Wyniki wyszukiwania - BazTech

1

Machine learning for analysis of gene expression data in fast- and slow-progressing amyotrophic lateral sclerosis murine models

Iadanza Ernesto, Fabbri Rachele, Goretti Francesco, Nardo Giovanni, Niccolai Elena, Bendotti Caterina, Amedei Amedeo

Biocybernetics and Biomedical Engineering

|

2022

|

Vol. 42, no. 1

273--284

EN

Amyotrophic lateral sclerosis is a fatal motor neuron disease characterised by degenerative changes in both upper and lower motor neurons. Current treatment options in the general cohort of ALS patients have only a minimal impact on survival. Only two approved medications are available today, just addressing the management of symptoms and supporting the respiration. In this work, gene expression data from genetically modified murine motor neurons have been analysed with machine learning techniques, with the scope of distinguishing between mice developing a fast progression of the disease, and mice showing a slower progression. Results showed high accuracy (above 80%) in all tasks, with peaks of accuracy for specific ones – such as distinguishing between fast and slow progression. In the above mentioned task the best performing algorithm reached an accuracy of 100%. This research group is currently working on three more investigations on data from mice, using similar approaches and methodology, focusing on thoracic and lumbar metabolomic data as well as microbiome data. We believe that, based on the findings in the murine models, machine learning could be used to discover ALS progression markers in humans by looking at features related to the immune response. This could pave the path for the discovery of druggable targets and disease biomarkers for homogeneous ALS patient subgroups.

2

A multi-layered incremental feature selection algorithm for adjuvant chemotherapy effectiveness/futileness assessment in non-small cell lung cancer

Naftchali R. E., Abadeh M. S.

Biocybernetics and Biomedical Engineering

|

2017

|

Vol. 37, no. 3

477--488

EN

Non-small cell lung cancer (NSCLC) is the most common type of lung cancer; and is one of the leading causes of death in the world. Surgery combined with chemotherapy is the recommended treatment for NSCLC. Since chemotherapy is an expensive treatment for either medical staff or patients suffering from pain, this study attempts to construct an intelligent predictive model to predict the adjuvant chemotherapy (ACT) effectiveness/ futileness in the patients, in order to help futile cases for unnecessary applications. There is a 2-step method: preprocessing and predicting. First a purposefully preprocessing tech-nique: chi-square test, SVM-RFE and correlation matrix, were employed in NSCLC gene expression dataset as a novel multi-layered feature selection method to defeat the curse of dimension and detect the chemotherapy target genes from tens of thousands features, based on which the patients can be classified into two groups, with NB classifier at second step. 10-Fold cross-validation was found with accuracy of 68.93% for 2 genes, TGFA (205015_s_at) and SEMA6C (208100_x_at), which is preferable compared to earlier studies, even though more than 2 input features are employed for the prediction. According to the results found in this study, one can concludes that the multi-layered feature selection approach has increased the classification accuracy in terms of finding the fitted patient for receiving ACT by reducing the number of features and has significant power to be used in medical datasets with small train samples and large number of features.

3

Methods of microarray data preprocessing for classification of tumors

Rubel T., Pawłowski Z., Zaremba K.

Biocybernetics and Biomedical Engineering

|

2005

|

Vol. 25, no. 4

37-57

EN

Gene arrays measure expression levels for thousands of genes simultaneously, providing a powerful tool for both basic research and clinical medicine. The aim of this paper was to present an optimal approach to preprocessing data from cancer microarray studies. The performance of different statistical methods used for the tumor classification was also compared. These methods include: the Bayes classifier, Fisher's classifier, minimum Euclidean and Mahalanobis distance classifiers and K-nearest neighbours classifier. The preprocessing algorithms and classification methods were applied to three datasets used for diagnosis of lymphoma, leukemia and lung cancer.