Direct antimicrobial resistance prediction from clinical MALDI-TOF mass spectra using machine learning

End-To-End Deep Learning Explains Antimicrobial Resistance in Peak-Picking-Free MALDI-MS Data

Mass spectrometry is used to determine infectious microbial species in thousands of clinical laboratories across the world. The vast amount of data allows modern data analysis methods that harvest more information and potentially answer new questions. Here, we present an end-to-end deep learning model for predicting antibiotic resistance using raw matrix assisted laser desorption ionization mass spectrometry (MALDI-MS) data. We used a 1-dimensional convolutional neural network to model (almost) raw data, skipping conventional peak-picking and directly predict resistance. The model's performance is state-of-the-art, having AUCs between 0.93 and 0.99 in all antimicrobial resistance phenotypes and validates across time and location. Feature attribution values highlight important insights into the model and how the end-to-end workflow can be improved further. This study showcases that reliable resistance phenotyping using MALDI-MS data is attainable and highlights the gains of using end-to-end deep learning for spectrometry data.

Pre-trained Maldi Transformers improve MALDI-TOF MS-based prediction

For the last decade, matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) has been the reference method for species identification in clinical microbiology. Hampered by a historical lack of open data, machine learning research towards models specifically adapted to MALDI-TOF MS remains in its infancy. Given the growing complexity of available datasets (such as large-scale antimicrobial resistance prediction), a need for models that (1) are specifically designed for MALDI-TOF MS data, and (2) have high representational capacity, presents itself. Here, we introduce Maldi Transformer, an adaptation of the state-of-the-art transformer architecture to the MALDI-TOF mass spectral domain. We propose the first self-supervised pre-training technique specifically designed for mass spectra. The technique is based on shuffling peaks across spectra, and pre-training the transformer as a peak discriminator. Extensive benchmarks confirm the efficacy of this novel design. The final result is a model exhibiting state-of-the-art (or competitive) performance on downstream prediction tasks. In addition, we show that Maldi Transformer's identification of noisy spectra may be leveraged towards higher predictive performance. All code supporting this study is distributed on PyPI and is packaged under: https://github.com/gdewael/maldi-nn.

Magnetic relaxation switch biosensor for detection of Vibrio parahaemolyticus based on photocleavable hydrogel

BACKGROUND

Foodborne pathogens, particularly Vibrio parahaemolyticus (VP) found in seafood, pose significant health risks, including abdominal pain, nausea, and even death. Rapid, accurate, and sensitive detection of these pathogens is crucial for food safety and public health. However, existing detection methods often require complex sample pretreatment, which limits their practical application. This study aims to overcome these limitations by developing a label-free magnetic relaxation switch (MRS) biosensor for the detection of VP, utilizing a photocleavable sol-gel phase transition system for improved efficiency and accuracy.

RESULTS

In this work, a tag-free magnetic relaxation switch (MRS) biosensor was designed for the detection of Vibrio parahaemolyticus (VP), based on a photocleavable sol-gel phase transition system. A large amount of lithium acyl hypophosphite (LAP), gold nanoparticles (AuNPs), and single-stranded DNA (ssDNA) loaded on the surface of Ti3C2Tx MXene acted as the signal unit LAP-MXene@AuNPs-ssDNA. The pipette tip served as a reaction vessel, and when VP was present, Apt specifically captured VP and released the signal units. The released signal units were then injected into the low-field nuclear magnetic resonance (LF-NMR) test solution, a gel formed by crosslinking of disulfide bonds. The gel was cleaved by LAPs on the signal units under ultraviolet (UV) irradiation, triggering a gel-sol phase transition, which increased transverse relaxation time (T2), thus enabling the detection of VP. Under the optimal experimental conditions, the linear range and detection limit for VP were 102 ∼ 108 CFU/mL and 10 CFU/mL, respectively.

SIGNIFICANCE AND NOVELTY

The simplified biometric identification process in the pipette tip reduces errors from multiple sample transfers, enhancing efficiency. The use of photocleavable hydrogel for signal output eliminates issues associated with magnetic material aggregation, significantly improving detection precision. The assay is of good selectivity, stability reproducibility, and convenience, having a broad application prospect in the rapid detection of pathogenic bacteria in the field.

Multidrug-resistant Gram-negative bacterial infections

Multidrug-resistant Gram-negative bacterial infections cause significant morbidity and mortality globally. These pathogens easily acquire antimicrobial resistance (AMR), further highlighting their clinical significance. Third-generation cephalosporin-resistant and carbapenem-resistant Enterobacterales (eg, Escherichia coli and Klebsiella spp), multidrug-resistant Pseudomonas aeruginosa, and carbapenem-resistant Acinetobacter baumannii are the most problematic and have been identified as priority pathogens. In response, several new diagnostic technologies aimed at rapidly detecting AMR have been developed, including biochemical, molecular, genomic, and proteomic techniques. The last decade has also seen the licensing of multiple antibiotics that have changed the treatment landscape for these challenging infections.

Challenges and applications of artificial intelligence in infectious diseases and antimicrobial resistance

Artificial intelligence (AI) has transformed infectious disease control, enhancing rapid diagnosis and antibiotic discovery. While conventional tests delay diagnosis, AI-driven methods like machine learning and deep learning assist in pathogen detection, resistance prediction, and drug discovery. These tools improve antibiotic stewardship and identify effective compounds such as antimicrobial peptides and small molecules. This review explores AI applications in diagnostics, therapy, and drug discovery, emphasizing both strengths and areas needing improvement.

Integrating MALDI-TOF Mass Spectrometry with Machine Learning Techniques for Rapid Antimicrobial Resistance Screening of Foodborne Bacterial Pathogens

Although MALDI-TOF mass spectrometry (MS) is considered as the gold standard for rapid and cost-effective identification of microorganisms in routine laboratory practices, its capability for antimicrobial resistance (AMR) detection has received limited focus. Nevertheless, recent studies explored the predictive performance of MALDI-TOF MS for detecting AMR in clinical pathogens when machine learning techniques are applied. This chapter describes a routine MALDI-TOF MS workflow for the rapid screening of AMR in foodborne pathogens, with Campylobacter spp. as a study model.

Integration of MALDI-TOF MS and machine learning to classify enterococci: A comparative analysis of supervised learning algorithms for species prediction

This research focused on distinguishing distinct matrix assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) spectral signatures of three Enterococcus species. We evaluated and compared the predictive performance of four supervised machine learning algorithms, K-nearest neighbor (KNN), support vector machine (SVM), and random forest (RF), to accurately classify Enterococcus species. This study involved a comprehensive dataset of 410 strains, generating 1640 individual spectra through on-plate and off-plate protein extraction methods. Although the commercial database correctly identified 76.9% of the strains, machine learning classifiers demonstrated superior performance (accuracy 0.991). In the RF model, top informative peaks played a significant role in the classification. Whole-genome sequencing showed that the most informative peaks are biomarkers connected to proteins, which are essential for understanding bacterial classification and evolution. The integration of MALDI-TOF MS and machine learning provides a rapid and accurate method for identifying Enterococcus species, improving healthcare and food safety.

Multi-label classification to predict antibiotic resistance from raw clinical MALDI-TOF mass spectrometry data

Antimicrobial resistance (AMR) poses a significant global health challenge, necessitating advanced predictive models to support clinical decision-making. In this study, we explore multi-label classification as a novel approach to predict antibiotic resistance across four clinically relevant bacteria: E. coli, S. aureus, K. pneumoniae, and P. aeruginosa. Using multiple datasets from the DRIAMS repository, we evaluated the performance of four algorithms - Multi-Layer Perceptron, Support Vector Classifier, Random Forest, and Extreme Gradient Boosting - under both single-label and multi-label frameworks. Our results demonstrate that the multi-label approach delivers competitive performance compared to traditional single-label models, with no statistically significant differences in most cases. The multi-label framework naturally captures the complex, interconnected nature of AMR data, reflecting real-world scenarios more accurately. We further validated the models on external datasets (DRIAMS B and C), confirming their generalizability and robustness. Additionally, we investigated the impact of oversampling techniques and provided a reproducible methodology for handling MALDI-TOF data, ensuring scalability for future studies. These findings underscore the potential of multi-label classification to enhance predictive accuracy in AMR research, offering valuable insights for developing diagnostic tools and guiding clinical interventions.

Early detection of bacterial pneumonia by characteristic induced odor signatures

INTRODUCTION

The ability to detect pathogenic bacteria before the onsets of severe respiratory symptoms and to differentiate bacterial infection allows to improve patient-tailored treatment leading to a significant reduction in illness severity, comorbidity as well as antibiotic resistance. As such, this study refines the application of the non-invasive Secondary Electrospray Ionization-High Resolution Mass Spectrometry (SESI-HRMS) methodology for real-time and early detection of human respiratory bacterial pathogens in the respiratory tract of a mouse infection model.

METHODS

A real-time analysis of changes in volatile metabolites excreted by mice undergoing a lung infection by Staphylococcus aureus or Streptococcus pneumoniae were evaluated using a SESI-HRMS instrument. The infection status was confirmed using classical CFU enumeration and tissue histology. The detected VOCs were analyzed using a pre- and post-processing algorithm along with ANOVA and RASCA statistical evaluation methods.

RESULTS

Characteristic changes in the VOCs emitted from the mice were detected as early as 4-6 h post-inoculation. Additionally, by using each mouse as its own baseline, we mimicked the inherent variation within biological organism and reported significant variations in 25 volatile organic compounds (VOCs) during the course of a lung bacterial infection.

CONCLUSION

the non-invasive SESI-HRMS enables real-time detection of infection specific VOCs. However, further refinement of this technology is necessary to improve clinical patient management, treatment, and facilitate decisions regarding antibiotic use due to early infection detection.

Machine learning algorithm for predict the in-hospital mortality in critically ill patients with congestive heart failure combined with chronic kidney disease

BACKGROUND

The objective of this study was to develop and validate a machine learning (ML) model for predict in-hospital mortality among critically ill patients with congestive heart failure (CHF) combined with chronic kidney disease (CKD).

METHODS

After employing least absolute shrinkage and selection operator regression for feature selection, six distinct methodologies were employed in the construction of the model. The selection of the optimal model was based on the area under the curve (AUC). Furthermore, the interpretation of the chosen model was facilitated through the utilization of SHapley Additive exPlanation (SHAP) values and the Local Interpretable Model-Agnostic Explanations (LIME) algorithm.

RESULTS

This study collected data and enrolled 5041 patients on CHF combined with CKD from 2008 to 2019, utilizing the Medical Information Mart for Intensive Care Unit. After selection, 22 of the 47 variables collected post-intensive care unit admission were identified as mortality-associated and subsequently utilized in the development of ML models. Among the six models generated, the eXtreme Gradient Boosting (XGBoost) model demonstrated the highest AUC at 0.837. Notably, the SHAP values highlighted the sequential organ failure assessment score, age, simplified acute physiology score II, and urine output as the four most influential variables in the XGBoost model. In addition, the LIME algorithm explains the individualized predictions.

CONCLUSIONS

In conclusion, our study accomplished the successful development and validation of ML models for predicting in-hospital mortality in critically ill patients with CHF combined with CKD. Notably, the XGBoost model emerged as the most efficacious among all the ML models employed.

An antimicrobial drug recommender system using MALDI-TOF MS and dual-branch neural networks

Timely and effective use of antimicrobial drugs can improve patient outcomes, as well as help safeguard against resistance development. Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) is currently routinely used in clinical diagnostics for rapid species identification. Mining additional data from said spectra in the form of antimicrobial resistance (AMR) profiles is, therefore, highly promising. Such AMR profiles could serve as a drop-in solution for drastically improving treatment efficiency, effectiveness, and costs. This study endeavors to develop the first machine learning models capable of predicting AMR profiles for the whole repertoire of species and drugs encountered in clinical microbiology. The resulting models can be interpreted as drug recommender systems for infectious diseases. We find that our dual-branch method delivers considerably higher performance compared to previous approaches. In addition, experiments show that the models can be efficiently fine-tuned to data from other clinical laboratories. MALDI-TOF-based AMR recommender systems can, hence, greatly extend the value of MALDI-TOF MS for clinical diagnostics. All code supporting this study is distributed on PyPI and is packaged at https://github.com/gdewael/maldi-nn.

Pioneering Klebsiella Pneumoniae Antibiotic Resistance Prediction With Artificial Intelligence-Clinical Decision Support System-Enhanced Matrix-Assisted Laser Desorption/Ionization Time-of-Flight Mass Spectrometry: Retrospective Study

BACKGROUND

The rising prevalence and swift spread of multidrug-resistant gram-negative bacteria (MDR-GNB), especially Klebsiella pneumoniae (KP), present a critical global health threat highlighted by the World Health Organization, with mortality rates soaring approximately 50% with inappropriate antimicrobial treatment.

OBJECTIVE

This study aims to advance a novel strategy to develop an artificial intelligence-clinical decision support system (AI-CDSS) that combines machine learning (ML) with matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS), aiming to significantly improve the accuracy and speed of diagnosing antibiotic resistance, directly addressing the grave health risks posed by the widespread dissemination of pan drug-resistant gram-negative bacteria across numerous countries.

METHODS

A comprehensive dataset comprising 165,299 bacterial specimens and 11,996 KP isolates was meticulously analyzed using MALDI-TOF MS technology. Advanced ML algorithms were harnessed to sculpt predictive models that ascertain resistance to quintessential antibiotics, particularly levofloxacin and ciprofloxacin, by using the amassed spectral data.

RESULTS

Our ML models revealed remarkable proficiency in forecasting antibiotic resistance, with the random forest classifier emerging as particularly effective in predicting resistance to both levofloxacin and ciprofloxacin, achieving the highest area under the curve of 0.95. Performance metrics across different models, including accuracy, sensitivity, specificity, positive predictive value, negative predictive value, and F1-score, were detailed, underlining the potential of these algorithms in aiding the development of precision treatment strategies.

CONCLUSIONS

This investigation highlights the synergy between MALDI-TOF MS and ML as a beacon of hope against the escalating threat of antibiotic resistance. The advent of AI-CDSS heralds a new era in clinical diagnostics, promising a future in which rapid and accurate resistance prediction becomes a cornerstone in combating infectious diseases. Through this innovative approach, we answered the challenge posed by KP and other multidrug-resistant pathogens, marking a significant milestone in our journey toward global health security.

Hypervirulent and carbapenem-resistant Klebsiella pneumoniae: A global public health threat

The evolution of hypervirulent and carbapenem-resistant Klebsiella pneumoniae can be categorized into three main patterns: the evolution of KL1/KL2-hvKp strains into CR-hvKp, the evolution of carbapenem-resistant K. pneumoniae (CRKp) strains into hv-CRKp, and the acquisition of hybrid plasmids carrying carbapenem resistance and virulence genes by classical K. pneumoniae (cKp). These strains are characterized by multi-drug resistance, high virulence, and high infectivity. Currently, there are no effective methods for treating and surveillance this pathogen. In addition, the continuous horizontal transfer and clonal spread of these bacteria under the pressure of hospital antibiotics have led to the emergence of more drug-resistant strains. This review discusses the evolution and distribution characteristics of hypervirulent and carbapenem-resistant K. pneumoniae, the mechanisms of carbapenem resistance and hypervirulence, risk factors for susceptibility, infection syndromes, treatment regimens, real-time surveillance and preventive control measures. It also outlines the resistance mechanisms of antimicrobial drugs used to treat this pathogen, providing insights for developing new drugs, combination therapies, and a "One Health" approach. Narrowing the scope of surveillance but intensifying implementation efforts is a viable solution. Monitoring of strains can be focused primarily on hospitals and urban wastewater treatment plants.

Accurate prediction of antimicrobial resistance and genetic marker of Staphylococcus aureus clinical isolates using MALDI-TOF MS and machine learning - across DRIAMS and Taiwan database

BACKGROUND

The use of matrix-assisted laser desorption/ionisation-time-of-flight mass spectra (MALDI-TOF MS) with machine learning (ML) has been explored for predicting antimicrobial resistance. This study evaluates the effectiveness of MALDI-TOF MS paired with various ML classifiers and establishes optimal models for predicting antimicrobial resistance and the presence of mecA gene among Staphylococcus aureus.

MATERIALS AND METHODS

Antimicrobial resistance against tier 1 antibiotics and MALDI-TOF MS of S. aureus were analysed using data from the Database of Resistance against Antimicrobials with MALDI-TOF Mass Spectrometry (DRIAMS) and one medical centre (CS database). Five ML classifiers were used to analyse performance metrics. The Shapley value quantified the predictive contribution of individual features.

RESULTS

The LightGBM demonstrated superior performance in predicting antimicrobial resistance for most tier 1 antibiotics among oxacillin-resistant S. aureus (ORSA) compared with all S. aureus and oxacillin-susceptible S. aureus (OSSA) in both databases. In DRIAMS, Multilayer Perceptron (MLP) was associated with excellent predictive performance, expressed as accuracy/AUROC/AUPR, for clindamycin (0.74/0.81/0.90), tetracycline (0.86/0.87/0.94), and trimethoprim-sulfamethoxazole (0.95/0.72/0.97). In the CS database, Ada and Light Gradient Boosting Machine (LightGBM) showed excellent performance for erythromycin (0.97/0.92/0.86) and tetracycline (0.68/0.79/0.86). Mass-to-charge ratio (m/z) features of 2411-2414 and 2429-2432 correlated with clindamycin resistance, whereas 5033-5036 was linked to erythromycin resistance in DRIAMS. In the CS database, overlapping features of 2423-2426, 4496-4499, and 3764-3767 simultaneously predicted the presence of mecA and oxacillin resistance.

CONCLUSION

The predictive performance of antimicrobial resistance against S. aureus using MALDI-TOF MS depends on database characteristics and the ML algorithm selected. Specific and overlapping mass spectra features are excellent predictive markers for mecA and specific antimicrobial resistance.

A new fusion strategy for rapid strain differentiation based on MALDI-TOF MS and Raman spectra

Typing of bacterial subspecies is urgently needed for the diagnosis and efficient treatment during disease outbreaks. Physicochemical spectroscopy can provide a rapid analysis but its identification accuracy is still far from satisfactory. Herein, a novel feature-extractor-based fusion-assisted machine learning strategy has been developed for high accuracy and rapid strain differentiation using matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) and Raman spectroscopy. Based on this fusion approach, rapid and reliable identification and analysis can be performed within 24 hours. Validation on a panel of important pathogens comprising Staphylococcus aureus, Klebsiella pneumoniae, Escherichia coli, and Acinetobacter baumannii showed that the identification accuracies of k-nearest neighbors (KNNs), support vector machines (SVMs) and artificial neural networks (ANNs) were 100%. In particular, when benchmarked against a MALDI-TOF MS spectral dataset, the new approach improved the identification accuracy of Acinetobacter baumannii from 87.67% to 100%. This work demonstrates the effectiveness of combining MALDI-TOF MS and Raman spectroscopy fusion data in pathogenic bacterial subtyping.

Repurposing MALDI-TOF MS for effective antibiotic resistance screening in Staphylococcus epidermidis using machine learning

The emergence of Staphylococcus epidermidis as a significant nosocomial pathogen necessitates advancements in more efficient antimicrobial resistance profiling. However, existing culture-based and PCR-based antimicrobial susceptibility testing methods are far too slow or costly. This study combines machine learning with matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) to develop predictive models for various antibiotics using a comprehensive dataset containing thousands of S. epidermidis isolates. Optimized machine learning models utilized feature selection and achieved high AUROC scores ranging from 0.80 to 0.95 while maintaining AUPRC scores up to 0.97. Shapley Additive exPlanations were employed to analyze relevant features and assess the significance of corresponding protein biomarkers while also verifying that predictive power was derived from the detection of proteins rather than noise. Antimicrobial resistance models were validated externally to evaluate model performance outside the original data collection site. The approaches and findings in this study demonstrate a significant advancement in rapid, cost-effective antimicrobial resistance profiling, offering a promising solution for improving treatments for nosocomial infections and being potentially applicable to other microbial pathogens in the future.

Machine learning and matrix-assisted laser desorption/ionization time-of-flight mass spectra for antimicrobial resistance prediction: A systematic review of recent advancements and future development

BACKGROUND

The use of matrix-assisted laser desorption/ionization time-of-flight mass spectra (MALDI-TOF MS) combined with machine learning techniques has recently emerged as a method to address the public health crisis of antimicrobial resistance. This systematic review, conducted following Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) guidelines, aims to evaluate the current state of the art in using machine learning for the detection and classification of antimicrobial resistance from MALDI-TOF mass spectrometry data.

METHODS

A comprehensive review of the literature on machine learning applications for antimicrobial resistance detection was performed using databases such as Web Of Science, Scopus, ScienceDirect, IEEE Xplore, and PubMed. Only original articles in English were included. Studies applying machine learning without using MALDI-TOF mass spectra were excluded.

RESULTS

Forty studies met the inclusion criteria. Staphylococcus aureus, Klebsiella pneumoniae and Escherichia coli were the most frequently cited bacteria. The antibiotics resistance most studied corresponds to methicillin for S. aureus, cephalosporins for K. pneumoniae, and aminoglycosides for E. coli. Random forest, support vector machine and logistic regression were the most employed algorithms to predict antimicrobial resistance. Additionally, seven studies reported using artificial neural networks. Most studies reported metrics such as accuracy, sensitivity, specificity, and the area under the receiver operating characteristic (AUROC) above 0.80.

CONCLUSIONS

Our study indicates that random forest, support vector machine, and logistic regression are effective for predicting antimicrobial resistance using MALDI-TOF MS data. Recent studies also highlight the potential of deep learning techniques in this area. We recommend further exploration of deep learning and multi-label supervised learning for comprehensive antibiotic resistance prediction in clinical practice.

Classification of Latilactobacillus sakei subspecies based on MALDI-TOF MS protein profiles using machine learning models

Latilactobacillus sakei is an important bacterial species used as a starter culture for fermented foods; however, two subspecies within this species exhibit different properties in the foods. Matrix-assisted laser desorption/ionization-time of flight mass spectrometer (MALDI-TOF MS) is the gold standard for microbial fingerprinting. However, the resolution power is down to the species level. This study was to combine MALDI-TOF mass spectra and machine learning to develop a new method to identify two L. sakei subspecies (L. sakei subsp. sakei and L. sakei subsp. carnosus) and non-L. sakei species. Totally, 227 strains were collected, with 908 spectra obtained via on- and off-plate protein extraction. Only 68.7% of strains were correctly identified at the subspecies level in the Biotyper database; however, a high level of performance was observed from the machine learning models. Partial least squares-discriminant analysis (PLS-DA), principal component analysis-K-nearest neighbor (PCA-KNN), and support vector machine (SVM) demonstrated 0.823, 0.914, and 0.903 accuracies, respectively, whereas the random forest (RF) achieved an accuracy of 0.954, with an area under the receiver operating characteristic (AUROC) curve of 0.99, outperforming the other algorithms in distinguishing the subspecies. The machine learning proved to be a promising technique for the rapid and high-resolution classification of L. sakei subspecies using MALDI-TOF MS.

IMPORTANCE

Latilactobacillus sakei plays a significant role in the realm of food bacteria. One particular subspecies of L. sakei is employed as a protective agent during food fermentation, whereas another strain is responsible for food spoilage. Hence, it is crucial to precisely differentiate between the two subspecies of L. sakei. In this study, machine learning models based on protein mass peaks were developed for the first time to distinguish L. sakei subspecies. Furthermore, the efficacy of three commonly used machine learning algorithms for microbial classification was evaluated. Our results provide the foundation for future research on developing machine learning models for the classification of microbial species or subspecies. In addition, the developed model can be used in the food industry to monitor L. sakei subspecies in fermented foods in a time- and cost-effective method for food quality and safety.

Application of machine learning based genome sequence analysis in pathogen identification

Infectious diseases caused by pathogenic microorganisms pose a serious threat to human health. Despite advances in molecular biology, genetics, computation, and medicinal chemistry, infectious diseases remain a significant public health concern. Addressing the challenges posed by pathogen outbreaks, pandemics, and antimicrobial resistance requires concerted interdisciplinary efforts. With the development of computer technology and the continuous exploration of artificial intelligence(AI)applications in the biomedical field, the automatic morphological recognition and image processing of microbial images under microscopes have advanced rapidly. The research team of Institute of Microbiology, Chinese Academy of Sciences has developed a single cell microbial identification technology combining Raman spectroscopy and artificial intelligence. Through laser Raman acquisition system and convolutional neural network analysis, the average accuracy rate of 95.64% has been achieved, and the identification can be completed in only 5 min. These technologies have shown substantial advantages in the visible morphological detection of pathogenic microorganisms, expanding anti-infective drug discovery, enhancing our understanding of infection biology, and accelerating the development of diagnostics. In this review, we discuss the application of AI-based machine learning in image analysis, genome sequencing data analysis, and natural language processing (NLP) for pathogen identification, highlighting the significant role of artificial intelligence in pathogen diagnosis. AI can improve the accuracy and efficiency of diagnosis, promote early detection and personalized treatment, and enhance public health safety.

Identification methods as a factor affecting the performance of clinical microbiology laboratories participating in an external quality assessment program: a cross-sectional, retrospective analysis

Introduction. Laboratory participation in external quality assessment (EQA) programmes including proficiency testing (PT) is a requirement of clinical laboratory conformance to ISO 15189:2022 Medical laboratories - Requirements for quality and competence. PT is one EQA method whereby laboratories are sent blinded samples for characterization by routine laboratory diagnostic methods. Importantly, PT enables a laboratory's performance to be evaluated in comparison to the standard reference methods and to the performance of other peer laboratories using similar diagnostic methods.Gap statement. The desired outcome of participating in PT is to help laboratories identify possible sources of error in each step of the total testing process and particularly in their test methods during the analytical phase.Aim. This cross-sectional study investigated the impact of using matrix-assisted laser desorption ionization time-of-flight mass spectrometry (MALDI-TOF MS) compared to conventional phenotypic biochemical testing on laboratory performance in a clinical bacteriology PT scheme.Methodology. During a 6-year period from 2017-2022, the Canadian Microbiology Proficiency Testing implemented 112 PT challenges comprising 22 different sample types and included 61 different bacterial species. This was translated into 5883 graded test events for analysis. Multiple logistic regression techniques were employed to explore the association between the test method employed and laboratory performance. The sample type and aerobic classification of challenge organisms were included as confounding variables.Results. Laboratories using MALDI-TOF MS performed significantly better in characterizing microorganisms than laboratories using phenotypic biochemical testing alone [odds ratio OR = 5.68, confidence interval (CI): 3.92, 8.22] regardless of the sample type and aerobic classification. Notably, our analysis identified a significant association between anaerobic organisms and laboratory performance (OR: 0.24, CI: 0.17-0.35), suggesting that culturing and identifying fastidious organisms remains a significant obstacle for many clinical microbiology laboratories.Conclusions. Although no method is infallible and its performance will depend on the validation and quality assurance procedures, this finding may help the management in the decision for implementing MALDI-TOF MS in the microbiology laboratory. This study highlights the important role PT providers play in the objective assessment of laboratory performance and how it can provide evidence for quality improvement.

Healthcare as a driver, reservoir and amplifier of antimicrobial resistance: opportunities for interventions

Antimicrobial resistance (AMR) is a global health challenge that threatens humans, animals and the environment. Evidence is emerging for a role of healthcare infrastructure, environments and patient pathways in promoting and maintaining AMR via direct and indirect mechanisms. Advances in vaccination and monoclonal antibody therapies together with integrated surveillance, rapid diagnostics, targeted antimicrobial therapy and infection control measures offer opportunities to address healthcare-associated AMR risks more effectively. Additionally, innovations in artificial intelligence, data linkage and intelligent systems can be used to better predict and reduce AMR and improve healthcare resilience. In this Review, we examine the mechanisms by which healthcare functions as a driver, reservoir and amplifier of AMR, contextualized within a One Health framework. We also explore the opportunities and innovative solutions that can be used to combat AMR throughout the patient journey. We provide a perspective on the current evidence for the effectiveness of interventions designed to mitigate healthcare-associated AMR and promote healthcare resilience within high-income and resource-limited settings, as well as the challenges associated with their implementation.

Artificial Intelligence in Infectious Disease Clinical Practice: An Overview of Gaps, Opportunities, and Limitations

The integration of artificial intelligence (AI) in clinical medicine marks a revolutionary shift, enhancing diagnostic accuracy, therapeutic efficacy, and overall healthcare delivery. This review explores the current uses, benefits, limitations, and future applications of AI in infectious diseases, highlighting its specific applications in diagnostics, clinical decision making, and personalized medicine. The transformative potential of AI in infectious diseases is emphasized, addressing gaps in rapid and accurate disease diagnosis, surveillance, outbreak detection and management, and treatment optimization. Despite these advancements, significant limitations and challenges exist, including data privacy concerns, potential biases, and ethical dilemmas. The article underscores the need for stringent regulatory frameworks and inclusive databases to ensure equitable, ethical, and effective AI utilization in the field of clinical and laboratory infectious diseases.

Predicting Pseudomonas aeruginosa drug resistance using artificial intelligence and clinical MALDI-TOF mass spectra

Matrix-assisted laser desorption/ionization-time of flight mass spectrometry (MALDI-TOF MS) is widely used in clinical microbiology laboratories for bacterial identification but its use for detection of antimicrobial resistance (AMR) remains limited. Here, we used MALDI-TOF MS with artificial intelligence (AI) approaches to successfully predict AMR in Pseudomonas aeruginosa, a priority pathogen with complex AMR mechanisms. The highest performance was achieved for modern β-lactam/β-lactamase inhibitor drugs, namely, ceftazidime/avibactam and ceftolozane/tazobactam. For these drugs, the model demonstrated area under the receiver operating characteristic curve (AUROC) of 0.869 and 0.856, specificity of 0.925 and 0.897, and sensitivity of 0.731 and 0.714, respectively. As part of this work, we developed dynamic binning, a feature engineering technique that effectively reduces the high-dimensional feature set and has wide-ranging applicability to MALDI-TOF MS data. Compared to conventional feature engineering approaches, the dynamic binning method yielded highest performance in 7 of 10 antimicrobials. Moreover, we showcased the efficacy of transfer learning in enhancing the AUROC performance for 8 of 11 antimicrobials. By assessing the contribution of features to the model's prediction, we identified proteins that may contribute to AMR mechanisms. Our findings demonstrate the potential of combining AI with MALDI-TOF MS as a rapid AMR diagnostic tool for Pseudomonas aeruginosa.IMPORTANCEPseudomonas aeruginosa is a key bacterial pathogen that causes significant global morbidity and mortality. Antimicrobial resistance (AMR) emerges rapidly in P. aeruginosa and is driven by complex mechanisms. Drug-resistant P. aeruginosa is a major challenge in clinical settings due to limited treatment options. Early detection of AMR can guide antibiotic choices, improve patient outcomes, and avoid unnecessary antibiotic use. Matrix-assisted laser desorption/ionization-time of flight mass spectrometry (MALDI-TOF MS) is widely used for rapid species identification in clinical microbiology. In this study, we repurposed mass spectra generated by MALDI-TOF and used them as inputs for artificial intelligence approaches to successfully predict AMR in P. aeruginosa for multiple key antibiotic classes. This work represents an important advance toward using MALDI-TOF as a rapid AMR diagnostic for P. aeruginosa in clinical settings.

Applying Machine Learning for Antibiotic Development and Prediction of Microbial Resistance

Antimicrobial resistance (AMR) poses a serious threat to human health worldwide. It is now more challenging than ever to introduce a potent antibiotic to the market considering rapid emergence of antimicrobial resistance, surpassing the rate of antibiotic drug discovery. Hence, new approaches need to be developed to accelerate the rate of drug discovery process and meet the demands for new antibiotics, while reducing the cost of their development. Machine learning holds immense promise of becoming a useful tool, especially since in the last two decades, exponential growth has occurred in computational power and biological big data analytics. Recent advancements in machine learning algorithms for drug discovery have provided significant clues for potential antibiotic classes. Apart from discovery of new scaffolds, the machine learning protocols will significantly impact prediction of AMR patterns and drug metabolism. In this review, we outline power of machine learning in antibiotic drug discovery, metabolic fate, and AMR prediction to support researchers engaged and interested in this field.

Resolution of peri-implant mucositis following standard treatment: A prospective split-mouth study

BACKGROUND

Peri-implant mucositis (PIM) is a pathological precursor of peri-implantitis, but its pattern of conversion to peri-implantitis is unclear and complicated to diagnose clinically, while none of the available protocols yield complete disease resolution. The aim of this study was the evaluation of PIM responsiveness to standard anti-infective mechanical treatment (AIMT) at clinical and biomarker levels, and estimation of the diagnostic capacity of bone markers as surrogate endpoints and predictors.

METHODS

Systemically healthy outpatients presenting one implant exhibiting clinical signs of inflammation confined within the soft tissue (PIM) and one healthy control (HC) implant at a non-adjacent position were included. Clinical parameters and peri-implant crevicular fluid samples were collected baseline and 6 months following mechanical therapy, to assess the levels of RANKL, OPG, and IGFBP2. PIM clustering was performed using machine learning algorithms.

RESULTS

Overall, 38 patients met the inclusion criteria. Therapy resulted in the reduction of all clinical and biological indicators, but respective values remained significantly higher compared to HC. Clinical examination noted 30% disease resolution at the 6-month follow-up, while 43% showed no active bone resorption. OPG showed positive prognostic value for treatment outcome, while the clustering based on active bone resorption did not differ in terms of therapeutic effectiveness.

CONCLUSION

AIMT is effective in reducing the clinical and biological indicators of PIM, but complete clinical resolution was achieved in only 30% of the cases. Around one third of PIM patients exhibited active bone resorption bellow clinical detectability that was not associated with disease progression and poor treatment responsiveness.

Machine Learning Assisted MALDI Mass Spectrometry for Rapid Antimicrobial Resistance Prediction in Clinicals

Antimicrobial susceptibility testing (AST) plays a critical role in assessing the resistance of individual microbial isolates and determining appropriate antimicrobial therapeutics in a timely manner. However, conventional AST normally takes up to 72 h for obtaining the results. In healthcare facilities, the global distribution of vancomycin-resistant Enterococcus fecium (VRE) infections underscores the importance of rapidly determining VRE isolates. Here, we developed an integrated antimicrobial resistance (AMR) screening strategy by combining matrix-assisted laser desorption ionization mass spectrometry (MALDI-MS) with machine learning to rapidly predict VRE from clinical samples. Over 400 VRE and vancomycin-susceptible E. faecium (VSE) isolates were analyzed using MALDI-MS at different culture times, and a comprehensive dataset comprising 2388 mass spectra was generated. Algorithms including the support vector machine (SVM), SVM with L1-norm, logistic regression, and multilayer perceptron (MLP) were utilized to train the classification model. Validation on a panel of clinical samples (external patients) resulted in a prediction accuracy of 78.07%, 80.26%, 78.95%, and 80.54% for each algorithm, respectively, all with an AUROC above 0.80. Furthermore, a total of 33 mass regions were recognized as influential features and elucidated, contributing to the differences between VRE and VSE through the Shapley value and accuracy, while tandem mass spectrometry was employed to identify the specific peaks among them. Certain ribosomal proteins, such as A0A133N352 and R2Q455, were tentatively identified. Overall, the integration of machine learning with MALDI-MS has enabled the rapid determination of bacterial antibiotic resistance, greatly expediting the usage of appropriate antibiotics.

Navigating the future: machine learning's role in revolutionizing antimicrobial stewardship and infection prevention and control

PURPOSE OF REVIEW

This review examines the current state and future prospects of machine learning (ML) in infection prevention and control (IPC) and antimicrobial stewardship (ASP), highlighting its potential to transform healthcare practices by enhancing the precision, efficiency, and effectiveness of interventions against infections and antimicrobial resistance.

RECENT FINDINGS

ML has shown promise in improving surveillance and detection of infections, predicting infection risk, and optimizing antimicrobial use through the development of predictive analytics, natural language processing, and personalized medicine approaches. However, challenges remain, including issues related to data quality, model interpretability, ethical considerations, and integration into clinical workflows.

SUMMARY

Despite these challenges, the future of ML in IPC and ASP is promising, with interdisciplinary collaboration identified as a key factor in overcoming existing barriers. ML's role in advancing personalized medicine, real-time disease monitoring, and effective IPC and ASP strategies signifies a pivotal shift towards safer, more efficient healthcare environments and improved patient care in the face of global antimicrobial resistance challenges.

From Data to Decisions: Leveraging Artificial Intelligence and Machine Learning in Combating Antimicrobial Resistance - a Comprehensive Review

The emergence of drug-resistant bacteria poses a significant challenge to modern medicine. In response, Artificial Intelligence (AI) and Machine Learning (ML) algorithms have emerged as powerful tools for combating antimicrobial resistance (AMR). This review aims to explore the role of AI/ML in AMR management, with a focus on identifying pathogens, understanding resistance patterns, predicting treatment outcomes, and discovering new antibiotic agents. Recent advancements in AI/ML have enabled the efficient analysis of large datasets, facilitating the reliable prediction of AMR trends and treatment responses with minimal human intervention. ML algorithms can analyze genomic data to identify genetic markers associated with antibiotic resistance, enabling the development of targeted treatment strategies. Additionally, AI/ML techniques show promise in optimizing drug administration and developing alternatives to traditional antibiotics. By analyzing patient data and clinical outcomes, these technologies can assist healthcare providers in diagnosing infections, evaluating their severity, and selecting appropriate antimicrobial therapies. While integration of AI/ML in clinical settings is still in its infancy, advancements in data quality and algorithm development suggest that widespread clinical adoption is forthcoming. In conclusion, AI/ML holds significant promise for improving AMR management and treatment outcome.

Deep-learning-enabled antibiotic discovery through molecular de-extinction

Molecular de-extinction aims at resurrecting molecules to solve antibiotic resistance and other present-day biological and biomedical problems. Here we show that deep learning can be used to mine the proteomes of all available extinct organisms for the discovery of antibiotic peptides. We trained ensembles of deep-learning models consisting of a peptide-sequence encoder coupled with neural networks for the prediction of antimicrobial activity and used it to mine 10,311,899 peptides. The models predicted 37,176 sequences with broad-spectrum antimicrobial activity, 11,035 of which were not found in extant organisms. We synthesized 69 peptides and experimentally confirmed their activity against bacterial pathogens. Most peptides killed bacteria by depolarizing their cytoplasmic membrane, contrary to known antimicrobial peptides, which tend to target the outer membrane. Notably, lead compounds (including mammuthusin-2 from the woolly mammoth, elephasin-2 from the straight-tusked elephant, hydrodamin-1 from the ancient sea cow, mylodonin-2 from the giant sloth and megalocerin-1 from the extinct giant elk) showed anti-infective activity in mice with skin abscess or thigh infections. Molecular de-extinction aided by deep learning may accelerate the discovery of therapeutic molecules.

TGC-ARG: Anticipating Antibiotic Resistance via Transformer-Based Modeling and Contrastive Learning

In various domains, including everyday activities, agricultural practices, and medical treatments, the escalating challenge of antibiotic resistance poses a significant concern. Traditional approaches to studying antibiotic resistance genes (ARGs) often require substantial time and effort and are limited in accuracy. Moreover, the decentralized nature of existing data repositories complicates comprehensive analysis of antibiotic resistance gene sequences. In this study, we introduce a novel computational framework named TGC-ARG designed to predict potential ARGs. This framework takes protein sequences as input, utilizes SCRATCH-1D for protein secondary structure prediction, and employs feature extraction techniques to derive distinctive features from both sequence and structural data. Subsequently, a Siamese network is employed to foster a contrastive learning environment, enhancing the model's ability to effectively represent the data. Finally, a multi-layer perceptron (MLP) integrates and processes sequence embeddings alongside predicted secondary structure embeddings to forecast ARG presence. To evaluate our approach, we curated a pioneering open dataset termed ARSS (Antibiotic Resistance Sequence Statistics). Comprehensive comparative experiments demonstrate that our method surpasses current state-of-the-art methodologies. Additionally, through detailed case studies, we illustrate the efficacy of our approach in predicting potential ARGs.

Biomarker identification by interpretable maximum mean discrepancy

MOTIVATION

In many biomedical applications, we are confronted with paired groups of samples, such as treated versus control. The aim is to detect discriminating features, i.e. biomarkers, based on high-dimensional (omics-) data. This problem can be phrased more generally as a two-sample problem requiring statistical significance testing to establish differences, and interpretations to identify distinguishing features. The multivariate maximum mean discrepancy (MMD) test quantifies group-level differences, whereas statistically significantly associated features are usually found by univariate feature selection. Currently, few general-purpose methods simultaneously perform multivariate feature selection and two-sample testing.

RESULTS

We introduce a sparse, interpretable, and optimized MMD test (SpInOpt-MMD) that enables two-sample testing and feature selection in the same experiment. SpInOpt-MMD is a versatile method and we demonstrate its application to a variety of synthetic and real-world data types including images, gene expression measurements, and text data. SpInOpt-MMD is effective in identifying relevant features in small sample sizes and outperforms other feature selection methods such as SHapley Additive exPlanations and univariate association analysis in several experiments.

AVAILABILITY AND IMPLEMENTATION

The code and links to our public data are available at https://github.com/BorgwardtLab/spinoptmmd.

Multidimensional mass profiles increase confidence in bacterial identification when using low-resolution mass spectrometers

Field-forward analytical technologies, such as portable mass spectrometry (MS), enable essential capabilities for real-time monitoring and point-of-care diagnostic applications. Significant and recent investments improving the features of miniaturized mass spectrometers enable various new applications outside of small molecule detection. Most notably, the addition of tandem mass spectrometry scans (MS/MS) allows the instrument to isolate and fragment ions and increase the analytical specificity by measuring unique chemical signatures for ions of interest. Notwithstanding these technological advancements, low-cost, portable systems still struggle to confidently identify clinically significant organisms of interest, such as bacteria, viruses, and proteinaceous toxins, due to the limitations in resolving power. To overcome these limitations, we developed a novel multidimensional mass fingerprinting technique that uses tandem mass spectrometry to increase the chemical specificity for low-resolution mass spectral profiles. We demonstrated the method's capabilities for differentiating four different bacteria, including attentuated strains of Yersinia pestis. This approach allowed for the accurate (>92%) identification of each organism at the strain level using de-resolved matrix-assisted laser desorption ionization-time of flight (MALDI-TOF) data to mimic the performance characteristics of miniaturized mass spectrometers. This work demonstrates that low-resolution mass spectrometers, equipped with tandem MS acquisition modes, can accurately identify clinically relevant bacteria. These findings support the future application of these technologies for field-forward and point-of-care applications where high-performance mass spectrometers would be cost-prohibitive or otherwise impractical.

Applications and advances in molecular diagnostics: revolutionizing non-tuberculous mycobacteria species and subspecies identification

Non-tuberculous mycobacteria (NTM) infections pose a significant public health challenge worldwide, affecting individuals across a wide spectrum of immune statuses. Recent epidemiological studies indicate rising incidence rates in both immunocompromised and immunocompetent populations, underscoring the need for enhanced diagnostic and therapeutic approaches. NTM infections often present with symptoms similar to those of tuberculosis, yet with less specificity, increasing the risk of misdiagnosis and potentially adverse outcomes for patients. Consequently, rapid and accurate identification of the pathogen is crucial for precise diagnosis and treatment. Traditional detection methods, notably microbiological culture, are hampered by lengthy incubation periods and a limited capacity to differentiate closely related NTM subtypes, thereby delaying diagnosis and the initiation of targeted therapies. Emerging diagnostic technologies offer new possibilities for the swift detection and accurate identification of NTM infections, playing a critical role in early diagnosis and providing more accurate and comprehensive information. This review delineates the current molecular methodologies for NTM species and subspecies identification. We critically assess the limitations and challenges inherent in these technologies for diagnosing NTM and explore potential future directions for their advancement. It aims to provide valuable insights into advancing the application of molecular diagnostic techniques in NTM infection identification.

Recent Developments in Machine Learning for Mass Spectrometry

Statistical analysis and modeling of mass spectrometry (MS) data have a long and rich history with several modern MS-based applications using statistical and chemometric methods. Recently, machine learning (ML) has experienced a renaissance due to advents in computational hardware and the development of new algorithms for artificial neural networks (ANN) and deep learning architectures. Moreover, recent successes of new ANN and deep learning architectures in several areas of science, engineering, and society have further strengthened the ML field. Importantly, modern ML methods and architectures have enabled new approaches for tasks related to MS that are now widely adopted in several popular MS-based subdisciplines, such as mass spectrometry imaging and proteomics. Herein, we aim to provide an introductory summary of the practical aspects of ML methodology relevant to MS. Additionally, we seek to provide an up-to-date review of the most recent developments in ML integration with MS-based techniques while also providing critical insights into the future direction of the field.

Contribution of MALDI-TOF mass spectrometry and machine learning including deep learning techniques for the detection of virulence factors of Clostridioides difficile strains

Clostridioides difficile (CD) infections are defined by toxins A (TcdA) and B (TcdB) along with the binary toxin (CDT). The emergence of the 'hypervirulent' (Hv) strain PR 027, along with PR 176 and 181, two decades ago, reshaped CD infection epidemiology in Europe. This study assessed MALDI-TOF mass spectrometry (MALDI-TOF MS) combined with machine learning (ML) and Deep Learning (DL) to identify toxigenic strains (producing TcdA, TcdB with or without CDT) and Hv strains. In total, 201 CD strains were analysed, comprising 151 toxigenic (24 ToxA+B+CDT+, 22 ToxA+B+CDT+ Hv+ and 105 ToxA+B+CDT-) and 50 non-toxigenic (ToxA-B-) strains. The DL-based classifier exhibited a 0.95 negative predictive value for excluding ToxA-B- strains, showcasing accuracy in identifying this strain category. Sensitivity in correctly identifying ToxA+B+CDT- strains ranged from 0.68 to 0.91. Additionally, all classifiers consistently demonstrated high specificity (>0.96) in detecting ToxA+B+CDT+ strains. The classifiers' performances for Hv strain detection were linked to high specificity (≥0.96). This study highlights MALDI-TOF MS enhanced by ML techniques as a rapid and cost-effective tool for identifying CD strain virulence factors. Our results brought a proof-of-concept concerning the ability of MALDI-TOF MS coupled with ML techniques to detect virulence factor and potentially improve the outbreak's management.

Interpretable machine learning model for predicting acute kidney injury in critically ill patients

BACKGROUND

This study aimed to create a method for promptly predicting acute kidney injury (AKI) in intensive care patients by applying interpretable, explainable artificial intelligence techniques.

METHODS

Population data regarding intensive care patients were derived from the Medical Information Mart for Intensive Care IV database from 2008 to 2019. Machine learning (ML) techniques with six methods were created to construct the predicted models for AKI. The performance of each ML model was evaluated by comparing the areas under the curve (AUC). Local Interpretable Model-Agnostic Explanations (LIME) method and Shapley Additive exPlanation values were used to decipher the best model.

RESULTS

According to inclusion and exclusion criteria, 53,150 severely sick individuals were included in the present study, of which 42,520 (80%) were assigned to the training group, and 10,630 (20%) were allocated to the validation group. Compared to the other five ML models, the eXtreme Gradient Boosting (XGBoost) model greatly predicted AKI following ICU admission, with an AUC of 0.816. The top four contributing variables of the XGBoost model were SOFA score, weight, mechanical ventilation, and the Simplified Acute Physiology Score II. An AKI and Non-AKI cases were predicted separately using the LIME algorithm.

CONCLUSION

Overall, the constructed clinical feature-based ML models are excellent in predicting AKI in intensive care patients. It would be constructive for physicians to provide early support and timely intervention measures to intensive care patients at risk of AKI.

MSDeepAMR: antimicrobial resistance prediction based on deep neural networks and transfer learning

Introduction

Antimicrobial resistance (AMR) is a global health problem that requires early and effective treatments to prevent the indiscriminate use of antimicrobial drugs and the outcome of infections. Mass Spectrometry (MS), and more particularly MALDI-TOF, have been widely adopted by routine clinical microbiology laboratories to identify bacterial species and detect AMR. The analysis of AMR with deep learning is still recent, and most models depend on filters and preprocessing techniques manually applied on spectra.

Methods

This study propose a deep neural network, MSDeepAMR, to learn from raw mass spectra to predict AMR. MSDeepAMR model was implemented for Escherichia coli, Klebsiella pneumoniae, and Staphylococcus aureus under different antibiotic resistance profiles. Additionally, a transfer learning test was performed to study the benefits of adapting the previously trained models to external data.

Results

MSDeepAMR models showed a good classification performance to detect antibiotic resistance. The AUROC of the model was above 0.83 in most cases studied, improving the results of previous investigations by over 10%. The adapted models improved the AUROC by up to 20% when compared to a model trained only with external data.

Discussion

This study demonstrate the potential of the MSDeepAMR model to predict antibiotic resistance and their use on external MS data. This allow the extrapolation of the MSDeepAMR model to de used in different laboratories that need to study AMR and do not have the capacity for an extensive sample collection.

Machine learning for microbiologists

Machine learning is increasingly important in microbiology where it is used for tasks such as predicting antibiotic resistance and associating human microbiome features with complex host diseases. The applications in microbiology are quickly expanding and the machine learning tools frequently used in basic and clinical research range from classification and regression to clustering and dimensionality reduction. In this Review, we examine the main machine learning concepts, tasks and applications that are relevant for experimental and clinical microbiologists. We provide the minimal toolbox for a microbiologist to be able to understand, interpret and use machine learning in their experimental and translational activities.

Using novel micropore technology combined with artificial intelligence to differentiate Staphylococcus aureus and Staphylococcus epidermidis

Methods for identifying bacterial pathogens are broadly categorised into conventional culture-based microbiology, nucleic acid-based tests, and mass spectrometry. The conventional method requires several days to isolate and identify bacteria. Nucleic acid-based tests and mass spectrometry are relatively rapid and reliable, but they require trained technicians. Moreover, mass spectrometry requires expensive equipment. The development of a novel, inexpensive, and simple technique for identifying bacterial pathogens is needed. Through combining micropore technology and assembly machine learning, we developed a novel classifier whose receiver operating characteristic (ROC) curve showed an area under the ROC curve of 0.94, which rapidly differentiated between Staphylococcus aureus and Staphylococcus epidermidis in this proof-of-concept study. Morphologically similar bacteria belonging to an identical genus can be distinguished using our method, which requires no specific training, and may facilitate the diagnosis and treatment of patients with bacterial infections in remote areas and in developing countries.

Benefit analysis of the auto-verification system of intelligent inspection for microorganisms

In recent years, the automatic machine for microbial identification and antibiotic susceptibility tests has been introduced into the microbiology laboratory of our hospital, but there are still many steps that need manual operation. The purpose of this study was to establish an auto-verification system for bacterial naming to improve the turnaround time (TAT) and reduce the burden on clinical laboratory technologists. After the basic interpretation of the gram staining results of microorganisms, the appearance of strain growth, etc., the 9 rules were formulated by the laboratory technologists specialized in microbiology for auto-verification of bacterial naming. The results showed that among 70,044 reports, the average pass rate of auto-verification was 68.2%, and the reason for the failure of auto-verification was further evaluated. It was found that the main causes reason the inconsistency between identification results and strain appearance rationality, the normal flora in the respiratory tract and urine that was identified, the identification limitation of the mass spectrometer, and so on. The average TAT for the preliminary report of bacterial naming was 35.2 h before, which was reduced to 31.9 h after auto-verification. In summary, after auto-verification, the laboratory could replace nearly 2/3 of manual verification and issuance of reports, reducing the daily workload of medical laboratory technologists by about 2 h. Moreover, the TAT on the preliminary identification report was reduced by 3.3 h on average, which could provide treatment evidence for clinicians in advance.

Development and validation of a quick, automated, and reproducible ATR FT-IR spectroscopy machine-learning model for Klebsiella pneumoniae typing

The reliability of Fourier-transform infrared (FT-IR) spectroscopy for Klebsiella pneumoniae typing and outbreak control has been previously assessed, but issues remain in standardization and reproducibility. We developed and validated a reproducible FT-IR with attenuated total reflectance (ATR) workflow for the identification of K. pneumoniae lineages. We used 293 isolates representing multidrug-resistant K. pneumoniae lineages causing outbreaks worldwide (2002-2021) to train a random forest classification (RF) model based on capsular (KL)-type discrimination. This model was validated with 280 contemporaneous isolates (2021-2022), using wzi sequencing and whole-genome sequencing as references. Repeatability and reproducibility were tested in different culture media and instruments throughout time. Our RF model allowed the classification of 33 capsular (KL)-types and up to 36 clinically relevant K. pneumoniae lineages based on the discrimination of specific KL- and O-type combinations. We obtained high rates of accuracy (89%), sensitivity (88%), and specificity (92%), including from cultures obtained directly from the clinical sample, allowing to obtain typing information the same day bacteria are identified. The workflow was reproducible in different instruments throughout time (>98% correct predictions). Direct colony application, spectral acquisition, and automated KL prediction through Clover MS Data analysis software allow a short time-to-result (5 min/isolate). We demonstrated that FT-IR ATR spectroscopy provides meaningful, reproducible, and accurate information at a very early stage (as soon as bacterial identification) to support infection control and public health surveillance. The high robustness together with automated and flexible workflows for data analysis provide opportunities to consolidate real-time applications at a global level. IMPORTANCE We created and validated an automated and simple workflow for the identification of clinically relevant Klebsiella pneumoniae lineages by FT-IR spectroscopy and machine-learning, a method that can be extremely useful to provide quick and reliable typing information to support real-time decisions of outbreak management and infection control. This method and workflow is of interest to support clinical microbiology diagnostics and to aid public health surveillance.

Antimicrobial resistance crisis: could artificial intelligence be the solution?

Antimicrobial resistance is a global public health threat, and the World Health Organization (WHO) has announced a priority list of the most threatening pathogens against which novel antibiotics need to be developed. The discovery and introduction of novel antibiotics are time-consuming and expensive. According to WHO's report of antibacterial agents in clinical development, only 18 novel antibiotics have been approved since 2014. Therefore, novel antibiotics are critically needed. Artificial intelligence (AI) has been rapidly applied to drug development since its recent technical breakthrough and has dramatically improved the efficiency of the discovery of novel antibiotics. Here, we first summarized recently marketed novel antibiotics, and antibiotic candidates in clinical development. In addition, we systematically reviewed the involvement of AI in antibacterial drug development and utilization, including small molecules, antimicrobial peptides, phage therapy, essential oils, as well as resistance mechanism prediction, and antibiotic stewardship.

Machine learning-based antibiotic resistance prediction models: An updated systematic review and meta-analysis

BACKGROUND

The widespread use of antibiotics has led to a gradual adaptation of bacteria to these drugs, diminishing the effectiveness of treatments.

OBJECTIVE

To comprehensively assess the research progress of antibiotic resistance prediction models based on machine learning (ML) algorithms, providing the latest quantitative analysis and methodological evaluation.

METHODS

Relevant literature was systematically retrieved from databases, including PubMed, Embase and the Cochrane Library, from inception up to December 2023. Studies meeting predefined criteria were selected for inclusion. The prediction model risk of bias assessment tool was employed for methodological quality assessment, and a random-effects model was utilised for meta-analysis.

RESULTS

The systematic review included a total of 22 studies with a combined sample size of 43,628; 10 studies were ultimately included in the meta-analysis. Commonly used ML algorithms included random forest, decision trees and neural networks. Frequently utilised predictive variables encompassed demographics, drug use history and underlying diseases. The overall sensitivity was 0.57 (95% CI: 0.42-0.70; p< 0.001; I2= 99.7%), the specificity was 0.95 (95% CI: 0.79-0.99; p< 0.001; I2 = 99.9%), the positive likelihood ratio was 10.7 (95% CI: 2.9-39.5), the negative likelihood ratio was 0.46 (95% CI: 0.34-0.61), the diagnostic odds ratio was 23 (95% CI: 7-81) and the area under the receiver operating characteristic curve was 0.78 (95% CI: 0.74-0.81; p< 0.001), indicating a good discriminative ability of ML models for antibiotic resistance. However, methodological assessment and funnel plots suggested a high risk of bias and publication bias in the included studies.

CONCLUSION

This meta-analysis provides a current and comprehensive evaluation of ML models for predicting antibiotic resistance, emphasising their potential application in clinical practice. Nevertheless, stringent research design and reporting are warranted to enhance the quality and credibility of future studies. Future research should focus on methodological innovation and incorporate more high-quality studies to further advance this field.

Molecular antimicrobial susceptibility testing in sepsis

Rapidly detecting and identifying pathogens is crucial for appropriate antimicrobial therapy in patients with sepsis. Conventional diagnostic methods have been a great asset to medicine, though they are time consuming and labor intensive. This work will enable healthcare professionals to understand the bacterial community better and enhance their diagnostic capacity by using novel molecular methods that make obtaining quicker, more precise results possible. The authors discuss and critically assess the merits and drawbacks of molecular testing and the added value of these tests, including the shift turnaround time, the implication for clinicians' decisions, gaps in knowledge, future research directions and novel insights or innovations. The field of antimicrobial molecular testing has seen several novel insights and innovations to improve the diagnosis and management of infectious diseases.

Plasma/Serum Proteomics based on Mass Spectrometry

Human blood is a window of physiology and disease. Examination of biomarkers in blood is a common clinical procedure, which can be informative in diagnosis and prognosis of diseases, and in evaluating treatment effectiveness. There is still a huge demand on new blood biomarkers and assays for precision medicine nowadays, therefore plasma/serum proteomics has attracted increasing attention in recent years. How to effectively proceed with the biomarker discovery and clinical diagnostic assay development is a question raised to researchers who are interested in this area. In this review, we comprehensively introduce the background and advancement of technologies for blood proteomics, with a focus on mass spectrometry (MS). Analyzing existing blood biomarkers and newly-built diagnostic assays based on MS can shed light on developing new biomarkers and analytical methods. We summarize various protein analytes in plasma/serum which include total proteome, protein post-translational modifications, and extracellular vesicles, focusing on their corresponding sample preparation methods for MS analysis. We propose screening multiple protein analytes in the same set of blood samples in order to increase success rate for biomarker discovery. We also review the trends of MS techniques for blood tests including sample preparation automation, and further provide our perspectives on their future directions.

Biofilm antimicrobial susceptibility testing: where are we and where could we be going?

Our knowledge about the fundamental aspects of biofilm biology, including the mechanisms behind the reduced antimicrobial susceptibility of biofilms, has increased drastically over the last decades. However, this knowledge has so far not been translated into major changes in clinical practice. While the biofilm concept is increasingly on the radar of clinical microbiologists, physicians, and healthcare professionals in general, the standardized tools to study biofilms in the clinical microbiology laboratory are still lacking; one area in which this is particularly obvious is that of antimicrobial susceptibility testing (AST). It is generally accepted that the biofilm lifestyle has a tremendous impact on antibiotic susceptibility, yet AST is typically still carried out with planktonic cells. On top of that, the microenvironment at the site of infection is an important driver for microbial physiology and hence susceptibility; but this is poorly reflected in current AST methods. The goal of this review is to provide an overview of the state of the art concerning biofilm AST and highlight the knowledge gaps in this area. Subsequently, potential ways to improve biofilm-based AST will be discussed. Finally, bottlenecks currently preventing the use of biofilm AST in clinical practice, as well as the steps needed to get past these bottlenecks, will be discussed.

Rapid discrimination of Bifidobacterium longum subspecies based on MALDI-TOF MS and machine learning

Although MALDI-TOF mass spectrometry (MS) is widely known as a rapid and cost-effective reference method for identifying microorganisms, its commercial databases face limitations in accurately distinguishing specific subspecies of Bifidobacterium. This study aimed to explore the potential of MALDI-TOF MS protein profiles, coupled with prediction methods, to differentiate between Bifidobacterium longum subsp. infantis (B. infantis) and Bifidobacterium longum subsp. longum (B. longum). The investigation involved the analysis of mass spectra of 59 B. longum strains and 41 B. infantis strains, leading to the identification of five distinct biomarker peaks, specifically at m/z 2,929, 4,408, 5,381, 5,394, and 8,817, using Recurrent Feature Elimination (RFE). To facilate classification between B. longum and B. infantis based on the mass spectra, machine learning models were developed, employing algorithms such as logistic regression (LR), random forest (RF), and support vector machine (SVM). The evaluation of the mass spectrometry data showed that the RF model exhibited the highest performace, boasting an impressive AUC of 0.984. This model outperformed other algorithms in terms of accuracy and sensitivity. Furthermore, when employing a voting mechanism on multi-mass spectrometry data for strain identificaton, the RF model achieved the highest accuracy of 96.67%. The outcomes of this research hold the significant potential for commercial applications, enabling the rapid and precise discrimination of B. longum and B. infantis using MALDI-TOF MS in conjunction with machine learning. Additionally, the approach proposed in this study carries substantial implications across various industries, such as probiotics and pharmaceuticals, where the precise differentiation of specific subspecies is essential for product development and quality control.

Prediction of methicillin-resistant Staphylococcus aureus and carbapenem-resistant Klebsiella pneumoniae from flagged blood cultures by combining rapid Sepsityper MALDI-TOF mass spectrometry with machine learning

This study investigated combination of the Rapid Sepsityper Kit and a machine learning (ML)-based matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) approach for rapid prediction of methicillin-resistant Staphylococcus aureus (MRSA) and carbapenem-resistant Klebsiella pneumoniae (CRKP) from positive blood culture bottles. The study involved 461 patients with monomicrobial bloodstream infections. Species identification was performed using the conventional MALDI-TOF MS Biotyper system and the Rapid Sepsityper protocol. The data underwent preprocessing steps, and ML models were trained using preprocessed MALDI-TOF data and corresponding labels. The interpretability of the model was enhanced using SHapely Additive exPlanations values to identify significant features. In total, 44 S. aureus isolates comprising 406 MALDI-TOF MS files and 126 K. pneumoniae isolates comprising 1249 MALDI-TOF MS files were evaluated. This study demonstrated the feasibility of predicting MRSA among S. aureus and CRKP among K. pneumoniae isolates using MALDI-TOF MS and Sepsityper. Accuracy, area under the receiver operating characteristic curve, and F1 score for MRSA/methicillin-susceptible S. aureus were 0.875, 0.898 and 0.904, respectively; for CRKP/carbapenem-susceptible K. pneumoniae, these values were 0.766, 0.828 and 0.795, respectively. In conclusion, the novel ML-based MALDI-TOF MS approach enables rapid identification of MRSA and CRKP from flagged blood cultures within 1 h. This enables earlier initiation of targeted antimicrobial therapy, reducing deaths due to sepsis. The favourable performance and reduced turnaround time of this method suggest its potential as a rapid detection strategy in clinical microbiology laboratories, ultimately improving patient outcomes.

Multimodal learning in clinical proteomics: enhancing antimicrobial resistance prediction models with chemical information

MOTIVATION

Large-scale clinical proteomics datasets of infectious pathogens, combined with antimicrobial resistance outcomes, have recently opened the door for machine learning models which aim to improve clinical treatment by predicting resistance early. However, existing prediction frameworks typically train a separate model for each antimicrobial and species in order to predict a pathogen's resistance outcome, resulting in missed opportunities for chemical knowledge transfer and generalizability.

RESULTS

We demonstrate the effectiveness of multimodal learning over proteomic and chemical features by exploring two clinically relevant tasks for our proposed deep learning models: drug recommendation and generalized resistance prediction. By adopting this multi-view representation of the pathogenic samples and leveraging the scale of the available datasets, our models outperformed the previous single-drug and single-species predictive models by statistically significant margins. We extensively validated the multi-drug setting, highlighting the challenges in generalizing beyond the training data distribution, and quantitatively demonstrate how suitable representations of antimicrobial drugs constitute a crucial tool in the development of clinically relevant predictive models.

AVAILABILITY AND IMPLEMENTATION

The code used to produce the results presented in this article is available at https://github.com/BorgwardtLab/MultimodalAMR.