Deep learning for diagnosis and survival prediction in soft tissue sarcoma.

BACKGROUND: Clinical management of soft tissue sarcoma (STS) is particularly challenging. Here, we used digital pathology and deep learning (DL) for diagnosis and prognosis prediction of STS. PATIENTS AND METHODS: Our retrospective, multicenter study included a total of 506 histopathological slides from 291 patients with STS. The Cancer Genome Atlas cohort (240 patients) served as training and validation set. A second, multicenter cohort (51 patients) served as an additional test set. The use of the DL model (DLM) as a clinical decision support system was evaluated by nine pathologists with different levels of expertise. For prognosis prediction, 139 slides from 85 patients with leiomyosarcoma (LMS) were used. Area under the receiver operating characteristic (AUROC) and accuracy served as main outcome measures. RESULTS: The DLM achieved a mean AUROC of 0.97 (±0.01) and an accuracy of 79.9% (±6.1%) in diagnosing the five most common STS subtypes. The DLM significantly improved the accuracy of the pathologists from 46.3% (±15.5%) to 87.1% (±11.1%). Furthermore, they were significantly faster and more certain in their diagnosis. In LMS, the mean AUROC in predicting the disease-specific survival status was 0.91 (±0.1) and the accuracy was 88.9% (±9.9%). Cox regression showed the DLM's prediction to be a significant independent prognostic factor (P = 0.008, hazard ratio 5.5, 95% confidence interval 1.56-19.7) in these patients, outperforming other risk factors. CONCLUSIONS: DL can be used to accurately diagnose frequent subtypes of STS from conventional histopathological slides. It might be used for prognosis prediction in LMS, the most prevalent STS subtype in our cohort. It can also help pathologists to make faster and more accurate diagnoses. This could substantially improve the clinical management of STS patients.

Machine learning in infection management using routine electronic health records: tools, techniques, and reporting of future technologies.

BACKGROUND: Machine learning (ML) is increasingly being used in many areas of health care. Its use in infection management is catching up as identified in a recent review in this journal. We present here a complementary review to this work. OBJECTIVES: To support clinicians and researchers in navigating through the methodological aspects of ML approaches in the field of infection management. SOURCES: A Medline search was performed with the keywords artificial intelligence, machine learning, infection∗, and infectious disease∗ for the years 2014-2019. Studies using routinely available electronic hospital record data from an inpatient setting with a focus on bacterial and fungal infections were included. CONTENT: Fifty-two studies were included and divided into six groups based on their focus. These studies covered detection/prediction of sepsis (n = 19), hospital-acquired infections (n = 11), surgical site infections and other postoperative infections (n = 11), microbiological test results (n = 4), infections in general (n = 2), musculoskeletal infections (n = 2), and other topics (urinary tract infections, deep fungal infections, antimicrobial prescriptions; n = 1 each). In total, 35 different ML techniques were used. Logistic regression was applied in 18 studies followed by random forest, support vector machines, and artificial neural networks in 18, 12, and seven studies, respectively. Overall, the studies were very heterogeneous in their approach and their reporting. Detailed information on data handling and software code was often missing. Validation on new datasets and/or in other institutions was rarely done. Clinical studies on the impact of ML in infection management were lacking. IMPLICATIONS: Promising approaches for ML use in infectious diseases were identified. But building trust in these new technologies will require improved reporting. Explainability and interpretability of the models used were rarely addressed and should be further explored. Independent model validation and clinical studies evaluating the added value of ML approaches are needed.

Staphylococcus aureus spa type t437: identification of the most dominant community-associated clone from Asia across Europe.

Methicillin-resistant Staphylococcus aureus (MRSA) belonging to the multilocus sequence type clonal complex 59 (MLST CC59) is the predominant community-associated MRSA clone in Asia. This clone, which is primarily linked with the spa type t437, has so far only been reported in low numbers among large epidemiological studies in Europe. Nevertheless, the overall numbers identified in some Northern European reference laboratories have increased during the past decade. To determine whether the S. aureus t437 clone is present in other European countries, and to assess its genetic diversity across Europe, we analysed 147 S. aureus t437 isolates from 11 European countries collected over a period of 11 years using multiple locus variable number tandem repeat fingerprinting/analysis (MLVF/MLVA) and MLST. Additionally 16 S. aureus t437 isolates from healthy carriers and patients from China were included. Most isolates were shown to be monophyletic with 98% of the isolates belonging to the single MLVA complex 621, to which nearly all included isolates from China also belonged. More importantly, all MLST-typed isolates belonged to CC59. Our study implies that the European S. aureus t437 population represents a genetically tight cluster, irrespective of the year, country and site of isolation. This underpins the view that S. aureus CC59 has been introduced into several European countries, not being restricted to particular geographical regions or specific host environments. The European S. aureus t437 isolates thus bear the general hallmarks of a high-risk clone.

The dynamic changes of dominant clones of Staphylococcus aureus causing bloodstream infections in the European region: results of a second structured survey.

Staphylococcus aureus is one of the most important human pathogens and meticillin-resistant S. aureus (MRSA) presents a major cause of healthcare- and community-acquired infections. This study investigated the spatial and temporal changes of S. aureus causing bacteraemia in Europe over a five-year interval and explored the possibility of integrating pathogen-based typing data with epidemiological and clinical information at a European level. Between January 2011 and July 2011, 350 laboratories serving 453 hospitals in 25 countries collected 3,753 isolates (meticillin-sensitive S. aureus (MSSA) and MRSA) from patients with S. aureus bloodstream infections. All isolates were sent to the national staphylococcal reference laboratories and characterised by quality-controlled spa typing. Data were uploaded to an interactive web-based mapping tool. A wide geographical distribution of spa types was found, with some prevalent in all European countries. MSSA was more diverse than MRSA. MRSA differed considerably between countries with major international clones expanding or receding when compared to a 2006 survey. We provide evidence that a network approach of decentralised typing and visualisation of aggregated data using an interactive mapping tool can provide important information on the dynamics of S. aureus populations such as early signalling of emerging strains, cross-border spread and importation by travel.

Carbapenemase-producing Enterobacteriaceae in Europe: a survey among national experts from 39 countries, February 2013.

The spread of carbapenemase-producing Enterobacteriaceae (CPE) is a threat to healthcare delivery, although its extent differs substantially from country to country. In February 2013, national experts from 39 European countries were invited to self-assess the current epidemiological situation of CPE in their country. Information about national management of CPE was also reported. The results highlight the urgent need for a coordinated European effort on early diagnosis, active surveillance, and guidance on infection control measures.