A Deep-Learning Based System for Rapid Genus Identification of Pathogens under Hyperspectral Microscopic Images

Abstract

Infectious diseases have always been a major threat to the survival of humanity. Additionally, they bring an enormous economic burden to society. The conventional methods for bacteria identification are expensive, time-consuming and laborious. Therefore, it is of great importance to automatically rapidly identify pathogenic bacteria in a short time. Here, we constructed an AI-assisted system for automating rapid bacteria genus identification, combining the hyperspectral microscopic technology and a deep-learning-based algorithm Buffer Net. After being trained and validated in the self-built dataset, which consists of 11 genera with over 130,000 hyperspectral images, the accuracy of the algorithm could achieve 94.9%, which outperformed 1D-CNN, 2D-CNN and 3D-ResNet. The AI-assisted system we developed has great potential in assisting clinicians in identifying pathogenic bacteria at the single-cell level with high accuracy in a cheap, rapid and automatic way. Since the AI-assisted system can identify the pathogenic genus rapidly (about 30 s per hyperspectral microscopic image) at the single-cell level, it can shorten the time or even eliminate the demand for cultivating. Additionally, the system is user-friendly for novices.

Dual-Readout Ultrasensitive Lateral Flow Immunosensing of Salmonella typhimurium in Dairy Products by Doping Engineering-Powered Nanoheterostructure with Enhanced Photothermal Performance

“…12 Although dependable detection results can be obtained, the detection process is time-consuming and laborious, and that is difficult to fulfill the pressing needs of immediate detection. 13,14 Therefore, it is of considerable significance to establish a rapid and accurate specific detection method for the early screening and disease control of S. typhimurium.…”

“…typhimurium can induce nosocomial infection and fulminant food poisoning with a high fatality rate. , Although antimicrobial intervention is an effective means of addressing bacterial contamination, eggs and milk with Salmonella are identified during foodborne and point source infection. , The source of pollution can be traced back to the farm through transportation chain . Although dependable detection results can be obtained, the detection process is time-consuming and laborious, and that is difficult to fulfill the pressing needs of immediate detection. , Therefore, it is of considerable significance to establish a rapid and accurate specific detection method for the early screening and disease control of S. typhimurium.…”

Rapid and robust bacterial species identification using hyperspectral microscopy and gram staining techniques

“…Hyperspectral imaging, which can obtain both spectral and spatial sample data, can identify and label species based on their spectral signatures. This technique has been extensively adopted in biological investigations for cancer cells [7,8], fungi [9,10], and bacteria [11][12][13]. Predominantly, hyperspectral imaging studies targeting bacterial identification have relied on unstained samples.…”

Hyperspectral upgrade solution for biomicroscope combined with Transformer network to classify infectious bacteria

“…In 2022, Chenglong Tao et al combined hyperspectral microscopy and the deep learning-based algorithm Buffer Net to build an AI-assisted system for automated rapid bacterial genus identification and compared the effects of different network structures on classification results. After training and validation on a self-constructed dataset, the Buffer Net could achieve an accuracy of 94.9%, outperforming 1D-CNN, 2D-CNN, and 3D-ResNet [18].…”