Microbiology 2.0-A behind the scenes consideration for artificial intelligence applications for interpretive culture plate reading in routine diagnostic laboratories

Article Medical Laboratory Technology

Deep Convolutional Neural Networks Implementation for the Analysis of Urine Culture

David J. Alouani et al.

Summary: This study developed a deep learning model, BacterioSight, that can improve the consistency of urine culture image interpretations. The model achieved comparable accuracy to standard-of-care-trained technologists, with a range from 97% when compared to a single technologist interpretation to 100% when compared to a consensus reached by a group of technologists. The study also identified variability in image interpretation by trained technologists and proposed that deep learning can enhance patient care by identifying annotation variability and improving personnel training.

CLINICAL CHEMISTRY (2022)

Add to Collection

Article Infectious Diseases

Efficient automated semi-quantitative urine culture analysis via BD Urine Culture App

Yoshifumi Uwamino et al.

Summary: The study evaluated the clinical utility of BD Kiestra (TM) Urine Culture App (UCA) and found high concordance rates between results obtained by medical technologists and using UCA, indicating potential efficiency improvement in obtaining semi-quantitative urine culture results.

DIAGNOSTIC MICROBIOLOGY AND INFECTIOUS DISEASE (2022)

Add to Collection

Article Microbiology

Digital Image Analysis for the Detection of Group B Streptococcus from ChromID Strepto B Medium Using PhenoMatrix Algorithms

Justin Baker et al.

Summary: This study evaluated the clinical performance of PhenoMatrix (PM) chromogenic detection module (CDM) digital imaging software in detecting Group B Streptococcus (GBS), which is the leading infectious cause of early neonatal morbidity and mortality. Compared to traditional methods like manual culture review and molecular detection, the combination of ChromID with PM CDM showed similar sensitivity to molecular detection and identified additional true positive specimens missed by manual reading. Overall, nucleic acid amplification testing (NAAT) had the highest sensitivity in GBS detection, with ChromID plus PM CDM showing comparable sensitivity and ability to detect missed true positive specimens.

JOURNAL OF CLINICAL MICROBIOLOGY (2021)

Add to Collection

Article Microbiology

Comparison of an Automated Plate Assessment System (APAS Independence) and Artificial Intelligence (AI) to Manual Plate Reading of Methicillin-Resistant and Methicillin-Susceptible Staphylococcus aureus CHROMagar Surveillance Cultures

Natalie Gammel et al.

Summary: The study demonstrated that the automated plate assessment system is accurate in triaging MRSA and S. aureus cultures and showed high sensitivity compared to human interpretation. The APAS can reliably identify positive and negative cultures, improving laboratory efficiency.

JOURNAL OF CLINICAL MICROBIOLOGY (2021)

Add to Collection

Editorial Material Microbiology