AI and IoT-based Intelligent Health Care & Sanitation

A form of cancer known as leukemia, attacks the body's blood cells and bone marrow. This happens when cancer cells multiply rapidly in the bone marrow. The uploaded image is analyzed by the website, and if leukemia is present, the user is notified-a collection of pictures depicting leukemia as well as healthy bones and blood. Once collected from Kaggle, the data is preprocessed using methods like image scaling and enhancement. To create a Deep Learning (DL) model, we use the VGG-16 model. The processed data is used to “train” the model until optimal results are achieved. A Hypertext Markup Language (HTML) based website is built to showcase the model. Using a DL model, this website returns a response indicating whether or not the user's uploaded photograph shows signs of leukemia. The primary aim of this site is to lessen the likelihood that cancer cells may multiply while the patient waits for test results or is otherwise unaware of their condition. Waiting for results after a leukemia test can cause further stress and even other health problems, even if the person is found to be leukemia-free. This problem can be fixed if this website is used as a screening tool for leukemia. 

The Heart Rate Monitoring system is developed using Artificial Intelligence (ANN) and Internet of Things technology to detect the heartbeat and SpO2 of the patient to monitor the risk of heart attack and also make the person do regular checkups. Health monitoring is very important to us to make sure our health is in which condition. The proposed framework examines the sources of information gathered from sensors fit with the patients to discover the crisis circumstance. The IoT technology is employed; it helps anyone monitor the patient's health from anywhere. A deep learning algorithm such as ANN is utilized to identify whether the person’s health is normal or abnormal. In case an abnormality in health is noticed, an alert will be popped out. The proposed framework with artificial intelligence and IoT can reduce the death occurring due to heart rate, and other related issues can also be avoided.

Diabetes is a chronic ailment characterized by abnormal blood glucose levels. Diabetes is caused by insufficient insulin synthesis or by cells' insensitivity to insulin activity. Glucose is essential to health since it is the primary source of energy for the cells that make up a person's muscles and tissues. On the condition that if a person has diabetes, his or her body either does not create enough insulin or cannot utilize the insulin that is produced. When there isn't enough insulin or cells stop responding to insulin, many dextroses accumulate in the person's vascular framework. As time passes, this could lead to diseases such as kidney disease, vision loss, and coronary disease. Although there is no cure for diabetes, losing weight, eating nutritious foods, being active, and closely monitoring the diabetes level can all assist. In this research, we used Artificial Neural Network to create a Deep Learning (DL) model for predicting Diabetes. Then it was validated using an accuracy of 92%. In addition, with the help of the MIT website, a mobile application was constructed. This project will now assist in predicting the effects of diabetes and deliver personalized warnings. Early detection of pre-diabetes can be extremely beneficial to patients since studies have shown that symptoms of early diabetic difficulties frequently exist at the time of diagnosis.

Health monitoring may be required regularly in everyday life, which might help predict the significant health consequences. Accurate surveillance is required for effective health parameters like temperature, stress, heart rate and blood pressure (BP) in the medical and healthcare domains. The Ideal health-related characteristics for efficient persistent health monitoring are established in this study. The primary goal of the device is to monitor the health parameters of a person in everyday life, facilitating psycho-physiological supervision to examine the relationship between underlying emotional states, including changing stress levels, and the progression and prognosis of cardiovascular disease. Non-invasive sensors are employed here to observe the mentioned health-related variables. The observed data will be stored in the cloud for further processing. IoT technology has been used to process and store the measured parameters in the cloud. At the same time, the device will give a notification in the form of an alarm to the concerned person. The data can be frequently monitored by the guardian and the concerned doctor. This may help to keep an eye on the people even if they are far away from the person and the stored data can be viewed at any time from anywhere. Thus, the wearable device will record the health parameters of a person, which may assist them to know their mental and physical health, as well as give alerts in case of abnormalities. Implementation of this system will be helpful for the people to get an awareness about their health condition and also make them stay healthy.

The evaporator used in the pharmaceutical industry is for drug preparation. The purpose of the evaporator in drug manufacturing is to extract the water content in the material through the heating process. In this research, the SISO evaporator is taken, which contains temperature as input and dry matter content as output. The mathematical modelling of the drug preparation evaporator is done with the help of the system identification method. Controlling and maintaining the temperature inside an evaporator is a tedious process. In this regard, the Neural Network predictive controller (NNPC) is designed and implemented for drug preparation. It helps to predict the future performance of the evaporator and tune the control signal based on that. The setpoint tracking challenge is given to the designed controller. For analysing the performance of the controller, the error metrics, such as integral square error (ISE), integral absolute error (IAE), integral time square error (ITSE), and integral time, absolute error (ITAE), are employed. The time-domain specification, such as rise time, settling time, and overshoot, is also used to better understand controller performance. From the above two analyses, the conclusion is made that the predictive controller is performing well in comparison with the conventional PID controller in the drug preparation pharmaceutical industry. 

 Breast cancer is the 2nd frequent occurrence of cancer among women, after skin cancer, according to the American Cancer Society. By using mammography, it is possible to detect breast cancer before it has spread to other parts of the body. It primarily affects females, though males can be affected as well. Early identification of breast cancer improves survival chances significantly, however, the detection procedure remains difficult in clinical studies. To solve this problem, a Machine Learning (ML) algorithm is used to detect breast cancer in mammogram images. In this study, 100 images from the mini-MIAS mammogram database were used, 50 of which were malignant and 50 of which were benign breast cancer mammograms. Before training the model, the sample image datasets are pre-processed using numerous techniques. The required features are then extracted from the sample images using Feature Extraction (FE) techniques, such as Daubechies (DB4) and HAAR. Finally, the extracted features are fed into ML classifiers such as Linear Discriminant Analysis (LDA), Support Vector Machine (SVM), and Random Forest (RF) to create a model. Several performance metrics are used to evaluate FE and classification. According to the results of the analysis, the HAAR FE with the RF model is the ideal combination, with an accuracy level of 91%.

Glaucoma is an eye disease that can result in permanent blindness if not detected and treated in the early stages of the disease. The worst part of Glaucoma is that it does not come up with a lot of visible symptoms, instead, it can go from the preliminary stage to a serious issue quickly. A Deep Learning (DL) model is capable of detecting the presence of Glaucoma by analyzing the image of the retina which is uploaded by the user. In this research, two DL algorithms were used to detect the presence of Glaucoma in the uploaded image. The DL algorithms include the convolutional neural network or the CNN and the transfer learning algorithm. The transfer learning algorithm is implemented by the VGG-19 model. Once two DL models were developed using the above-mentioned algorithms, the models were trained and tested using the images of the retina. The trained models are tested to find the better model. The efficiency of the model is measured based on some metrics. These metrics include the True Positive (TP), True Negative (TN), False Positive (FP), and False Negative (FN). Using these metrics, the true positive rate, the true negative rate, the false-positive rate, and the false-negative rate are calculated. From the above values, the DL algorithm, which is more efficient than the other one in identifying Glaucoma, can be found.

Lung cancer is a form of carcinoma that develops as a result of aberrant cell growth or mutation in the lungs. Most of the time, this occurs due to daily exposure to hazardous chemicals. However, this is not the only cause of lung cancer; additional factors include smoking, indirect smoke exposure, family medical history, and so on. Cancer cells, unlike normal cells, proliferate inexorably and cluster together to create masses or tumors. The symptoms of this disease do not appear until cancer cells have moved to other parts of the body and are interfering with the healthy functioning of other organs. As a solution to this problem, Machine Learning (ML) algorithms are used to diagnose lung cancer. The image datasets for this study were obtained from Kaggle. The images are preprocessed using various approaches before being used to train the image model. Texture-based Feature Extraction (FE) algorithms such as Generalized Low-Rank Models (GLRM) and Gray-level co-occurrence matrix (GLCM) are then used to extract the essential characteristics from the image dataset. To develop a model, the collected features are given into ML classifiers like the Support Vector Machine (SVM) and the k-nearest neighbor's algorithm (k-NN). To evaluate FE and classification, several performance metrics are used, such as accuracy, error rate, sensitivity specificity, and so on.

It is often difficult to diagnose several lung illnesses, such as atelectasis and cardiomegaly, as well as Pneumonia, in hospitals due to a scarcity of radiologists who are educated in diagnostic imaging. If pneumonia is diagnosed early enough, the survival rate of pulmonary patients suffering from the disease can be improved. Most of the time, chest X-ray (CXR) pictures are used to detect and diagnose pneumonia. When it comes to detecting pneumonia on CXR images, even an experienced radiologist may have difficulty. It is vital to have an automated diagnostic system to improve the accuracy of diagnostic results. It is estimated that automated pneumonia detection in energy-efficient medical systems has a substantial impact on the quality and cost of healthcare, as well as on response time. To detect pneumonia, we employed deep transfer learning techniques such as ResNet-18 and VGG-16. Each of the model's four standard metrics, namely accuracy, precision, recall, and f1-score, are used to evaluate. The best model is established by the use of metrics. To make pneumonia detection simple, the website is designed by employing the best model.

Melanoma, the deadliest form of skin cancer, is becoming more common every year, according to the American Cancer Society. As a result of the artifacts, low contrast, and similarity to other lesions, such as moles and scars on the skin, diagnosing skin cancer from the lesions might be difficult. Skin cancer can be diagnosed using a variety of techniques, including dermatology, dermoscopic examination, biopsy and histological testing. Even though the vast majority of skin cancers are non-cancerous and do not constitute a threat to survival, certain more malignant tumors can be fatal if not detected and treated on time. In reality, it is not feasible for every patient to have a dermatologist do a complete examination of his or her skin at every visit to the doctor's office or clinic. To solve this challenge, numerous investigations are being conducted to provide computer-aided diagnoses. In this work, skin cancer can be predicted from an image of the skin using deep learning techniques such as convolutional neural networks. The accuracy and loss functions of the model are used to evaluate its overall performance. The mobile app is created to detect skin cancer using the developed model. As soon as the images have been submitted, the app can communicate with the user about their progress.

Dementia is a state of mind in which the sufferer tends to forget important data like memories, language, etc.. This is caused due to the brain cells that are damaged. The damaged brain cells and the intensity of the damage can be detected by using Magnetic Resonance Imaging. In this process, two extraction techniques, Gray Level Co-Occurrence Matrix (GLCM) and the Gray Level Run-Length matrix (GLRM), are used for the clear extraction of data from the image of the brain. Then the data obtained from the extraction techniques are further analyzed using four machine learning classifiers named Support Vector Machine (SVM), K-Nearest Neighbor (KNN), Random Forest (RF), and the combination of two classifiers (SVM+KNN). The results are further analyzed using a confusion matrix to find accuracy, precision, TPR/FPR - True and False Positive Rate, and TNR/FNR – True and False Negative Rate. The maximum accuracy of 93.53% is obtained using the GLRM Feature Extraction (FE) technique with the combination of the SVM and KNN algorithm.

There has been an increase in new diseases in recent years, which has had both economic and societal consequences. Patients in the modern environment require not only constant monitoring but also all-encompassing smart healthcare solutions. These systems keep track of the patient's health, store data, and send alerts when critical conditions arise. Healthcare may be considerably improved with the use of Artificial Intelligence and Machine Learning (ML) systems. These systems can help with earlier diagnosis of diseases, as well as more specific treatment plans. As big data, the Internet of Things with many more smart technologies grows more widespread; deep learning is becoming more popular. Due to the apparent rising complexity and volume of data in healthcare, artificial intelligence (AI) will be used more frequently. This work aims to develop a deep learning-based smart healthcare monitoring system. This system keeps track of patients' health, analyses numerous parameters, categorizes data, and organizes requirements. The algorithm using the python program is developed and discussed to track the health of several patients with various illnesses. This method also aids in the categorization of data, organization of pharmacological requirements. This approach yields satisfactory performance, and the results are also provided.

Cardiovascular disease has a significant global impact. Cardiovascular disease is the primary cause of disability and mortality in most developed countries. Cardiovascular disease is a condition that disturbs the structures and functionality of the heart and can also be called heart disease. Cardiovascular diseases require more precise, accurate, and reliable detection and forecasting because even a small inaccuracy might lead to fatigue or mortality. There are very few death occurrences related to cardio sickness, and the amount is expanding rapidly. Predicting this disease at its early stage can be done by employing Machine Learning (ML) algorithms, which may help reduce the number of deaths. Data pre-processing can be employed here to eliminate randomness in data, replace missing data, fill in default values if appropriate, and categorize features for forecasting and making decisions at various levels. This research investigates various parameters that are related to the cause of heart disease. Several models discussed here will come under the supervised learning type of algorithms like Support Vector Machine (SVM), K-nearest neighbor (KNN), and Naïve Bayes (NB) algorithm. The existing dataset of heart disease patients from the Kaggle has been used for the analysis. The dataset includes 300 instances and 13 parameters and 1 label are used for prediction and testing the performance of various algorithms. Predicting the likelihood that a given patient will be affected by the cardiac disease is the goal of this research. The most important purpose of the study is to make better efficiency and precision for the detection of cardiovascular disease in which the target output ultimately matters whether or not a person has heart disease.

Nowadays, an image must be verified or analyzed keenly for further processing methods. Few numbers of images can be analyzed manually for a particular object or a human being. But when millions and millions of images are present in a dataset, and every single one of them must be verified and classified based on the objects present in the image, it is necessary to find an algorithm or a technique to assist this process. Of many types and applications of object detection, this research aims at pedestrian detection. Pedestrian detection is a special way that only aims to detect the human beings in the uploaded image. The Mask R-CNN algorithm is used in pedestrian detection. The Mask R-CNN model is a Deep Learning (DL) model that can detect a certain object from an image and can also be used for image augmentation. The Mask R-CNN stands for Mask Regional Convolutional Neural Network (CNN). This is one of the classification techniques which can be designed using the Convolutional Network theories. The Mask R-CNN algorithm is the updated version of the Faster RCNN. The main difference between the faster R-CNN and the Mask R-CNN is that the mask R-CNN can bind the borders of the object detected while the faster R-CNN uses a box to identify the object. One of the major advantages of mask R-CNN is that it can provide high-quality image augmentation and is also one of the fastest image segmentation algorithms. This model can be easily implemented when compared to other object detection techniques. Python contains various inbuilt dependencies which can be installed with the help of repositories. These dependencies are used to design the model. The model is then trained and tested with various inputs for better accuracy. Image segmentation is used in various fields like medical imaging, video surveillance, traffic control, etc., so the mask R-CNN technique would be an extremely efficient DL algorithm.

 Agriculture provides a living for half of India's people. The infection in crops poses a danger to food security, but quick detection is hard due to a lack of facilities. Nowadays, Deep learning will automatically diagnose plant diseases from raw image data. It assists the farmer in determining plant health, increasing productivity, deciding whether pesticides are necessary, and so on. The potato leaf is used in this study for analysis. Among the most devastating crop diseases is potato leaf blight, which reduces the quantity and quality of potato yields, significantly influencing both farmers and the agricultural industry as a whole. Potato leaves taken in the research contain three categories, such as healthy, early blight, and late blight. Convolution Neural Network (CNN), and Convolution Neural Network- Long Short Term Memory(CNN-LSTM) are two neural network models employed to classify plant diseases. Various performance evaluation approaches are utilized to determine the best model.

Agriculture is the backbone of human civilization since it is a requirement of every living entity. Paddy agriculture is extremely important to humans, particularly in Asia. Farmers are currently facing a deficit in agricultural yield owing to a variety of factors, one of which is illness. The composition of paddy crop diseases is complicated, and their presentation in various species is highly similar, making classification challenging. These agricultural infections must be discovered and diagnosed as soon as feasible to avoid disease transmission. The disease significantly impacts crop productivity, and early detection of paddy infections is critical to avoiding these consequences. These issues arise as a result of a lack of awareness regarding health. Identifying the disease needs the best expertise or previous knowledge to regulate it. This is both difficult and costly. To address the aforementioned problem, a Deep Learning (DL) model was created utilizing a Convolutional Neural Network (CNN) and the transfer learning approach. The model is trained using an image of a paddy crop as input. By comparing metrics like accuracy and loss, the optimum technique is identified.

A brain tumor is defined by the proliferation of aberrant brain cells, some of which may progress to malignancy. A brain tumor is usually diagnosed via a magnetic resonance imaging (MRI) examination. These images demonstrate the recently observed aberrant brain tissue proliferation. Several academics have examined the use of machine learning and Deep Learning (DL) algorithms to diagnose brain tumors accurately A radiologist may also profit from these forecasts, which allow them to make more timely decisions. The VGG-16 pre-trained model is employed to detect the brain tumor in this study. Using the outcomes of training and validation, the model is completed by employing two critical metrics: accuracy and loss. Normal people confront numerous challenges in scheduling a doctor's appointment (financial support, work pressure, lack of time). There are various possibilities for bringing doctors to patients' homes, including teleconferencing and other technologies. This research creates a website that allows people to upload a medical image and have the website predict the ailment. The Google Cloud Platform (GCP) will be utilized to install the DL model due to its flexibility and compatibility. The customized brain tumor detection website is then constructed utilizing HTML code. 

Different types of brain illnesses can affect many parts of the brain at the same time. Alzheimer's disease is a chronic illness characterized by brain cell deterioration, which results in memory loss. Amnesia and ambiguity are two of the most prevalent Alzheimer's disease symptoms, and both are caused by issues with cognitive reasoning. This paper proposes several feature extractions as well as Machine Learning (ML) algorithms for disease detection. The goal of this study is to detect Alzheimer's disease using magnetic resonance imaging (MRI) of the brain. The Alzheimer's disease dataset was obtained from the Kaggle website. Following that, the unprocessed MRI picture is subjected to several pre-processing procedures. Feature extraction is one of the most crucial stages in extracting important attributes from processed images. In this study, wavelet and texture-based methods are used to extract characteristics. Gray Level Co-occurrence Matrix (GLCM) is utilized for the texture approach, and HAAR is used for the wavelet method. The extracted data from both procedures are then fed into ML algorithms. The Support Vector Machine (SVM) and Linear Discriminant Analysis (LDA) are used in this investigation. The values of the confusion matrix are utilized to identify the best technique.