Artificial intelligence (AI) is changing, at a fast pace, all aspects of science, technology, and society in general, giving rise to what is known as the 4th Industrial Revolution. In this chapter, we review the literature regarding AI applications to bone tissue engineering, and more particularly, to cell adhesion in bone scaffolds. The works found are very few (only six works), and we classify them according to the AI technique used. The question we want to address in this chapter is what AI techniques were used and what exactly have they been used for. The chapter shows that the most used AI tools were the artificial neural network, in their different types, followed by cellular automata and multiagent systems. The intended use varies, but it is mainly related to understanding the variables involved and adjusting a model that provides insight and allows for a better and more informed design process of the scaffold.
This chapter presents various cloud platforms that are available in market offerings from different vendors. IBM provided a machine learning (ML) platform “IBM Watson Studio” (formerly “Data Science Experience”), and this is considered here for the field of study. An overview of artificial intelligence, ML, and deep learning (DL) with their relationship is deliberated. Discussion on popular DL architectures with elementary comparison is also considered.
Today information technologies have led to a vast number of innovations and developments in almost every field of advancement. In this context, recommender systems (RS) have been setting milestones in the service industry. If we look at the web-based services, it can be said that RS has contributed majorly to increase the reachability of products and to provide a sea of options for potential customers. This RS can also be used as a tool to support decision-making by the decision-makers. With this advancement of RS in the service industry, healthcare systems also do not lag. Health RS (HRS) are becoming an important platform for providing healthcare services, which would cut down the hectic schedule of visiting the doctors and waiting for hours to get checked. In the healthcare industry, RS already play a very significant role in terms of supporting decision-making process about any individual’s health. Keeping the very limited availability of resources in mind and the need for HRS to make its way into the chapter of milestones and innovations, it is important to introduce a set of knowledge and information for the researchers that are interested in HRS studies and can make huge advancements in this domain. Using HRS to suggest the most probable and appropriate medicines after taking into consideration the history of the individual can be a sub-domain to highly think about. Hence, this paper provides a literature study of the HRS domain in general which includes the literature, innovations, purpose, and methods of HRS, along with the new concept of HRS being used for medication purposes.
Healthcare is one of the domains which has truly advanced to use the latest technologies like machine learning to help in diagnosis. Owing to its complexity of medical images, extracting features correctly makes the problem even tougher. The earlier image processing algorithms using descriptors were unable to detect the disease accurately, and also, using the correct form of descriptors based on the dataset was even a bigger challenge which further reduced the accuracy of the machine learning algorithms trained over the dataset. However the recent advancements in the field of deep learning are able to give better results for these classifications. Convolutional neural networks (CNNs) have proved to be a great algorithm choice in case of extracting spatial features, making it suitable for the medical diagnosis. However, as the number of layers increase in CNNs, the complexity of the network increases and the information passed from one later to another eventually decreases, thus causing information loss. In order to overcome this dense CNN can be considered. We have shown couple of case studies which were performed. One using the local binary patterns and other using dense CNN on two different types of medical images. The dense CNN work was also recognized by the IEEE Computer Society.
The development, approbation, and acceptance of various social media tools and applications have opened new doors of opportunity for gaining crucial insight from unstructured information. Sentiment analysis and opinion mining have become popular in modern years and can be applied in diversified application areas like healthcare informatics, sports, financial sector, politics, tourism, and consumer activities and behavior. In this regard, this chapter presents how sentiment analysis can help for betterment of people suffering from critical diseases. Healthcare-related unstructured tweets relating to being shared on Twitter is becoming crowd-pleasing source of information for healthcare research. Sentiment analysis is becoming metric measurement to find out feelings or opinion of patient suffering from severe diseases. Various tools and methodologies are used, from which color-coded Word Cloud can be formed based on sentiment. Exploring the methods used for sentiment analysis on healthcare research can allow us to get better insight and understanding of human feelings and their psychology and mindset. The study shows various types of tools used in each case and different media sources and examines its impact and improvement in diseases like obesity, diabetes, cardiovascular disease, hypertension, schizophrenia, Alzheimer’s disease, and cancer using sentiment analysis and its impact on one’s life. Sentiment analysis helps in designing strategies to improve patients understanding and behavior.
In the analysis and preprocessing of images, image segmentation is a very important step. Due to their simplicity, robustness, reduced convergence times, and accuracy, standard multilevel thresholding techniques for bilevel thresholds are efficient. However, a number of computational expenditures are needed, and efficiency is broken down as extensive research is used to decide the optimum thresholds, resulting in the implementation of evolutionary algorithms and swarm intelligence (SI) to achieve the optimum thresholds. Object segmentation’s primary objective is to distinguish the foreground from the background. By optimizing Shannon or fuzzy entropy based on the neural network optimization algorithm, this chapter provided a multilevel image border for object segmentation. The suggested algorithm is evaluated on standard image sets using Firefly algorithm (FA), Differential Evolution (DE), and particle swarm optimization, and the results are compared with entropy approaches for Shannon or fuzzy. The suggested approach shows better efficiency in objective factor than state-of-the-art approaches, structural similarity index, Peak signal to noise ratio (PSNR), and standard derivation.
Machine learning (ML) is an application of AI (artificial intelligence), which deals with the study of capability of a computer to learn from the given data to gain knowledge in making predictions and decisions based on its experience. Such technology can benefit healthcare industry to a great extent. It is the fastest growing industry with high rates of progress in the field of health with new technologies emerging rapidly. These can be extended to a wide range of clinical tasks and prediction tasks since the performance of ML algorithms has been proved to be more than that of humans. Nowadays, all of the patient data has been recorded on computers, and the existing patient data can be used by the doctors and examiners for follow-ups. ML algorithms use this existing data and analyze them to identify patterns that are used to make precise diagnosis and provide better care to patients. With the invention of wearables, all the patient data has been monitored and stored, which is then used by ML for better patient management. ML algorithms are also being used to accurately predict the progress of a disease. This innovation can give chances to improve the proficiency and quality of healthcare.
This chapter is a recent survey of evolutionary computation (EC)-based feature selection (FS) techniques whose objective is mainly to improve the accuracy of the machine learning algorithm in minimized computation time. The idea is to bring forth the main strengths of EC as a naturally inspired optimization technique for FS in the machine learning process. The modeling of biological and natural intelligence that has made progressive advancements in the recent decade motivates us to review state-of-the-art FS techniques to add more to the area of computational intelligence. Owing to its importance, the use-case in support of health informatics is also exhibited.
A lot has changed in the world since the inception of the so-called deep learning (DL) era. Unlike conventional machine learning algorithms, where human learns more of feature extraction than machine, which just crunches numbers, DL algorithms are quite promising as they have revolutionized the way we deal with data and have become adept in taking human-like decisions. Based on the powerful notion of artificial neural networks, these learning algorithms learn to represent data or, if we rephrase, can extract features from raw data. DL has paved way for end-to-end systems where one learning algorithm does all the tasks. Convolutional neural networks have earned a lot of fame lately, especially in the domain of computer vision where in some cases its performance has beaten that of humans. A lot of work has been done on convNets and in this chapter we will demystify how convolutional neural networks work and will illustrate using one novel application in the field of astronomy where we will do galaxy classification using raw images as input and classify them based on its shape. In addition to this we will investigate what features the network is learning. We will also discuss how DL superseded other forms of learning and some recent algorithmic innovations centered on convolutional neural nets.
Health informatics primarily means dealing with the methodologies that help to acquire, store, and use information in health and medicine. Large database including heterogeneous and complex data of healthcare can be very beneficial but it is difficult for humans to interpret such a “big data.” With such a large dataset, machine learning (ML) algorithms can work very well in predicting the disease and treatment methods. ML algorithms include learning from past experience and making the predictions and decisions for the current problems. There are many challenges that are encountered while applying ML in healthcare and mostly in healthcare applications where dataset is very complex and is of varied type (ranging from texts to scans). A possible alternative is the use of interactive ML where doctors can be taken in loop. Hence, an integrated and extensive approach is required for application of ML in health informatics. This chapter deals with the various types of ML techniques, approaches, challenges, and its future scope in healthcare informatics. Further, these techniques can be used to make a model for quick and precise healthcare discovery.