Nearest neighbor search with applications in medical image manipulation

A fundamental problem in computer science, nearest neighbor search (NNS) is used in many different fields such as web search, image processing, and recommender systems. Due to diverse applications, this problem remains an active area of research in data science, data management, and medical image processing. Extensive research has been done on this topic, so it is necessary to organize a summary of existing results. We bridge this gap by doing a thorough analysis of the topic with an emphasis on different variants of the problem in different domains. Different data structures used for the problem along with applications have been listed. It has been shown in past studies that NNS can be used in medical imaging. Several times attackers in the online setting manipulate the medical images to perform fraud. This work uses an algorithm based on convoluted neural network (CNN) and the NNS to determine whether medical images have been modified or not. The obtained results describes that the proposed algorithm provides an efficient way to check whether medical images are manipulated or not. In the end, we list out many future research directions.