algorithm for machine learning that searches for previously unknown things or patterns in the data set where no labels have been assigned. There are several methods to employ AI with COVID-19. Nevertheless, our objective is to identify the most effective solutions to the COVID-19 problems, which have significantly hampered health systems. Because of this, this solution's High-Risk Group, Outburst and Control, and Identification and Diagnostic sections are split into three parts.
AI-based Covid-19 detection using Artificial Neural Networks (ANN) involves using a type of machine learning algorithm to analyze medical data, such as X-rays or CT scans of patients' lungs, in order to identify potential cases of Covid-19. ANN is a subset of machine learning inspired by the structure and functioning of the human brain's neural networks. Here's how the process generally works: The first step is to gather a large dataset of medical images, such as X-rays or CT scans, from both Covid-19 positive and negative cases. These images are labeled accordingly to indicate whether the patient has Covid-19 or not. Medical images often need preprocessing before being fed into the neural network. This might involve resizing, normalizing pixel values, and other techniques to ensure uniformity and enhance the model's ability to learn. An Artificial Neural Network is designed to mimic the structure of the human brain, with interconnected layers of artificial neurons. A common architecture used for image classification tasks is a Convolutional Neural Network (CNN). CNNs are particularly well-suited for image analysis due to their ability to capture spatial features. The network is trained using the labeled dataset. During training, the ANN learns to recognize patterns and features in the images that are indicative of Covid-19 infection. It adjusts its internal parameters (weights and biases) to minimize the difference between its predictions and the actual labels in the training data. The model's performance is monitored using a validation dataset that it hasn't seen during training. This helps to prevent overfitting (memorization of the training data) and allows for fine-tuning of hyperparameters, such as learning rate and number of layers, to improve the model's generalization ability. Once the model is trained and validated, it's tested on a separate set of images to evaluate its performance on unseen data. This provides an estimate of how well the model can generalize to new, real-world cases. After training, validation, and testing, the trained ANN can be used to analyze new medical images. It takes in an image as input, processes it through its layers, and produces an output prediction indicating whether the patient's image suggests a Covid-19 infection or not.
The working model of an AI-based COVID-19 detection system using Artificial Neural Networks (ANN) involves several steps, from data preparation and model training to inference and diagnosis. Here's a detailed explanation of the working process: Gather a dataset of medical images, such as chest X-rays or CT scans, that include COVID-19-positive and negative cases. Ensure the dataset is labeled accurately. Preprocess the images by resizing them to a consistent resolution and normalizing pixel values to a specific range. Utilize pre-trained CNN models (e.g., VGG, ResNet, or Inception) as feature extractors. Remove the classification layers of the pre-trained model and retain the convolutional layers that can extract high-level features from the images. Design an Artificial Neural Network (ANN) architecture that takes the extracted features as input. The output layer usually consists of two nodes, representing COVID-19-positive and negative classes. Initialize the ANN's weights randomly. Feed forward a batch of pre-processed images through the CNN layers to extract features. The extracted features are then fed into the fully connected layers for classification. Calculate the loss (difference between predicted and actual labels) using a loss function (e.g., cross-entropy). Use backpropagation to update the weights of the network using optimization algorithms like Adam or SGD to minimize the loss. Use a validation dataset to monitor the model's performance during training. Adjust hyperparameters (learning rate, batch size, etc.) based on the validation results to prevent overfitting. Once training is complete, use a separate testing dataset that the model has not seen before. Feed the test images through the network and obtain predictions. Evaluate the model's performance using metrics like accuracy, precision, recall, F1-score, and ROC curves. Deploy the trained model in a clinical setting. Medical professionals input X-ray or CT scan images into the model. The model provides predictions, indicating whether the image suggests COVID-19 infection. Collaboration between AI experts and medical professionals is crucial to validate the model's results and ensure its integration into the clinical workflow.
Overall, the AI-based COVID-19 detection system using ANN combines powerful feature extraction capabilities of CNNs with the learning capabilities of ANNs to aid medical professionals in accurate and timely diagnosis. It's important to follow rigorous validation processes and maintain ethical standards in healthcare when developing and deploying such systems.
It's important to note that AI-based Covid-19 detection using ANN is not a definitive diagnostic tool. While it can provide a valuable aid in the diagnostic process, it should not replace clinical evaluation and expert medical opinion. Additionally, the success of this approach depends on the availability of high-quality and diverse medical image datasets, as well as rigorous training and validation processes.
5 Claims & 1 Figure
Brief description of Drawing
In the figure which are illustrate exemplary embodiments of the invention.
Figure 1, The Process of proposed Covid-19 using AI-Based Techniques , Claims:The scope of the invention is defined by the following claims:

Claim:
1. A system/method for making autonomous system for agriculture using artificial intelligence and WSN, said system/method comprising the steps of:
a) The system starts with the data collections (1), the human values are measured (2) and gave to the machine learning models (3).
b) The system analyzes and matched with different measurements to predict the Covid-19 (4).
c) The output is matched with previous medical results (5) and medical experts (6) will give an alert to the human.
2. As mentioned in claim 1, The first step is to gather a large dataset of medical images, such as X-rays or CT scans, from both Covid-19 positive and negative cases. These images are labeled accordingly to indicate whether the patient has Covid-19 or not. Medical images often need preprocessing before being fed into the neural network.
3. According to claim 1, This might involve resizing, normalizing pixel values, and other techniques to ensure uniformity and enhance the model's ability to learn.
4. As per claim 1, An Artificial Neural Network is designed to mimic the structure of the human brain, with interconnected layers of artificial neurons. A common architecture used for image classification tasks is a Convolutional Neural Network (CNN).
5. According to claim 1, CNNs are particularly well-suited for image analysis due to their ability to capture spatial features. The network is trained using the labeled dataset. During training, the ANN learns to recognize patterns and features in the images that are indicative of Covid-19 infection.