Influence of microbiome species in hard-to-heal wounds on disease severity and treatment duration

By Dagmar Chudobová, Kristyna Cihalova, Roman Guráň, S. Dostalova, Kristyna Smerkova, Radek Veselý, J. Gumulec, M. Masařík, Z. Heger, V. Adam and 1 more. 2015

13 citations recorded by the source corpus.

Background Infections, mostly those associated with colonization of wound by different pathogenic microorganisms, are one of the most serious health complications during a medical treatment. Therefore, this study is focused on the isolation, characterization, and identification of microorganisms prevalent in superficial wounds of patients (n = 50) presenting with bacterial infection. Methods After successful cultivation, bacteria were processed and analyzed. Initially the identification of the strains was performed through matrix-assisted laser desorption/ionization time-of-flight mass spectrometry based on comparison of protein profiles (2–30 kDa) with database. Subsequently, bacterial strains from infected wounds were identified by both matrix-assisted laser desorption/ionization time-of-flight mass spectrometry and sequencing of 16S rRNA gene 108. Results The most prevalent species was Staphylococcus aureus (70%), and out of those 11% turned out to be methicillin-resistant (mecA positive). Identified strains were compared with patients’ diagnoses using the method of artificial neuronal network to assess the association between severity of infection and wound microbiome species composition. Artificial neuronal network was subsequently used to predict patients’ prognosis (n = 9) with 85% success. Conclusions In all of 50 patients tested bacterial infections were identified. Based on the proposed artificial neuronal network we were able to predict the severity of the infection and length of the treatment.

Subfields: Rehabilitation, Surgery

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Opportunities and obstacles for deep learning in biology and medicine (2017, background): “A standard multi-layer perceptron (MLP) was able to classify wound severity from microbial species present in the wound [251].”