Search

Tuberculosis is a respiratory disease that affects lungs and it is caused by the bacillus Mycobacterium tuberculosis (MTB), which is spread when people who are sick with tuberculosis expel bacteria into the air by coughing. Before the coronavirus (COVID-19) pandemic, tuberculosis was the leading cause of death from infectious agents even ranking above HIV/AIDS. Growing MTB on solid medium is the main diagnostic reference method. In addition, this is the standard method for identifying the bacterial resistance profile. However, the waiting period for the results is long, during which time a patient can be highly contagious. In turn, delaying treatment for tuberculosis can increase disease severity, but treating without diagnostic confirmation can lead to bacterial resistance, a higher rate of adverse events, and higher costs. Therefore, pulmonary tuberculosis disease must be rapidly diagnosed by a costeffective method. This paper proposes models for the detection of pulmonary tuberculosis by using different artificial intelligence techniques. These models can be used to support decision making of doctors and are intended to identify whether a patient has tuberculosis or not. In particular, four supervised learning techniques are used (neural networks, decision trees, and two ensemble methods). Each model allows predicting a positive or negative diagnosis of pulmonary tuberculosis based on previously recorded diagnostic variables taken from patients in Cali, Colombia. According to the results, the Extra Trees method reaches the highest accuracy compared to the other techniques used for the prediction of pulmonary tuberculosis with an area under the ROC curve of 95,63%. [ABSTRACT FROM AUTHOR]

In this paper, a new protein structure prediction method is presented. Unlike the current methods, we use an approach based on supervised classification algorithms during the protein structure prediction. The accuracy of the proposed method was compared to traditional methods such as LFF (Local Feature Frequency) when using the SCOP 2.05 dataset. The results indicate that there is a significant difference between these two methods. The proposed method reaches accuracy values of 92.13%, 96.32%, 93.05%, and 76.35%, at class, fold, superfamily, and family levels, respectively, and the LFF method reaches accuracy values of 85.90%, 90.54%, 79.85% and 67.38%, for the same structural levels. [ABSTRACT FROM AUTHOR]

Copyright of Revista Educación en Ingeniería is the property of Asociacion Colombiana de Facultades de Ingenieria and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

In this paper, a new program for protein secondary structure prediction is proposed. The program, which is called NN-Pred, allows multiple sequences to be submitted and it returns predictions from five secondary structure prediction algorithms. In addition, NN-Pred calculates a consensus prediction, which is based on a neural network strategy that is used in this paper to improve the prediction accuracy. NN-Pred was obtained by using a methodology called consensus strategy, which tries to make a better prediction by integrating some of the most remarkable existing techniques. The NN-Pred program provides a three-state (alpha-helix, beta-sheet, and other) prediction of secondary structure. According to the test sets, the prediction accuracy of NN-Pred is at least 70%, surpassing most of the existing methods. The experiments showed that neural networks can be used as a consensus strategy to producing accurate models for protein secondary prediction. [ABSTRACT FROM AUTHOR]