Disease Prediction using Machine Learning, Deep Learning and Data Analytics

In the modern era, there is a boom in automating medical diagnosis by adopting emerging technologies and advanced applications of artificial intelligence. These technologies require a huge amount of data for training the models and precisely predicting the disease or disorder. Multiple organizations can contribute data for such systems but maintaining data privacy while sharing the data is a major challenge. Also, provisioning a large data corpus for the performance improvement of machine learning and deep learning models in the healthcare domain while keeping the patient’s medical confidentiality intact is a point of concern. Thus, there is a strong need to preserve the privacy of medical data. This calls for the use of up-to-the-minute technologies where the necessity of sharing raw data is completely eradicated, while each organization receives a catered infrastructure for processing data. A cross-silo federated learning model is based on the concept of decentralized data weights collection from multiple clients which are then processed on the central server for modeling and aggregation, thus maintaining data privacy in its true sense. The authors in this manuscript provide a detailed comparative study of the different deep learning-based models in federated learning and how efficiently they can classify lung X-Ray images into three classes: Covid-19, Pneumonia, and Normal. This study can provide a benchmark for the researchers looking forward to deep learning-based model applications of cross-silo federated learning in healthcare.

Three-dimensional geometry of a bone is important in the correct diagnosis of a disease, arthritis, or other bone deformities. The modalities such as Computer Tomography Scans and Magnetic Resonance Imaging are used for a three-dimensional view of a bone. Both the above- stated modalities have high costs and expose the patient to strong carcinogenic radiations. Computer Tomography captures an extensive number of images to collect the required information from a bone. Another modality Magnetic Resonance Imaging is more suitable for retrieving information from soft tissues rather than bones. Therefore, it becomes less effective to read the pathology from bones. This has motivated the authors to identify imaging techniques useful in detecting the pathology or deformity in bones. Also, this is the need of the hour to provide a low cost and safer technique of bone imaging. To address this need, we present a review of the bone imaging techniques and techniques applied for the conversion of two dimensional images into three-dimensional form. We also give the directions for developing the patient-specific and organ-specific optimized techniques for 3-D reconstruction.

The increasing number of health issues is a cause of concern for public as well as health services across the globe. However, a boom in the use of imaging techniques such as CT scans and chest radiographs has been observed for correct diagnosis. But, manual scanning of these modalities requires expertise in modality reading. It is also a time-consuming task. Artificial intelligence-based techniques have proven their potential in pattern recognition, object identification, and data analysis. Therefore, these techniques can be used to provide assisting tools for the primary screening of diseases from these modalities. It has been observed from the literature that a lot of research works are available on disease diagnosis and classification using machine learning, and deep learning. But, the disease severity detection is underexplored. Moreover, the techniques employed for the detection of the severity of diseases have lacunae that need immediate attention. These challenges motivated us to review the machine learning and deep learning-based technological solutions proposed in the literature for the detection of disease severity. The objective of this research is to present a comprehensive survey of research works available about disease severity detection. This research also presents a comparative analysis of the machine learning techniques and deep learning techniques employed, datasets used, and performance achieved. It also highlights the drawbacks of the technological solution proposed. Further, it provides the directions for future scope in the domain of disease severity detection.

The health expert’s crucial task is to interpret the output and treat the disease accordingly. They may delay the decision-making during emergencies. To address this issue, research on smart tools for biomedical applications is much needed which may help in making accurate decisions at the earliest stage. Discovery in medicinal research requires state-of-the-art computer-based tools for diagnosing and treating complex diseases such as cancer, COVID-19, SARS-Cov, MERS-Cov, tuberculosis, brain disorders, heart, and lung-related chronic infections. Among various diagnostic methods, image-based disease identification stands out as the most prominent approach for detecting new and complex diseases. A well-trained computerized biomedical system can provide physicians with enhanced support for early disease detection. Biomedical images are typically acquired from various sources, including CT, ultrasound, MRI, dermoscopy, X-ray, biopsy, and endoscopy. Presently, a wide range of image-analysis procedures are available for biomedical images. These procedures involve image acquisition, pre-processing, segmentation, feature extraction, and classification, all contributing to improved disease decision accuracy. Although many biomedical images are available online free of cost, the proper procedure must be followed to select appropriate images from databases and enhance their quality. This is important for effectively training image-processing algorithms and increasing their efficiency. This leads to improved instrument performance and more valuable insights into the diseases under study. It also handles complex and vast image data to detect early signs of unusual signals, growth, inflammation, cell damage, protein sequence changes, and blockages. Additionally, it should be user-friendly and convincing to health experts to identify hidden biological issues. This chapter emphasizes the power of computerized tools in image analysis and disease detection. It also focuses on recent developments in the field of medicinal research.

The brain is one of the most sensitive parts of the human body which transmits millions of signals every moment. Dementia is the most emerging brain health issue which involves memory loss, difficulty in problem-solving, handling complex tasks, etc. Dementia is a syndrome that causes a loss of mental ability. It affects memory, thinking, shape, comprehension, counting, reading ability, language, and judgment. Dementia affects millions of people and can be the leading cause of death. It is now the seventh leading cause of death worldwide, as well as one of the major causes of impairment and reliance on elderly people. There is no treatment for dementia at present. The importance of early detection and diagnosis in improving early and effective management is crucial. Predicting dementia in advance can lead us to a better life. To predict dementia, various Machine Learning models have been used. In this paper, Dementia is predicted on the basis of MRI Images, for this, three different datasets of MRI Images have been collected. Furthermore, for better prediction, various Machine learning models are used to predict dementia and validate the performance with statistical analysis like K-Nearest Neighbours, XG Boost, Support Vector Machine, Random Forest Algorithm (RFA), and Convolutional Neural Network (CNN). Out of all algorithms, Random Forest Algorithm and Convolutional Neural Network gave the best result with the accuracy of 93.2 and 99.9 respectively.

We present a clinical decision support system for the identification of asthmatics in two different cohorts representing rural and urban populations in India. The input data representing the two populations are cross-sectional in nature and are necessarily categorical in nature, with information on clinical history emphasizing clinical symptoms and patterns characterizing the disease. The system is described as hybrid as it combines the unsupervised and supervised learning techniques in a unique way as discussed in the work presented in the paper. The clustering information emphasizing the phenotypic characterization of asthma is an input to the classifier and a significant improvement is observed in the performance of the classifier. The results of the developed hybrid decision support system are quite promising for suitable deployment in a real-time scenario, as it explores the benefits of both supervised and unsupervised learning techniques. Further, the use of clustering information in the form of cluster evaluation scores as an input parameter to the classifiers can efficiently predict disease outcomes, especially with diseases such as asthma, as the disease is heterogeneous and exhibits several disease subtypes and heterogeneous phenotypes.

Medical image processing has a significant role in clinical investigation and recent medical research. An appropriate image-based medical assessment helps to analyze or detect critical diseases early, as it has a high value of medical information. In this study, medical imaging is reviewed for the diagnosis of eye diseases using computational intelligence. However, the identification of these diseases using traditional image processing is quite complicated. Nowadays, various machine learning and deep learning approaches are developed for the detection of different eye diseases which are helpful for the detection of the diseases at an early stage. Research showed that eye disorders are more serious in emerging or underdeveloped nations due to inadequate healthcare facilities and skilled health workers. An estimate of 45 million people around the world are blind and the tragic fact is that only 75% of these cases are curable. Moreover, the doctor-patient ratio around the globe is about 1: 10,000. Therefore, it takes an hour to create a screening system for the identification of these illnesses. Ophthalmology is close to making breakthroughs in evaluating, diagnosing, and treating eye diseases. Additionally, many eye and vision problems show no obvious signs. As a consequence, people are often unaware that problems exist. Early detection of diseases is a primary concern as they could be easily cured before leading to severity. This research paper focuses on detecting eye illnesses, such as Diabetic retinopathy, Diabetic Macular Edema, Glaucoma, Age macular Degeneration, Retinal Vascular Occlusions, and Retinal Detachment. The authors explore various algorithms, imaging modalities, and challenges in this context. The study aims to raise awareness about eye disorders leading to blindness using computer vision, image processing, and deep learning techniques. It also investigates how these machine learning and deep learning approaches can aid in early disease diagnoses for effective treatment before vision loss occurs.

A personal computer may be used, with input devices such as a keyboard, mouse, and joystick serving as an interface between the computers and the human. The euphemistic, physically challenged are unable to use these computer systems, therefore, BCI technology has advanced external applications to be managed without physical movements in order to assist these physically disabled people and address the limitations of HCI. The technological advancement in the field of cognitive neuroscience and brain imaging has enabled it to communicate directly with the human brain instead of using an interface. Rather than generating signals from muscle movements, these systems use brain activity to monitor computers or communication devices. Researchers in the field of Human-Computer Interaction (HCI) look at ways for machines to utilize as many sensory sources as possible. Furthermore, researchers have begun to consider implicit types of data, input that is not specifically performed to instruct a machine to perform a task. Systems can evolve dynamically based on this data in order to assist the user with the task at hand. Here we discussed components of Brain-Computer Interface, its characteristics and challenges. The researchers are attempting to replace conventional classifiers with Convolutional neural networks (CNNs) that would provide a promising advantage in classification. The EEG signals from the brain can be linked seamlessly to mechanical systems via BCI applications, making it a rapidly growing technology that has applications in fields such as Artificial Intelligence and Computational Intelligence.

Medical database is among the most crucial databases in terms of their applicability to human life. Many researchers are in search of knowledge that is abstracted within the data. Data mining is popular in today's world as it gives access to knowledge that is otherwise unavailable. The concealed knowledge which is offered as a result of it can help the individual to make better decisions. Data mining tools in health have great potential. These solutions may be divided into four categories: therapeutic efficacy assessment, patient care, customer service, and embezzlement monitoring. The authors discovered that giving decision assistance in the medical sector with an emphasis on electronic health records (EHRs) can save lives. Though offering decision assistance in EHRs using data mining is valuable, it needs consistency. As a result, the authors intend to use data mining methods on standardised EHRs to create a decision support system. This paper presents the state-of-the-art data mining approaches and their application in the healthcare sector. It provides an integrated summary and a comparison detail of the existing literature. This chapter surveys several issues that need to be handled before employing data mining on EHRs and further proposes a solution for dealing with these problems. The problems such as multiple origins, multiple formats, missing data, distinguished users, data granularity, flexibility, and sparseness need immediate attention from researchers. Resolving these problems is important to build an efficient standardized EHRs database.

In this technological era, the technology of databases is very essential to many aspects of modern life. To give the prospective medical practitioner, the finest in class and most recent medical knowledge, it seems mandatory that education in the health domain be well-integrated with the most recent databases. This is because there is a growing demand for it and there are benefits from the collaboration of healthrelated issues of the public and database technology. Database technology can help improve health in several ways, including connecting geographically separated health providers and patients, collecting data for research studies like drug and vaccine trials, keeping track of chronic diseases, and guaranteeing that patients follow their prescribed treatments. In this pandemic situation of COVID-19, which the whole world is currently suffering, the current paper attempts to emphasize the databases’ role. It illustrates how the COVID-19 Dataset can be stored, queried, and analyzed, and helps in providing decision support to various end-users. We have performed descriptive analysis by executing specific queries on the COVID-19 Dataset. Then, we performed predictive analysis using two data analysis techniques on the COVID-19 Dataset to approximate the situation in some major cities of India. Further, we have visualized our results to get valuable information from our analysis.