Machine Learning Approach for Candida albicans Fluconazole Resistance Detection Using Matrix-Assisted Laser Desorption/Ionization Time-of-Flight Mass Spectrometry

Superficial Fungal Infections and Artificial Intelligence: A Review on Current Advances and Opportunities: REVISION

BACKGROUND

Superficial fungal infections are among the most common infections in world, they mainly affect skin, nails and scalp without further invasion. Superficial fungal diseases are conventionally diagnosed with direct microscopy, fungal culture or histopathology, treated with topical or systemic antifungal agents and prevented in immunocompetent patients by improving personal hygiene. However, conventional diagnostic tests can be time-consuming, also treatment can be insufficient or ineffective and prevention can prove to be demanding. Artificial Intelligence (AI) refers to a digital system having an intelligence akin to a human being. The concept of AI has existed since 1956, but hasn't been practicalised until recently. AI has revolutionised medical research in the recent years, promising to influence almost all specialties of medicine.

OBJECTIVE

An increasing number of articles have been published about the usage of AI in cutaneous mycoses.

METHODS

In this review, the key findings of articles about utilisation of AI in diagnosis, treatment and prevention of superficial fungal infections are summarised. Moreover, the need for more research and development is highlighted.

RESULTS

Fifty-four studies were reviewed. Onychomycosis was the most researched superficial fungal infection. AI can be used diagnosing fungi in macroscopic and microscopic images and classify them to some extent. AI can be a tool and be used as a part of something bigger to diagnose superficial mycoses.

CONCLUSION

AI can be used in all three steps of diagnosing, treating and preventing. AI can be a tool complementary to the clinician's skills and laboratory results.

Genotyping of oral Candida albicans and Candida tropicalis strains in patients with orofacial clefts undergoing surgical rehabilitation by MALDI-TOF MS: Case-series study

Patients with orofacial clefts are more likely to develop oral fungal diseases due to anatomo-physiological changes and surgical rehabilitation treatment. This case-series study evaluated the genetic diversity and dynamics of oral colonization and spread of C. albicans and C. tropicalis in four patients with orofacial clefts, from the time of hospital admission, perioperative and outpatient follow-up, with specialized physician. Candida biotypes previously identified by CHROMagar Candida and PCR methods were studied by MALDI-TOF MS assays and clustering analyses. Possible correlations with pathogenicity characteristics were observed, including production of hydrolytic exoenzymes and the antifungal sensitivity profiles. Amphotericin B-sensitive and fluconazole-resistant (low frequency) C. tropicalis and C. albicans, including clinically compatible MIC of nystatin, were found in the oral cavity of these patients. Clusters of isolates revealed phenomena of (i) elimination in the operative phase, (ii) maintenance or (iii) acquisition of oral C. tropicalis in the perioperative period and specialized outpatient and medical follow-up. For C. albicans, these phenomena included (i) elimination in the operative phase, (ii) acquisition in the operative phase and propagation from the hospital environment, and (iii) maintenance during hospitalization and operative phase. Amphotericin B and nystatin were shown to be effective in cases of clinical treatment and/or prophylaxis, especially considering the pre-existence of fluconazole-resistant strains. This study confirmed the phenomena of septic maintenance, septic neocolonization and septic elimination involving the opportunistic pathogens. MALDI-TOF MS associated with clustering analysis may assist the monitoring of clinical isolates or groups of epidemiologically important microbial strains in the hospital setting.

Harnessing Machine Learning to Uncover Hidden Patterns in Azole-Resistant CYP51/ERG11 Proteins

Fungal resistance is a public health concern due to the limited availability of antifungal resources and the complexities associated with treating persistent fungal infections. Azoles are thus far the primary line of defense against fungi. Specifically, azoles inhibit the conversion of lanosterol to ergosterol, producing defective sterols and impairing fluidity in fungal plasmatic membranes. Studies on azole resistance have emphasized specific point mutations in CYP51/ERG11 proteins linked to resistance. Although very insightful, the traditional approach to studying azole resistance is time-consuming and prone to errors during meticulous alignment evaluation. It relies on a reference-based method using a specific protein sequence obtained from a wild-type (WT) phenotype. Therefore, this study introduces a machine learning (ML)-based approach utilizing molecular descriptors representing the physiochemical attributes of CYP51/ERG11 protein isoforms. This approach aims to unravel hidden patterns associated with azole resistance. The results highlight that descriptors related to amino acid composition and their combination of hydrophobicity and hydrophilicity effectively explain the slight differences between the resistant non-wild-type (NWT) and WT (nonresistant) protein sequences. This study underscores the potential of ML to unravel nuanced patterns in CYP51/ERG11 sequences, providing valuable molecular signatures that could inform future endeavors in drug development and computational screening of resistant and nonresistant fungal lineages.

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.

Artificial Intelligence: Exploring utility in detection and typing of fungus with futuristic application in fungal cytology

Artificial Intelligence (AI) is an emerging, transforming and revolutionary technology that has captured attention worldwide. It is translating research into precision oncology treatments. AI can analyse large or big data sets requiring high-speed specialized computing solutions. The data are big in terms of volume and multimodal with the amalgamation of images, text and structure. Machine learning has identified antifungal drug targets, and taxonomic and phylogenetic classification of fungi based on sequence analysis is now available. Real-time identification tools and user-friendly mobile applications for identifying fungi have been discovered. Akin to histopathology, AI can be applied to fungal cytology. AI has been fruitful in cytopathology of the thyroid gland, breast, urine and uterine cervical lesions. AI has a huge scope in fungal cytology and would certainly bear fruit with its accuracy, reproducibility and capacity for handling big data. The purpose of this systematic review was to highlight the AI's utility in detecting fungus and its typing with a special focus on future application in fungal cytology. We also touch upon the basics of AI in brief.

In Vitro Susceptibility Tests in the Context of Antifungal Resistance: Beyond Minimum Inhibitory Concentration in Candida spp

Antimicrobial resistance is a matter of rising concern, especially in fungal diseases. Multiple reports all over the world are highlighting a worrisome increase in azole- and echinocandin-resistance among fungal pathogens, especially in Candida species, as reported in the recently published fungal pathogens priority list made by WHO. Despite continuous efforts and advances in infection control, development of new antifungal molecules, and research on molecular mechanisms of antifungal resistance made by the scientific community, trends in invasive fungal diseases and associated antifungal resistance are on the rise, hindering therapeutic options and clinical cures. In this context, in vitro susceptibility testing aimed at evaluating minimum inhibitory concentrations, is still a milestone in the management of fungal diseases. However, such testing is not the only type at a microbiologist's disposal. There are other adjunctive in vitro tests aimed at evaluating fungicidal activity of antifungal molecules and also exploring tolerance to antifungals. This plethora of in vitro tests are still left behind and performed only for research purposes, but their role in the context of invasive fungal diseases associated with antifungal resistance might add resourceful information to the clinical management of patients. The aim of this review was therefore to revise and explore all other in vitro tests that could be potentially implemented in current clinical practice in resistant and difficult-to-treat cases.

Improving the Detection of Epidemic Clones in Candida parapsilosis Outbreaks by Combining MALDI-TOF Mass Spectrometry and Deep Learning Approaches

Identifying fungal clones propagated during outbreaks in hospital settings is a problem that increasingly confronts biologists. Current tools based on DNA sequencing or microsatellite analysis require specific manipulations that are difficult to implement in the context of routine diagnosis. Using deep learning to classify the mass spectra obtained during the routine identification of fungi by MALDI-TOF mass spectrometry could be of interest to differentiate isolates belonging to epidemic clones from others. As part of the management of a nosocomial outbreak due to Candida parapsilosis in two Parisian hospitals, we studied the impact of the preparation of the spectra on the performance of a deep neural network. Our purpose was to differentiate 39 otherwise fluconazole-resistant isolates belonging to a clonal subset from 56 other isolates, most of which were fluconazole-susceptible, collected during the same period and not belonging to the clonal subset. Our study carried out on spectra obtained on four different machines from isolates cultured for 24 or 48 h on three different culture media showed that each of these parameters had a significant impact on the performance of the classifier. In particular, using different culture times between learning and testing steps could lead to a collapse in the accuracy of the predictions. On the other hand, including spectra obtained after 24 and 48 h of growth during the learning step restored the good results. Finally, we showed that the deleterious effect of the device variability used for learning and testing could be largely improved by including a spectra alignment step during preprocessing before submitting them to the neural network. Taken together, these experiments show the great potential of deep learning models to identify spectra of specific clones, providing that crucial parameters are controlled during both culture and preparation steps before submitting spectra to a classifier.

Fungal biotechnology: From yesterday to tomorrow

Fungi have been used to better the lives of everyday people and unravel the mysteries of higher eukaryotic organisms for decades. However, comparing progress and development stemming from fungal research to that of human, plant, and bacterial research, fungi remain largely understudied and underutilized. Recent commercial ventures have begun to gain popularity in society, providing a new surge of interest in fungi, mycelia, and potential new applications of these organisms to various aspects of research. Biotechnological advancements in fungal research cannot occur without intensive amounts of time, investments, and research tool development. In this review, we highlight past breakthroughs in fungal biotechnology, discuss requirements to advance fungal biotechnology even further, and touch on the horizon of new breakthroughs with the highest potential to positively impact both research and society.

Matrix-assisted laser desorption/ionisation-time of flight mass spectrometry azole susceptibility assessment in Candida and Aspergillus species

BACKGROUND

Matrix-assisted laser desorption/ionisation-time of flight mass spectrometry (MALDI-TOF MS) allows rapid pathogen identification and potentially can be used for antifungal susceptibility testing (AFST).

OBJECTIVES

We evaluated the performance of the MALDI-TOF MS in assessing azole susceptibility, with reduced incubation time, by comparing the results with the reference method Broth Microdilution.

METHODS

Resistant and susceptible strains of Candida (n = 15) were evaluated against fluconazole and Aspergillus (n = 15) against itraconazole and voriconazole. Strains were exposed to serial dilutions of the antifungals for 15 h. Microorganisms' protein spectra against all drug concentrations were acquired and used to generate a composite correlation index (CCI) matrix. The comparison of autocorrelations and cross-correlations between spectra facilitated by CCI was used as a similarity parameter between them, enabling the inference of a minimum profile change concentration breakpoint. Results obtained with the different AFST methods were then compared.

FINDINGS

The overall agreement between methods was 91.11%. Full agreement (100%) was reached for Aspergillus against voriconazole and Candida against fluconazole, and 73.33% of agreement was obtained for Aspergillus against itraconazole.

MAIN CONCLUSIONS

This study demonstrates MALDI-TOF MS' potential as a reliable and faster alternative for AFST. More studies are necessary for method optimisation and standardisation for clinical routine application.

Investigation of fluconazole susceptibility to Candida albicans by MALDI-TOF MS and real-time PCR for CDR1, CDR2, MDR1 and ERG11

BACKGROUND

C. albicans is a pathogenic yeast that is the most common cause of fungal infections in humans. Unfortunately, the yeast's resistance to the antifungal medication fluconazole (FLC) is increasing; furthermore, testing its susceptibility to FLC by conventional methods takes time, resulting in treatment failure. The susceptibility of C. albicans to FLC was investigated using MALDI-TOF Mass Spectrometry and Real-time PCR tests for CDR1, CDR2, MDR1 and ERG11. Overall, 32 C. albicans strains made up of four reference strains (three FLC susceptible [S] and one FLC resistant [R], one spontaneous mutant strain [FLC susceptible-dose-dependent (SDD)] and 27 clinical strains obtained from two Thai University Hospitals) were tested for susceptibility to FLC. The following tests were performed: SensititreYeastOne and broth microdilution method, FLC resistant expression mechanism by Real-time PCR, and the major peak determination by MALDI-TOF MS.

RESULTS

The change of CDR1 and CDR2 mRNA expression was only significantly observed in SDD and R strains. MALDI-TOF MS was performed after incubation for six hours; the change of mass spectral intensity at range 3376-3382 m/z (major peak) was significantly related to FLC susceptibility as SDD (decreased at 4 µg/mL and increased at 8 µg/mL), S (all increased), and R (all slightly decreased or no change). All 27 clinical strains showed FLC minimum inhibitory concentrations (MIC range 0.25-2 µg/mL), no change in CDR1 and CDR2 expression and S major peak type. The FLC resistant C. albicans with CDR1and CDR2 expression may possibly affect the change of mass spectral intensity at range 3376-3382 m/z.

CONCLUSIONS

The MALDI-TOF MS may be used to simultaneously classify and predict FLC resistant C. albicans strains associated with CDR1 and CDR2 expression. Further studies are essential to clarify the methodology and improve the reliability of this assay for routine diagnosis.

How Yeast Antifungal Resistance Gene Analysis Is Essential to Validate Antifungal Susceptibility Testing Systems

Objectives

The antifungal susceptibility testing (AFST) of yeast pathogen alerts clinicians about the potential emergence of resistance. In this study, we compared two commercial microdilution AFST methods: Sensititre YeastOne read visually (YO) and MICRONAUT-AM read visually (MN) or spectrophotometrically (MNV), interpreted with Clinical and Laboratory Standards Institute and European Committee on Antimicrobial Susceptibility Testing criteria, respectively.

Methods

Overall, 97 strains from 19 yeast species were measured for nine antifungal drugs including a total of 873 observations. First, the minimal inhibitory concentration (MIC) was compared between YO and MNV, and between MNV and MN, either directly or by assigning them to five susceptibility categories. Those categories were based on the number of MIC dilutions around the breakpoint or epidemiological cut-off reference values (ECOFFs or ECVs). Second, YO and MNV methods were evaluated for their ability to detect the elevation of MICs due to mutation in antifungal resistance genes, thanks to pairs or triplets of isogenic strains isolated from a single patient along a treatment previously analyzed for antifungal resistance gene mutations. Reproducibility measurement was evaluated, thanks to three quality control (QC) strains.

Results

YO and MNV direct MIC comparisons obtained a global agreement of 67%. Performing susceptibility category comparisons, only 22% and 49% of the MICs could be assigned to categories using breakpoints and ECOFFs/ECVs, respectively, and 40% could not be assigned due to the lack of criteria in both consortia. The YO and MN susceptibility categories gave accuracies as low as 50%, revealing the difficulty to implement this method of comparison. In contrast, using the antifungal resistance gene sequences as a gold standard, we demonstrated that both methods (YO and MN) were equally able to detect the acquisition of resistance in the Candida strains, even if MN showed a global lower MIC elevation than YO. Finally, no major differences in reproducibility were observed between the three AFST methods.

Conclusion

This study demonstrates the valuable use of both commercial microdilution AFST methods to detect antifungal resistance due to point mutations in antifungal resistance genes. We highlighted the difficulty to conduct conclusive analyses without antifungal gene sequence data as a gold standard. Indeed, MIC comparisons taking into account the consortia criteria of interpretation remain difficult even after the effort of harmonization.

Echinocandins Susceptibility Patterns of 2,787 Yeast Isolates: Importance of the Thresholds for the Detection of FKS Mutations

Since echinocandins are recommended as first line therapy for invasive candidiasis, detection of resistance, mainly due to alteration in FKS protein, is of main interest. EUCAST AFST recommends testing both MIC of anidulafungin and micafungin, and breakpoints (BPs) have been proposed to detect echinocandin-resistant isolates. We analyzed MIC distribution for all three available echinocandins of 2,787 clinical yeast isolates corresponding to 5 common and 16 rare yeast species, using the standardized EUCAST method for anidulafungin and modified for caspofungin and micafungin (AM3-MIC). In our database, 64 isolates of common pathogenic species were resistant to anidulafungin, according to the EUCAST BP, and/or to caspofungin, using our previously published threshold (AM3-MIC ≥ 0.5 mg/L). Among these 64 isolates, 50 exhibited 21 different FKS mutations. We analyzed the capacity of caspofungin AM3-MIC and anidulafungin MIC determination in detecting isolates with FKS mutation. They were always identified using caspofungin AM3-MIC and the local threshold while some isolates were misclassified using anidulafungin MIC and EUCAST threshold. However, both methods misclassified four wild-type C. glabrata as resistant. Based on a large data set from a single center, the use of AM3-MIC testing for caspofungin looks promising in identifying non-wild-type C. albicans, C. tropicalis and P. kudiravzevii isolates, but additional multicenter comparison is mandatory to conclude on the possible superiority of AM3-MIC testing compared to the EUCAST method.

MALDI-TOF mass spectrometry of saliva samples as a prognostic tool for COVID-19

Background

Methods

Results

Conclusion

Artificial intelligence and machine learning assisted drug delivery for effective treatment of infectious diseases

In the era of antimicrobial resistance, the prevalence of multidrug-resistant microorganisms that resist conventional antibiotic treatment has steadily increased. Thus, it is now unquestionable that infectious diseases are significant global burdens that urgently require innovative treatment strategies. Emerging studies have demonstrated that artificial intelligence (AI) can transform drug delivery to promote effective treatment of infectious diseases. In this review, we propose to evaluate the significance, essential principles, and popular tools of AI in drug delivery for infectious disease treatment. Specifically, we will focus on the achievements and key findings of current research, as well as the applications of AI on drug delivery throughout the whole antimicrobial treatment process, with an emphasis on drug development, treatment regimen optimization, drug delivery system and administration route design, and drug delivery outcome prediction. To that end, the challenges of AI in drug delivery for infectious disease treatments and their current solutions and future perspective will be presented and discussed.

Investigation of MALDI-TOF Mass Spectrometry for Assessing the Molecular Diversity of Campylobacter jejuni and Comparison with MLST and cgMLST: A Luxembourg One-Health Study

There is a need for active molecular surveillance of human and veterinary Campylobacter infections. However, sequencing of all isolates is associated with high costs and a considerable workload. Thus, there is a need for a straightforward complementary tool to prioritize isolates to sequence. In this study, we proposed to investigate the ability of MALDI-TOF MS to pre-screen C. jejuni genetic diversity in comparison to MLST and cgMLST. A panel of 126 isolates, with 10 clonal complexes (CC), 21 sequence types (ST) and 42 different complex types (CT) determined by the SeqSphere+ cgMLST, were analysed by a MALDI Biotyper, resulting into one average spectra per isolate. Concordance and discriminating ability were evaluated based on protein profiles and different cut-offs. A random forest algorithm was trained to predict STs. With a 94% similarity cut-off, an AWC of 1.000, 0.933 and 0.851 was obtained for MLST_CC, MLST_ST and cgMLST profile, respectively. The random forest classifier showed a sensitivity and specificity up to 97.5% to predict four different STs. Protein profiles allowed to predict C. jejuni CCs, STs and CTs at 100%, 93% and 85%, respectively. Machine learning and MALDI-TOF MS could be a fast and inexpensive complementary tool to give an early signal of recurrent C. jejuni on a routine basis.

Can We Improve Antifungal Susceptibility Testing?

Systemic antifungal agents are increasingly used for prevention or treatment of invasive fungal infections, whose prognosis remains poor. At the same time, emergence of resistant or even multi-resistant strains is of concern as the antifungal arsenal is limited. Antifungal susceptibility testing (AFST) is therefore of key importance for patient management and antifungal stewardship. Current AFST methods, including reference and commercial types, are based on growth inhibition in the presence of an antifungal, in liquid or solid media. They usually enable Minimal Inhibitory Concentrations (MIC) to be determined with direct clinical application. However, they are limited by a high turnaround time (TAT). Several innovative methods are currently under development to improve AFST. Techniques based on MALDI-TOF are promising with short TAT, but still need extensive clinical validation. Flow cytometry and computed imaging techniques detecting cellular responses to antifungal stress other than growth inhibition are also of interest. Finally, molecular detection of mutations associated with antifungal resistance is an intriguing alternative to standard AFST, already used in routine microbiology labs for detection of azole resistance in Aspergillus and even directly from samples. It is still restricted to known mutations. The development of Next Generation Sequencing (NGS) and whole-genome approaches may overcome this limitation in the near future. While promising approaches are under development, they are not perfect and the ideal AFST technique (user-friendly, reproducible, low-cost, fast and accurate) still needs to be set up routinely in clinical laboratories.

Detection of azole resistance in Aspergillus fumigatus complex isolates using MALDI-TOF mass spectrometry

OBJECTIVES

The main goal of this study was to accurately detect azole resistance in species of the Aspergillus fumigatus complex by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS).

METHODS

Identification of isolates (n = 868) was done with MALDI-TOF MS using both commercial and in-house libraries. To determine azole susceptibility, the EUCAST E.Def. 9.3.2 method was applied as the reference standard. Identification of resistant isolates was confirmed by DNA sequence analysis. Protein spectra obtained by MALDI-TOF MS were analysed to differentiate species within the A. fumigatus complex and to detect azole-resistant A. fumigatus sensu stricto isolates.

RESULTS

Correct discrimination of A. fumigatus sensu stricto from cryptic species was accomplished in 100% of the cases applying principal component analysis (PCA) to protein spectra generated by MALDI-TOF MS. Furthermore, a specific peak (4586 m/z) was found to be present only in cryptic species. The application of partial least squares (PLS) discriminant analysis allowed 98.43% (±0.038) discrimination between susceptible and azole-resistant A. fumigatus sensu stricto isolates. Finally, based on PLS and SVM, A. fumigatus sensu stricto isolates with different cyp51A gene mutations were correctly clustered in 91.5% of the cases.

CONCLUSIONS

MALDI-TOF MS combined with peak analysis is a novel tool that allows the differentiation of A. fumigatus sensu stricto from other species within the A. fumigatus complex, as well as the detection of azole-resistant A. fumigatus sensu stricto. Although further studies are still needed, the results reported here show the great potential of MALDI-TOF and machine learning for the rapid detection of azole-resistant Aspergillus fumigatus isolates from clinical origins.

Matrix-Assisted Laser Desorption Ionization Time-of-Flight for Fungal Identification

Many studies have shown successful performance of matrix-assisted laser desorption ionization time-of-flight mass spectrometry for rapid yeast and mold identification, yet few laboratories have chosen to apply this technology into their routine clinical mycology workflow. This review provides an overview of the current status of matrix-assisted laser desorption ionization time-of-flight mass spectrometry for fungal identification, including key findings in the literature, processing and database considerations, updates in technology, and exciting future prospects. Significant advances toward standardization have taken place recently; thus, accurate species-level identification of yeasts and molds should be highly attainable, achievable, and practical in most clinical laboratories.

Identification of Fungi Isolated from Oral Cavity of Patients with HIV Using MALDI-TOF MS

BACKGROUND

A growing incidence of invasive fungal infections, especially among immunocompromised patients, has given increased significance to microbiological diagnostics of yeast-like fungi. More accurate and faster fungi identification methods that can compete with classical methods are being searched for. In this paper, classical microbiological methods are compared to MALDI-TOF MS (matrix-assisted laser desorption/ionization time-of-flight mass spectrometry).

METHODS

The diagnostic material was collected from buccal mucosa from 98 adults, including 69 with HIV. Only positive cultures were included in the study.

RESULTS

Matching results were obtained in 45 samples, and there were nonmatching results in 35 samples, with the majority of these in the study group, constituting 50% of identifications within this group. A particularly common mistake resulting from the use of classical methods is the false identification of C. dubliniensis as C. albicans. Additionally, C. tropicalis proves to be difficult to identify.

CONCLUSIONS

Our results and literature data suggest that MALDI-TOF MS should be considered an effective alternative to classical methods in terms of fungi identification, especially among HIV-positive patients, due to the different morphology of fungal colonies.

The Evolving Landscape of Fungal Diagnostics, Current and Emerging Microbiological Approaches

Invasive fungal infections are increasingly recognized in immunocompromised hosts. Current diagnostic techniques are limited by low sensitivity and prolonged turnaround times. We review emerging diagnostic technologies and platforms for diagnosing the clinically invasive disease caused by Candida, Aspergillus, and Mucorales.

Rapid Antifungal Susceptibility Testing of Yeasts and Molds by MALDI-TOF MS: A Systematic Review and Meta-Analysis

Due to the growing burden of fungal infections and a recent rise in antifungal resistance, antifungal susceptibility testing (AFST) is of increasing importance. The common methods of AFST have turnaround times of 24 to 48 h, and the available rapid methods are limited by applicability, cost-efficiency or accuracy. Given the urgency of adequate antifungal treatment in invasive mycoses, the need for the rapid and reliable detection of resistance is evident. In this systematic review and meta-analysis, we evaluated the diagnostic accuracy of AFST based on matrix-assisted laser desorption ionization time of flight mass spectrometry (MALDI-TOF MS). Twelve studies were reviewed, and data for the comparative analysis of their accuracy and methodology were systematically extracted. Compared to broth dilution as the gold standard, MALDI-TOF MS-based AFST reached a pooled sensitivity and specificity of 91% (95% Confidence Interval [CI], 84% to 96%) and 95% (95% CI, 90% to 98%), respectively. A comparative analysis showed that the sensitivity was higher for the semi-quantitative matrix-assisted laser desorption ionization Biotyper antibiotic susceptibility test rapid assay (MBT ASTRA) technique (96%) than for the correlate composite index (CCI) approach (85%), which is based on spectrum changes. Turnaround times below eight hours reached better diagnostic values than longer incubation periods, qualifying MALDI-TOF MS-based AFST as a rapid and accurate method for the detection of antifungal resistance.

Helix Matrix Transformation Combined With Convolutional Neural Network Algorithm for Matrix-Assisted Laser Desorption Ionization-Time of Flight Mass Spectrometry-Based Bacterial Identification

Matrix-assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF MS) analysis is a rapid and reliable method for bacterial identification. Classification algorithms, as a critical part of the MALDI-TOF MS analysis approach, have been developed using both traditional algorithms and machine learning algorithms. In this study, a method that combined helix matrix transformation with a convolutional neural network (CNN) algorithm was presented for bacterial identification. A total of 14 bacterial species including 58 strains were selected to create an in-house MALDI-TOF MS spectrum dataset. The 1D array-type MALDI-TOF MS spectrum data were transformed through a helix matrix transformation into matrix-type data, which was fitted during the CNN training. Through the parameter optimization, the threshold for binarization was set as 16 and the final size of a matrix-type data was set as 25 × 25 to obtain a clean dataset with a small size. A CNN model with three convolutional layers was well trained using the dataset to predict bacterial species. The filter sizes for the three convolutional layers were 4, 8, and 16. The kernel size was three and the activation function was the rectified linear unit (ReLU). A back propagation neural network (BPNN) model was created without helix matrix transformation and a convolution layer to demonstrate whether the helix matrix transformation combined with CNN algorithm works better. The areas under the receiver operating characteristic (ROC) curve of the CNN and BPNN models were 0.98 and 0.87, respectively. The accuracies of the CNN and BPNN models were 97.78 ± 0.08 and 86.50 ± 0.01, respectively, with a significant statistical difference (p < 0.001). The results suggested that helix matrix transformation combined with the CNN algorithm enabled the feature extraction of the bacterial MALDI-TOF MS spectrum, which might be a proposed solution to identify bacterial species.

Recent evolutions of machine learning applications in clinical laboratory medicine

Machine learning (ML) is gaining increased interest in clinical laboratory medicine, mainly triggered by the decreased cost of generating and storing data using laboratory automation and computational power, and the widespread accessibility of open source tools. Nevertheless, only a handful of ML-based products are currently commercially available for routine clinical laboratory practice. In this review, we start with an introduction to ML by providing an overview of the ML landscape, its general workflow, and the most commonly used algorithms for clinical laboratory applications. Furthermore, we aim to illustrate recent evolutions (2018 to mid-2020) of the techniques used in the clinical laboratory setting and discuss the associated challenges and opportunities. In the field of clinical chemistry, the reviewed applications of ML algorithms include quality review of lab results, automated urine sediment analysis, disease or outcome prediction from routine laboratory parameters, and interpretation of complex biochemical data. In the hematology subdiscipline, we discuss the concepts of automated blood film reporting and malaria diagnosis. At last, we handle a broad range of clinical microbiology applications, such as the reduction of diagnostic workload by laboratory automation, the detection and identification of clinically relevant microorganisms, and the detection of antimicrobial resistance.

Prediction of malaria transmission drivers in Anopheles mosquitoes using artificial intelligence coupled to MALDI-TOF mass spectrometry

Vector control programmes are a strategic priority in the fight against malaria. However, vector control interventions require rigorous monitoring. Entomological tools for characterizing malaria transmission drivers are limited and are difficult to establish in the field. To predict Anopheles drivers of malaria transmission, such as mosquito age, blood feeding and Plasmodium infection, we evaluated artificial neural networks (ANNs) coupled to matrix-assisted laser desorption ionization-time of flight (MALDI-TOF) mass spectrometry (MS) and analysed the impact on the proteome of laboratory-reared Anopheles stephensi mosquitoes. ANNs were sensitive to Anopheles proteome changes and specifically recognized spectral patterns associated with mosquito age (0–10 days, 11–20 days and 21–28 days), blood feeding and P. berghei infection, with best prediction accuracies of 73%, 89% and 78%, respectively. This study illustrates that MALDI-TOF MS coupled to ANNs can be used to predict entomological drivers of malaria transmission, providing potential new tools for vector control. Future studies must assess the field validity of this new approach in wild-caught adult Anopheles. A similar approach could be envisaged for the identification of blood meal source and the detection of insecticide resistance in Anopheles and to other arthropods and pathogens.