Biomarker Expression in Multifocal Vulvar High-Grade Squamous Intraepithelial Lesions

In patients with high-grade squamous intraepithelial lesion (HSIL) of the vulva, the presence of multiple lesions, called multifocal HSIL, is common. The aim of this exploratory study was to investigate biomarker expression profiles in multifocal HSIL. In total, 27 lesions from 12 patients with high-risk human papillomavirus (HPV)-positive multifocal HSIL were tested for HPV genotype, expression of p16INK4a and Ki-67, and DNA methylation of six genes. HPV16 was found most commonly in 21 (77.8%) HSILs. In two (16.4%) patients, HPV genotype differed between the lesions. All lesions demonstrated diffuse p16INK4a staining, of which three (11.1%) were combined with patchy staining. One patient (8.3%) demonstrated markedly different DNA methylation levels between lesions. Generally, heterogeneity in methylation profiles was observed between different patients, even when other biomarkers showed similar expression. In conclusion, this study is the first to demonstrate heterogeneity of individual lesions in patients with multifocal HSIL. The studied biomarkers have the potential to refine prognostic and predictive diagnostics. Future prospective, longitudinal studies are needed to further explore the potential of a biomarker profile for management of patients with multifocal HSIL.

Analysis of Predictive Maintenance for Tunnel Systems

Smart and automated maintenance could make the system and its parts more sustainable by extending their lifecycle, failure detection, smart control of the equipment, and precise detection and reaction to unexpected circumstances. This article focuses on the analysis of data, particularly on logs captured in several Czech tunnel systems. The objective of the analysis is to find useful information in the logs for predicting upcoming situations, and furthermore, to check the possibilities of predictive diagnostics and to design the process of predictive maintenance. The main goal of the article is to summarize the possibilities of optimizing system maintenance that are based on data analysis as well as expert analysis based on the experience with the equipment in the tunnel. The results, findings, and conclusions could primarily be used in the tunnels; secondarily, these principles could be applied in telematics and lead to the optimization and improvement of system sustainability.

Acute kidney injury in cardiac surgery: preoperative predictive diagnostics

<p>Acute kidney injury (AKI) and its delayed diagnosis often lead to an increase in the number of patients with chronic kidney disease. There is no denying the importance of studying the pathogenetic mechanisms of AKI, timely predictive identification of patients at high risk of CSA-AKI, as well as the need to search for and improve perioperative strategies to prevent CSA-AKI. The development of new approaches regarding predictive diagnostics of AKI and their widespread introduction into a wide clinical practice will improve the prognosis and survival of these patients.<br />The modern diagnostic continuum of AKI considers risk factors as pre-existing conditions against which adverse factors of the perioperative period are realised. Risk factors for AKI in cardiac surgery are divided into two categories: 1) patient-dependent and 2) operations-associated or modifiable risk factors for the development of AKI, to some extent secondary to iatrogenic effects (adverse factors of the perioperative period). A clear understanding of the significance of these factors regarding the development of AKI in cardiac surgery patients allows us to form risk scales for predicting CSA-AKI in the postoperative period. This review presents the following work, which is a milestone for the predictive diagnosis of AKI in cardiac surgery: model Association of Nephrology and the Association of Anaesthesiologists and Reanimatologists of Russia, the scale EuroSCORE II, the STS Score, Score Mehta, ‘risk index perioperative renal dysfunction/failure’, S. Aronson et al., scale S.Y. Ng et al., model K. Birnie et al.<br />The use of predictive models for the predictive diagnosis of CSA-AKI is an important strategy for identifying high-risk patients. This approach allows active application of preventive strategies regarding AKI in routine clinical practice. It also has distinct advantages regarding conducting cohort clinical studies of new renoprotective technologies. To date, there is no ‘gold standard’ scale for predicting the risk of cardiac AKI. The authors propose consideration of their application as a weighted ‘Solomon’ solution, according to a specific clinical situation.</p><p>Received 10 July 2020. Revised 9 September 2020. Accepted 10 September 2020.</p><p><strong>Funding:</strong> The study did not have sponsorship.</p><p><strong>Conflict of interest:</strong> Authors declare no conflict of interest.</p><p><strong>Author contributions</strong><br />Conception and design: N.O. Kamenshchikov, Yu.K. Podoksenov, M.L. Diakova<br />Drafting the article: N.O. Kamenshchikov, M.L. Diakova, A.M. Boyko<br />Critical revision of the article: M.L. Diakova, Yu.K. Podoksenov<br />Final approval of the version to be published: N.O. Kamenshchikov, Yu.K. Podoksenov, M.L. Diakova, A.M. Boyko, B.N. Kozlov</p>

Gut microbiome-based supervised machine learning for clinical diagnosis of inflammatory bowel diseases

Despite the availability of various diagnostic tests for inflammatory bowel diseases (IBD), misdiagnosis of IBD occurs frequently, and thus there is a clinical need to further improve the diagnosis of IBD. As gut dysbiosis is reported in IBD patients, we hypothesized that supervised machine learning (ML) could be used to analyze gut microbiome data for predictive diagnostics of IBD. To test our hypothesis, fecal 16S metagenomic data of 729 IBD and 700 non-IBD subjects from the American Gut Project were analyzed using five different ML algorithms. Fifty differential bacterial taxa were identified (LEfSe: LDA > 3) between the IBD and non-IBD groups, and ML classifications trained with these taxonomic features using random forest (RF) achieved a testing AUC of ~0.80. Next, we tested if operational taxonomic units (OTUs), instead of bacterial taxa, could be used as ML features for diagnostic classification of IBD. Top 500 high-variance OTUs were used for ML training and an improved testing AUC of ~0.82 (RF) was achieved. Lastly, we tested if supervised ML could be used for differentiating Crohn's disease (CD) and ulcerative colitis (UC). Using 331 CD and 141 UC samples, 117 differential bacterial taxa (LEfSe: LDA > 3) were identified, and the RF model trained with differential taxonomic features or high-variance OTU features achieved a testing AUC > 0.90. In summary, our study demonstrates the promising potential of artificial intelligence via supervised ML modeling for predictive diagnostics of IBD using gut microbiome data.

Belt conveyor predictive diagnostics based on the monitoring of the drive motor electrical parameters

The study focuses on the main malfunctions of the mechanical components transmitting rotational moment to belt conveyor pulley and the possibility of functional diagnostics of their technical condition. The results of electrical values of the drive were obtained for the modeled emergency modes of the conveyor operation in the Matlab Simulink programming environment.

Innovative Solutions for Partial Discharge Measurements

The article discusses systems developed using an innovative approach to measuring partial discharges, their advantages over traditional methods of monitoring cable lines and electrical equipment, as well as the principles of their operation. The design of the systems is based on a special broadband sensor-antenna which provides remote recording of both partial discharge pulses and the phase of alternating current flowing in the object of measurement without taking the equipment under test out of operation. Due to the wireless nature and system compact dimensions, partial discharge measurements can perform in previously inaccessible conditions, significantly expanding the capabilities of this diagnostic method. The partial discharge measurements are an up-to-date method for predictive diagnostics of cable systems and are necessary for the condition-based maintenance.