Contribution of lung ultrasound in diagnosis of community-acquired pneumonia in the emergency department: a prospective multicentre study

Lung ultrasound (LUS) can help clinicians make a timely diagnosis of community-acquired pneumonia (CAP).ObjectivesTo assess if LUS can improve diagnosis and antibiotic initiation in emergency department (ED) patients with suspected CAP.DesignA prospective observational study.SettingsFour EDs.ParticipantsThe study included 150 patients older than 18 years with a clinical suspicion of CAP, of which 2 were subsequently excluded (incorrect identification), leaving 148 patients (70 women and 78 men, average age 72±18 years). Exclusion criteria included a life-threatening condition with do-not-resuscitate-order or patient requiring immediate intensive care.InterventionsAfter routine diagnostic procedure (clinical, radiological and laboratory tests), the attending emergency physician established a clinical CAP probability according to a four-level Likert scale (definite, probable, possible and excluded). An LUS was then performed, and another CAP probability was established based on the ultrasound result. An adjudication committee composed of three independent experts established the final CAP probability at hospital discharge.Primary and secondary outcome measuresPrimary objective was to assess concordance rate of CAP diagnostic probabilities between routine diagnosis procedure or LUS and the final probability of the adjudication committee. Secondary objectives were to assess changes in CAP probability induced by LUS, and changes in antibiotic treatment initiation.ResultsOverall, 27% (95% CI 20 to 35) of the routine procedure CAP classifications and 77% (95% CI 71 to 84) of the LUS CAP classifications were concordant with the adjudication committee classifications. Cohen’s kappa coefficients between routine diagnosis procedure and LUS, according to adjudication committee, were 0.07 (95% CI 0.04 to 0.11) and 0.61 (95% CI 0.55 to 0.66), respectively. The modified probabilities for the diagnosis of CAP after LUS resulted in changes in antibiotic prescriptions in 32% (95% CI 25 to 40) of the cases.ConclusionIn our study, LUS was a powerful tool to improve CAP diagnosis in the ED, reducing diagnostic uncertainty from 73% to 14%.Trial registration numberNCT03411824.

Clinical and immunohistochemical assessment of proliferative activity in uterine myoma

A tumor of the uterus, consisting of smooth muscles, is called myoma and fibroids, and in practice, the terms are used synonymously. Myoma of the uterus is a common disease in women of childbearing age. The frequency of its detection without regard to age is 2.45%. With age, its prevalence increases and reaches 8.31% by the age of 50. The estimate of the final probability of contracting uterine fibroids in the population throughout life is 9.7%. According to sectional data, uterine fibroids, including small nodes, occur in 20% of women. In 50% of cases, these tumors are manifested by clinically pronounced disorders that lead a woman to a doctor.

Ensemble of Multiple Classifiers for Multilabel Classification of Plant Protein Subcellular Localization

The accurate prediction of protein localization is a critical step in any functional genome annotation process. This paper proposes an improved strategy for protein subcellular localization prediction in plants based on multiple classifiers, to improve prediction results in terms of both accuracy and reliability. The prediction of plant protein subcellular localization is challenging because the underlying problem is not only a multiclass, but also a multilabel problem. Generally, plant proteins can be found in 10–14 locations/compartments. The number of proteins in some compartments (nucleus, cytoplasm, and mitochondria) is generally much greater than that in other compartments (vacuole, peroxisome, Golgi, and cell wall). Therefore, the problem of imbalanced data usually arises. Therefore, we propose an ensemble machine learning method based on average voting among heterogeneous classifiers. We first extracted various types of features suitable for each type of protein localization to form a total of 479 feature spaces. Then, feature selection methods were used to reduce the dimensions of the features into smaller informative feature subsets. This reduced feature subset was then used to train/build three different individual models. In the process of combining the three distinct classifier models, we used an average voting approach to combine the results of these three different classifiers that we constructed to return the final probability prediction. The method could predict subcellular localizations in both single- and multilabel locations, based on the voting probability. Experimental results indicated that the proposed ensemble method could achieve correct classification with an overall accuracy of 84.58% for 11 compartments, on the basis of the testing dataset.

Uncertainty query sampling strategies for active learning of named entity recognition task

Active learning approach is well known method for labeling huge un-annotated dataset requiring minimal effort and is conducted in a cost efficient way. This approach selects and adds most informative instances to the training set iteratively such that the performance of learner improves with each iteration. Named entity recognition (NER) is a key task for information extraction in which entities present in sequences are labeled with correct class. The traditional query sampling strategies for the active learning only considers the final probability value of the model to select the most informative instances. In this paper, we have proposed a new active learning algorithm based on the hybrid query sampling strategy which also considers the sentence similarity along with the final probability value of the model and compared them with four other well known pool based uncertainty query sampling strategies based active learning approaches for named entity recognition (NER) i.e. least confident sampling, margin of confidence sampling, ratio of confidence sampling and entropy query sampling strategies. The experiments have been performed over three different biomedical NER datasets of different domains and a Spanish language NER dataset. We found that all the above approaches are able to reach to the performance of supervised learning based approach with much less annotated data requirement for training in comparison to that of supervised approach. The proposed active learning algorithm performs well and further reduces the annotation cost in comparison to the other sampling strategies based active algorithm in most of the cases.

The influence of geological and technological factors on the probability of determining zones of residual oil reserves

This article is devoted to the probability maps have been constructed for predicting the zones of residual oil reserves using the example of deposits in Shaim region. The refinement of the previously presented algorithm [4] has been made, which helps with a fairly high degree of probability to quickly localize the residual oil reserves based on 2D modeling. In the process of work, the influence of many geological and technological parameters on the final probability map was established, the influence of the observation zone on the value of the correlation coefficient of the map of residual mobile oil reserves with the map of current mobile oil reserves based on geological and hydrodynamic modeling was established.

Evaluating susceptibility of karst dolines (sinkholes) for collapse in Sango, Tennessee, USA

Dolines or sinkholes are earth depressions that develop in soluble rocks complexes such as limestone, dolomite, gypsum, anhydrite, and halite; dolines appear in a variety of shapes from nearly circular to complex structures with highly curved perimeters. The occurrence of dolines in the studied karst area is not random; they are the results of geomorphic, hydrologic, and chemical processes that have caused partial subsidence, even the total collapse of the land surface when voids and caves are present in the bedrock and the regolith arch overbridging these voids is unstable. In the study area, the majority of collapses occur in the regolith (bedrock cover) that bridges voids in the bedrock. Because these collapsing dolines may result in property damage and even cause the loss of lives, there is a need to develop methods for evaluating karst hazards. These methods can then be used by planners and practitioners for urban and economic development, especially in regions with a growing population. The purpose of the project reported in this paper is threefold: (1) to develop a karst feature database, (2) to investigate critical indicators associated with doline collapse, and (3) to develop a doline susceptibility model for potential doline collapse based on external morphometric data. The study has revealed the presence of short range spatial dependence in the distribution of the dolines’ morphometric parameters such as circularity, the geographic orientation of the main doline axes, and the length-to-width doline ratios; therefore, geostatistics can be used to spatially evaluate the susceptibility of the karst area for doline collapse. The partial susceptibility estimates were combined into a final probability map enabling the identification of areas where, until now, undetected dolines may cause significant hazards.

Application of Probabilistic Techniques for the Development of a Prognosis Model of Stroke Using Epidemiological Studies

Automated medical diagnosis systems based on knowledge-oriented descriptions have gained momentum with the emergence of recent artificial intelligence techniques. The objective of this paper is to propose a design of a probabilistic model for the prevention of stroke based on the most outstanding risk factors associated with this pathology. The authors gather probabilistic technologies to develop a new clinical support decision-making model. This development is part of a future system that aims to improve health-quality and prevent strokes. The Naïve Bayes model is proposed to calculate the probability of suffering a stroke in the future, based on epidemiological data. Due to a new design, the model is capable to determine the probability of suffering a stroke given some risk factors. The proposed model allows to calculate the final probability of suffering a specific disease for the preventive prognosis of the stroke based on risk factors. Our model enables query the probability of suffering a stroke giving as parameter the presence or absence of a specific indication, also setting if the indication can take several values with its presence, degree or value. With the obtained results the physician will be able to promote patients healthy living habits in order to prevent future stroke events.