Emotions play a vital part in human interactions and are important determinants of user happiness and consumer sentiment. The creation of human-computer interaction (HCI) systems depends on speech emotion recognition (SER) modules. Systems for analysing and classifying speech signals in order to identify the emotions they contain are known as speech emotion recognition (SER) systems. SER is a well-established and fast developing branch of signal processing. More data will be available to learn from as speech emotion detection systems become more popular, helping them to improve their performance. This study also includes a discussion of certain common machine learning approaches, such as ensemble methods, deep learning methods, and classifier enhancement methods. Attention-based deep neural networks (DNNs) are also highlighted in this paper, which is getting a lot of attention in this domain. The findings of this study show that this field of study is being investigated steadily. The scientific community in this area could benefit from this analysis’s inspiration and motivation.

Solid tumor survives by the generation of micro-vessels around it. This phenomenon is known as angiogenesis. Angiogenesis is governed by two factors, namely Tumor Angiogenic Factor (TAF) secreted by the tumor cells and tissue Fibronectin (FNT) concentration in the extra-cellular space. These two factors help in mobilization of endothelial cells from nearby blood vessels. At the initial phase of angiogenesis, neighboring blood vessels affect in formation of capillary sprouts. In this work we have develop an analytical model to make a correlation between tumor system, vasculature system of the cancer milieu and peripheral blood circulatory system. This, in turn, could be an effective and clinically relevant approach for tracking the tumor growth and its angiogenesis with a quantitative manner.

Any kind of heart problem is referred to as a “heart disease,” and involves the blood vessel system within it. It is currently among the world's main causes of death. The creation of a heart disease prediction system which could forecast the possibilities of a person developing a heart disease is the aim of this chapter. Without using feature selection techniques, datasets 1 and 2 had the highest accuracy rates, at 83.06% and 85.25%, respectively. A comparative study is done using two databases having different attributes to find out the most significant feature that involves in the development of heart disease. This is carried out by applying various feature selection methods to discover the appropriate feature for the system.