key: cord-0934320-w6sh7xpn authors: Egli, Adrian; Schrenzel, Jacques; Greub, Gilbert title: Digital microbiology date: 2020-06-27 journal: Clin Microbiol Infect DOI: 10.1016/j.cmi.2020.06.023 sha: 9e70f21b0cfdca13de7105bd8ff291bafeeb6409 doc_id: 934320 cord_uid: w6sh7xpn BACKGROUND: Digitalisation and artificial intelligence have an important impact on the way microbiology laboratories will work in the near future. Opportunities and challenges lay ahead to digitalise the microbiological workflows. Making an efficient use of big data, machine learning, and artificial intelligence in clinical microbiology requires a profound understanding of data handling aspects. OBJECTIVE: This review article summarizes the most important concepts of digital microbiology. The article provides microbiologists, clinicians and data scientists a viewpoint and practical examples along the diagnostic process. SOURCES: We used peer-reviewed literature identified by a Pubmed search for digitalisation, machine learning, artificial intelligence and microbiology. CONTENT: We describe the opportunities and challenges of digitalisation in microbiological diagnostic process with various examples. We also provide in this context key aspects of data structure and interoperability, as well as legal aspects. Finally, we outline the way for applications in a modern microbiology laboratory. IMPLICATIONS: We predict that digitalization and the usage of machine learning will have a profound impact on the daily routine of the laboratory staff. Along the analytical process, the most important steps should be identified, where digital technologies can be applied and provide a benefit. The education of all staff involved should be adapted to prepare for the advances in digital microbiology. inflammatory response syndrome (SIRS) and the presence of a central venous line, the risk of blood culture contamination can be assessed 20 . In the future, the combination of LIS and electronic health 2 record (EHR) data may allow more sophisticated feedback loops and provide automated quality 3 assessments reports to the microbiologist and clinician. Another important pre-analytical aspect is diagnostic stewardship. Diagnostic stewardship 5 incorporates the concept of recommending the best diagnostic approach for a given situation [21] [22] [23] . Digital solutions in this field may range from digital twins 24, 25 to machine-learning based algorithms in 7 smartphone app 26 or chatbots 27, 28 . Recently, chatbots have been developed to support the diagnostic 8 evaluation and recommending immediate measures, when patients are exposed to SARS-CoV-2 27 . Similarly to a microbiologist consultant, a chatbot may provide helpful diagnostic information and 10 advice e.g. on the correct transport media for a sample, assay costs, the expected turn-around time, and test performance in specific sample types. Such an interactive tool may be a first source of 12 information for routine and repetitive questions, and could support the pre-analytical quality Test performance and data generation within the laboratory are parts of analytics. As an example, automated microscopy allows to acquire high-resolution images of smears from positive blood 23 cultures and can categorize Gram staining with high sensitivity and specificity 30, 31 . Besides state-of-24 the-art automated microscopes, smartphones can also be used for image analysis of microscopy 25 data 32, 33 . Automated plate reading systems act similarly on pattern recognition and can reliably 26 recognize bacterial growth on a agar plate and could be used to pre-screen culture plates [34] [35] [36] [37] [38] . Such 27 automated plate reading systems are currently established in many European laboratories as part of 28 the ongoing automation process. Reading of E-tests and inhibition zone diameters around antibiotic-29 impregnated disks can also be automatized with well-developed reading software 39, 40 . clinical decision support systems based on machine learning to provide automated feedback 7 regarding empiric antibiotic prescription adapted to specific patient groups 46 . As a next step, also 8 more complex datasets will be analysed. As physiology and laboratory parameters can rapidly change 9 during an infection, time-series data greatly impact the predictive values of such algorithms -similar 10 to a doctor, who observers the patient during disease progression -machine learning algorithms will 11 also follow the patient's data stream. Recently, a series of studies has shown the impact of highfrequency physiological parameters in ICUs on the prediction of sepsis 47-49 or meningitis 50, 51 . These 13 studies are retrospective analysis and prospective controlled validation studies are largely missing in 14 the field. Therefore, although our expectations for digital microbiology may be high, we should remain 15 critical and carefully address the associated challenges. Challenges of digitalisation in the microbiology diagnostic process The collection, quality control and cleaning, storage, security and protection, stewardship and governance, interoperability and interconnection, reporting and visualization, versioning, and sharing 20 of data pose considerable challenges for big data in microbiology diagnostic laboratories. Some of 21 these data handling aspects may be managed with a profound understanding of the laboratory and Due to the increasing quantity of data (explosion of information), it will soon become almost A first step: Data structure and interoperability diseases (Figure 1 ) [72] [73] [74] . Machine learning algorithms require large, structured, interoperable, and 2 interconnected datasets. Healthcare data has to be further standardized and annotated with 3 international recognized definitions 75, 76 . Ontologies help to structure data in such a way by using a 4 common vocabulary, and allow to determine relations of variables within a data model 77 . The previously mentioned concepts for data handling have been used for a series of large healthcare University Hospitals. The goal is to discover digital biomarkers for early sepsis recognition and 8 prediction of mortality using machine learning algorithms (www.sphn.ch/). Epidemiological databases can also benefit from structured data. For example, Pulsenet is a large 14 15 predictions or decisions without being explicitly programmed to perform that task 9, 112 . Machine 18 learning algorithms may be used at each step of the microbiological diagnostic process from pre-to 19 post-analytics, helping us to deal with the increasing quantities and complexity of data 113,114 (Table 1) . Human analytical capacity has reached its limits to (i) grasp the huge amount of available complex process management is key, (ii) data handling is easiest at the point where the data is actually diagnostic tests. In general, incentives are needed to further support all aspects of data handling in 3 laboratory medicine -including standardization data structures and machine learning algorithms. Conclusion 6 Digitalisation in healthcare shows already a profound impact on patients. It is expected, that the 7 developments started will further gain momentum. Machine learning radically changes the way we 8 handle healthcare-related data -including data of clinical microbiology and infectious diseases. Likely, we will move from the internet-of-things environment (interconnected datasets in a patient with in a disease-free time. In addition, developments of molecular diagnostics such as metagenomics will 12 increase the data complexity. Current trends indicate, that the importance of laboratory diagnostics We have to develop strategies for the next five to ten years to face the opportunities and challenges 1 2 Table S1 . Glossary Basel) for critically feedback regarding the manuscript. Conflict of interest disclosure: None of the authors had a conflict of interest. Quality control How reliable is the analytical performance of a test? -Surveillance of reagent lots performance with internal and external controls and automated reported in connection to specific used lots of time. Imaging Are there bacteria on the microscope slide? -Automated image acquisition with a microscope and scan for pathogen-like structures and category 30, 32, 33 Plate reading Is there bacterial growth on the plate? -Automated image acquisition and scan for colonies and subsequent identification (telebacteriology). Expert system Does the detected resistance profile make sense? -Medical validation of antibiotic resistance profiles with expert database. Public Health Is there a potential outbreak? -Automated screening for pathogen similarities e.g. resistance profile or automated bioinformatics 130,131 Is there a potential bacterial phenotype? -Detection of resistance by analysing MALDI-TOF spectra 43, 44 Sepsis treatment What is the best treatment for the patient? -Prediction of sepsis, and best treatment e.g. volume and antibiotics for the patient 47-49 Tracking Strains in the Microbiome: Insights from Metagenomics and Design and evaluation of a bacterial clinical infectious diseases ontology Ontologies for clinical and translational research: 35 Introduction Semantic data interoperability, digital medicine, and e-37 health in infectious disease management: a review The Need for a Global Language -SNOMED CT Introduction. Stud Health Technol MIMIC-III, a freely accessible critical care database The eICU Collaborative Research Database, a freely available multi-center 14 database for critical care research PulseNet and the Changing Paradigm of Laboratory-Based Surveillance Microreact: visualizing and sharing data for genomic epidemiology and 19 phylogeography Nextstrain: real-time tracking of pathogen evolution Improving the quality and workflow of bacterial genome sequencing and 23 analysis: paving the way for a Switzerland-wide molecular epidemiological surveillance 24 platform A comprehensive collection of systems biology data characterizing Infectious Diseases and Associated Ethical Impacts Big data and machine learning in 35 critical care: Opportunities for collaborative research Privacy in the age of medical big data The COMPARE Data Hubs. Database (Oxford) 2019, justice point of view Introduction to machine learning Machine learning in infection management using routine electronic health Machine learning for clinical decision support in infectious diseases: 13 a narrative review of current applications Machine Learning for Healthcare: On the Verge of a Major Shift in 118 Supervised machine learning for the prediction of infection on admission 18 to hospital: a prospective observational cohort study Using artificial intelligence to reduce 21 diagnostic workload without compromising detection of urinary tract infections Unsupervised extraction of epidemic syndromes from participatory influenza stewardship: fair enough? The Promise of the Internet of 4 Things in Healthcare: How Hard Is It to Keep? Stud Wearable Devices in Medical Internet of Things: Scientific 7 Research and Commercially Available Devices