Deep Learning-Based Culture-Free Bacteria Detection in Urine Using Large-Volume Microscopy.

Biosensors (Basel)

Biodesign Center for Biosensors and Bioelectronics, Arizona State University, Tempe, AZ 85287, USA.

Published: February 2024

Bacterial infections, increasingly resistant to common antibiotics, pose a global health challenge. Traditional diagnostics often depend on slow cell culturing, leading to empirical treatments that accelerate antibiotic resistance. We present a novel large-volume microscopy (LVM) system for rapid, point-of-care bacterial detection. This system, using low magnification (1-2×), visualizes sufficient sample volumes, eliminating the need for culture-based enrichment. Employing deep neural networks, our model demonstrates superior accuracy in detecting uropathogenic compared to traditional machine learning methods. Future endeavors will focus on enriching our datasets with mixed samples and a broader spectrum of uropathogens, aiming to extend the applicability of our model to clinical samples.

Download full-text PDF

Source
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC10887190PMC
http://dx.doi.org/10.3390/bios14020089DOI Listing

Publication Analysis

Top Keywords

large-volume microscopy
8
deep learning-based
4
learning-based culture-free
4
culture-free bacteria
4
bacteria detection
4
detection urine
4
urine large-volume
4
microscopy bacterial
4
bacterial infections
4
infections increasingly
4

Similar Publications

Single-shot 3D optical microscopy that can capture high-resolution information over a large volume has broad applications in biology. Existing 3D imaging methods using point-spread-function (PSF) engineering often have limited depth of field (DOF) or require custom and often complex design of phase masks. We propose a new, to the best of our knowledge, PSF approach that is easy to implement and offers a large DOF.

View Article and Find Full Text PDF

Until recently, the lack of three-dimensional visualisation of whole cells at the electron microscopic (EM) level has led to a significant gap in our understanding of the interaction of cellular organelles and their interconnection. This is particularly true with regard to the role of the endoplasmic reticulum (ER). In this study, we perform three-dimensional reconstructions of serial FIB/SEM stacks and anaglyphs derived from volume rendering, cryo-scanning electron microscopy (cryo-SEM) and state-of-the-art electron microscopy immobilisation and imaging techniques.

View Article and Find Full Text PDF

Solid-State Photoswitching of Hydrazones Based on Excited-State Intramolecular Proton Transfer.

J Am Chem Soc

January 2025

Department of Organic Chemistry, Faculty of Natural Sciences, Comenius University, Ilkovičova 6, Bratislava SK-842 15, Slovakia.

The development of new photochromic systems is motivated by the possibility of controlling the properties and functions of materials with high spatial and temporal resolution in a reversible manner. While there are several classes of photoswitches operating in solution, the design of systems efficiently operating in the solid state remains highly challenging, mainly due to limitations related to confinement effects. Triaryl-hydrazones represent a relatively new subclass of bistable hydrazone photoswitches exhibiting efficient / photochromism in solution.

View Article and Find Full Text PDF

Taking advantage of the good mechanical strength of expanded Drosophila brains and to tackle their relatively large size that can complicate imaging, we apply potassium (poly)acrylate-based hydrogels for expansion microscopy (ExM), resulting in a 40x plus increased resolution of transgenic fluorescent proteins preserved by glutaraldehyde fixation in the nervous system. Large-volume ExM is realized by using an axicon-based Bessel lightsheet microscope, featuring gentle multi-color fluorophore excitation and intrinsic optical sectioning capability, enabling visualization of Tm5a neurites and L3 lamina neurons with photoreceptors in the optic lobe. We also image nanometer-sized dopaminergic neurons across the same intact iteratively expanded Drosophila brain, enabling us to measure the 3D expansion ratio.

View Article and Find Full Text PDF

Cryo-electron tomography (cryo-ET) has become a key technique for obtaining structures of macromolecular complexes in their native environment, assessing their local organization and describing the molecular sociology of the cell. While microorganisms and adherent mammalian cells are common targets for tomography studies, appropriate sample preparation and data acquisition strategies for larger cellular assemblies such as tissues, organoids or small model organisms have only recently become sufficiently practical to allow for in-depth structural characterization of such samples in situ. These advances include tailored lift-out approaches using focused ion beam (FIB) milling, and improved data acquisition schemes.

View Article and Find Full Text PDF

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!