Application of Artificial Intelligence in Laboratory and Microbiological Diagnostics
Code: G-1200
Authors: Mojtaba Asadi * ℗, Zahra Khajehzadeh Yavari
Schedule: Not Scheduled!
Tag: Biomedical Signal Processing
Download: Download Poster
Abstract:
Abstract
Background and aims: In recent years, artificial intelligence (AI) has emerged as an innovative and comprehensive approach in laboratory and microbiological diagnostics. Traditional methods for microbial identification heavily rely on culture-based techniques, which are often time-consuming and susceptible to contamination and error. Recent advancements in AI have provided innovative solutions to these challenges, offering not only enhanced accuracy but also rapid diagnostic capabilities. Method: AI encompasses various technologies, such as machine learning (ML), deep learning (DL), and natural language processing (NLP). By analyzing large volumes of clinical and genomic data, AI demonstrates the ability to quickly and accurately identify microorganisms through machine learning. This technology can classify 16S rRNA sequencing data, which serves as the basis for bacterial species identification, with an accuracy exceeding 95%, significantly outperforming traditional methods. Results: AI has shown remarkable performance, particularly in image analysis, where it can be utilized for the automated analysis and detection of microscopic images, such as Gram-stained microscopy for distinguishing between Gram-positive and Gram-negative bacteria. AI employs machine learning algorithms and multilayer convolutional neural networks (CNNs) to analyze polymerase chain reaction (PCR) images and gel electrophoresis results. Furthermore, BioFire technology can be used for image analysis of PCR. AI enhances the accuracy and efficiency of result interpretation and facilitates the automation of diagnostic processes. Conclusion: However, several prerequisites are necessary for the widespread adoption and increased efficacy of AI, such as the preprocessing and preparation of laboratory data for AI utilization. Given that various tests and medications can have different nomenclatures, there is a critical need for standardization of scientific terminologies for tests and their measurement units at an international level. This standardization will enable machine learning-based AI to leverage this information, yielding more reliable results.
Keywords
Artificial Intelligence, Microbial Diagnosis, Machine
