Artificial intelligence (AI) is not very prevalent in the healthcare sector. It will promote and enhance automation in various aspects of patient care by promoting diagnosis of disease at a subclinical stage which can be otherwise missed by human clinicians. There is a rapidly increasing interest in machine learning (ML) applications in medical care. Precision medicine, neural networks, and deep learning methods of ML have gained importance in the healthcare domain. Based upon the patient attributes, the prediction about the prognosis of disease is possible through precision medicine approach of ML. Likewise, neural networks and deep learning methods are sufficiently capable of predicting the outcomes of the patient disease, which is otherwise less predictable due to the lack of prediction models in the clinical practice. WSN, IoT, IoMT have gained popularity among all the stakeholders in the hospital settings. Monitoring of the patients has become more viable with an application of wireless communication, which is more cost effective and energy saving for the patients.
Top1. Introduction
Artificial Intelligence (AI) is the ability of the intelligent computer programming to adopt machine algorithms based on human intelligent processes of cognition, meta-cognition, learning and auto-correction, for the decision making related to diagnosis and prediction of outcomes of various bodily system diseases. The modernization of the clinical development process has been outlined by the integration of AI and machine learning (ML) based algorithms and digital methods (Ahmed et al.,2020). It can be quite useful and popular in the health care industry due to rapid rise in data complexity and it has been noticed in published literature that the demand of inclusion of AI has increased at many instances. The use of AI- & ML- have been explored in the following major areas –
- 1.
Drug discovery using machine-based learning.
- 2.
Diagnosis at sub-clinical stage and monitoring of progression of disease; preparation of algorithms for computational expansion of prevailing datasets (clinical and imaging).
- 3.
Detection of new prediction model using Deep learning methods.
Wireless sensor networks (WSN) have gained an importance as a major connector of all key stakeholders of the hospital. Internet of Things (IoT) has become popularized among healthcare sector due to its cost effectiveness and characteristics of autonomous sensor operations methods to monitor vitals such as blood pressure, temperature, heart rate etc. and to take an emergency action when required (Nogueira, 2019). Wireless Body Area Network (WBAN), through low powered biosensor modes monitor the health status of patients, thereby preventing the consumption of energy and costs of communication. WSN will allow patients to live independently in their homes and for longer periods with family members and will enable to improve the quality of life of patients.
The organization of the chapter is as follows. Section 2 highlights an overview of various ML methods. Section 3 presents an overview of wireless communication networks with a special emphasis on Internet of things/Internet of Medical Things (IoT/IoMT), WBAN and Bluetooth Low Energy (BLE). Section 4 delineates the emerging trends and applications of ML and WSN. Section 5 addresses the challenges and ethical implications of AI and WSN. Finally, the conclusion highlighting the key points and future aspects of AI and WSN in the healthcare.
Top2. Overview Of Machine Learning Methods
Machine Learning (ML) is a broad muti-disciplinary specific technique of AI which has proliferated in the healthcare industry. It is the method of procuring information from the training data. It is omnipresent and is used widely in the various fields like finance, IT industry, security and medical care science. The important characteristics attributed to ML which has made it popularized in the health industry are higher precision, scalability, adaptability and robustness. In the last decade, the application of ML has been accelerated in the healthcare settings due to significant improvement in resources of data, computational analysis, innovative methods. Handling of large and heterogenous data, revealing of complicated and mystical forms and forecast of complex outcomes is possible these days due to availability of various ML methods. In the clinical trials, ML is efficient and capable to enhance the patient – centeredness, external validation, success of results. As a specific AI technique, it has proven to be intelligent in processing the multi-sensor data in the hospital settings, to augment better care to the geriatric population; to optimize and monitor the physical condition of the children, and in the prediction of various life – threatening cardiovascular diseases at a very early stage and thus contributes in the early prevention and treatment plan. ML can be categorized under the following categories as shown in Figure 1.