Artificial intelligence in the service of genetics: applications and challenges
Artificial intelligence in the service of genetics: applications and challenges
Shiva Pouyanfar,1,*Nafiseh Soudi,2
1. Student Research Committee, Alborz University of Medical Sciences, Karaj, Iran 2. Department of Laboratory Sciences, Tabriz Faculty of Paramedicine, Tabriz University of Medical Sciences, Tabriz, Iran
Introduction: Advances in technology in recent years have had many and varied effects on medical science. Artificial intelligence is a rapidly growing technology that, in addition to its various roles in everyday human life, also plays an important role in different fields of science (including medicine). On the other hand, one of the fields of medicine that has always been developing rapidly is the field of genetics, which is closely related to technology. The combination of artificial intelligence and genetics is a topic that can be very challenging and regarded as an applied field of study.
Methods: by studying the researches conducted in this field in recent years, we gained an overview of the subject.
Results: Artificial intelligence has wide applications in medicine; including collecting patient information and suggesting possible treatment methods, classifying patient information in statistical studies, facilitation of communication between autistic children and their therapist, diagnosing some fetal abnormalities, assessing fetal viability based on fetal imaging analysis, diagnosing diabetic retinopathy through retinal imaging analysis and classifying patients based on the diagnosis, etc.
Particularly in the field of genetics, artificial intelligence and its various forms (machine learning, deep learning, natural language processing, etc.) are used in information gathering, risk assessment, genome sequencing, pedigree drawing, designing algorithms for specific genetic diagnostic testing, polygenic risk scores for complex conditions, gene-editing CRISPR, literature mining, next-generation sequencing (NGS), variant calling, genome annotation and variant classification, genotype-to-phenotype prediction, phenotype-to-genotype mapping, etc.
Despite the widespread use of artificial intelligence in medicine and genetics, there are limitations and challenges, including the need for large datasets, the black box nature, regulatory issues, privacy, human resistance, data and machine bias, and so on.
Of course, the constant challenge in the field of artificial intelligence has been the fear of humans being replaced by machines, which still exists.
Conclusion: Artificial intelligence is an evolving technology that can reduce the extra workload of geneticists by performing certain tasks, and by saving more of their time and energy, can play an important role in providing better and more accurate services.