Evaluation of IoT-Enabled hybrid model for genome sequence analysis of patients in healthcare 4.0

Abstract

Genome sequence matching is vital for health analytics and treatment in healthcare 4.0. It focuses on finding whether a given sequence resembles other sequences that can help identify disease outbreaks faster. The healthcare 4.0 primarily require more efficient systems that can easily connect and interact with big data. The Internet of things (IoT) can potentially realize smart devices that can perform this role in the health sector. This research presents an IoT-enabled hybrid model for Genome sequence analysis of patients in healthcare 4.0. The proposed model utilizes an improved pattern-matching technique that uses Hadoop's ideas, OpenCL, and APAR (API). The primary local alignment search tool (BLAST) is the core algorithm extensively used to compare and find preliminary biological sequence information and calculate the matches' statistical importance. BLAST enables many strings to be distributed in the form of blocks; after that, an instance of the Mapper is mapped to process it after the output of all the mappers combined through the reducer. This mapper and reducer process has inter-node parallelism to speed up the process and efficiently utilize the shared resources. In this work, fine-grained parallelism has been introduced in Hadoop-based BLAST. The proposed work also provided inter-node and intra-node parallelism. The Mapper and reducer accelerated the conventional BLAST algorithm. The proposed model achieved better results than other traditional algorithms for different datasets in all cases.