..

Revista de informática y biología de sistemas

Volumen 2, Asunto 4 (2009)

Artículo de investigación

Optimizing Number of Inputs to Classify Breast Cancer Using Artificial Neural Network

Bindu Garg, M.M. Sufian Beg and A.Q. Ansari

The Objective of this research work is to prove significant role of each attribute to decide breast cancer type using Computer Aided Diagnosis. One of major challenges in medical domain is the extraction of intelligible knowledge from medical diagnostic data in minimum time and cost This research shows that out of these attributes stated, some attributes can be ignored to decide the type Breast Cancer as if the number of inputs are less then it reduces the time and cost in analyzing the breast cancer. In this paper, significant role of each attribute is proved by experiment in matlab.

Artículo de investigación

MicroRNA Set : A Novel Way to Uncover the Potential Black Box of Chronic Heart Failure in MicroRNA Microarray Analysis

Guomin Shi, Qinghua Cui and Youyi Zhang

As the prime criminal among all the cardio-vascular diseases, chronic heart failure (CHF) is still far from being fully understood after decades of study by researchers. The booming bioinformatics studies, especially the microRNA (miRNA) microarray analysis, have significantly accelerated the uncovering of underlying mechanisms of human diseases. However, these miRNA researches mainly focus on single miRNA, paying less attention to the group characteristics of miRNAs, which may ignore the group characteristics of miRNAs. Here we introduce a novel miRNA set concept incorporation with a group analysis method of CHF miRNA microarray expression. Our results show great accordance with previous studies, and also reveal potential characteristics of miRNAs in CHF. Furthermore, this novel miRNA set approach may give us new insights into other diseases studies as well.

Artículo de revisión

Data Mining Techniques in High Content Screening: A Survey

Karol Kozak, Aagya Agrawal, Nikolaus Machuy and Gabor Csucs

Advanced microscopy and corresponding image analysis have evolved in recent years as a compelling tool for studying molecular and morphological events in cells and tissues. Cell-based High-Content Screening (HCS) is an upcoming technique for the investigation of cellular processes and their alteration by multiple chemical or genetic perturbations. The analysis of the large amount of data generated in HCS experiments represents a significant challenge and is currently a bottleneck in many screening projects. This article reviews the different ways to analyse large sets of HCS data, including the questions that can be asked and the challenges in interpreting the measurements. The main data mining approaches used in HCS are image descriptors, computations, normalization, quality control methods and classification algorithms.

Indexado en

arrow_upward arrow_upward