ABSTRACT
Nowadays healthcare data analytics plays an important role due to the tremendous increase in volume of multimodal clinical data such as biomedical images, electronic health records, and data from sensors, signals, etc. Mining disease patterns from such multimodal data helps in making decisions by clinicians. Medical imaging plays a major role in the current scenario due to its enormous competency in capturing high-resolution images of anatomical structures in humans. Meritoriously examining such high-resolution images can be useful for clinicians in ailment monitoring, surgical planning, and diagnosis. The imaging modalities to obtain biomedical images are magnetic resonance imaging (MRI), computed tomography (CT), positron emission tomography (PET), and ultrasound (U/S). The main objective of biomedical image analysis is to mine quantitative data about the disease patterns that can provide more insights about the medical condition of the patient. There are various challenges since the images may contain noisy values, irregularities in shapes. Numerous challenges arise in analyzing biomedical images such as image segmentation, image registration, object detection, and feature extraction and pattern recognition. Artificial intelligence and deep learning are driving advances in the healthcare industry. Application of deep learning algorithms helps in recognizing the complex patterns from biomedical images. Deep learning techniques help in processing the healthcare data, which results in early detection of disease patterns, personalized optimal treatment plans, fast and more accurate diagnosis. This chapter summarizes the overview of deep learning algorithms for biomedical image analysis.
