Automatic Diagnosis of Disc Herniation in Two-Dimensional MR Images with Combination of Distinct Features Using Machine Learning Methods

Hepatocellular Carcinoma (HCC) is the most common malignant liver tumor, being present in 70% of liver cancer cases. It usually evolves on the top of the cirrhotic parenchyma. The most reliable method for HCC diagnosis is the needle biopsy, which is an invasive, dangerous method. In our research, specific techniques for non-invasive, computerized HCC diagnosis are developed, by exploiting the information from ultrasound images. In this work, the possibility of performing the automatic diagnosis of HCC within B-mode ultrasound and Contrast-Enhanced Ultrasound (CEUS) images, using advanced machine learning methods based on Convolutional Neural Networks (CNN), was assessed. The recognition performance was evaluated separately on B-mode ultrasound images and on CEUS images, respectively, as well as on combined B-mode ultrasound and CEUS images. For this purpose, we considered the possibility of combining the input images directly, performing feature level fusion, then providing the resulted data at the entrances of representative CNN classifiers. In addition, several multimodal combined classifiers were experimented, resulted by the fusion, at classifier, respectively, at the decision levels of two different branches based on the same CNN architecture, as well as on different CNN architectures. Various combination methods, and also the dimensionality reduction method of Kernel Principal Component Analysis (KPCA), were involved in this process. These results were compared with those obtained on the same dataset, when employing advanced texture analysis techniques in conjunction with conventional classification methods and also with equivalent state-of-the-art approaches. An accuracy above 97% was achieved when our new methodology was applied.

Download Full-text

Probing Protein Allostery as a Residue-specific Concept via Residue Perturbation Maps

10.1101/355370 ◽

2018 ◽

Author(s):

Hamed S Hayatshahi ◽

Emilio Ahuactzin ◽

Peng Tao ◽

Shouyi Wang ◽

Jin Liu

Keyword(s):

Machine Learning ◽

Molecular Dynamics ◽

Two Dimensional ◽

Learning Methods ◽

Machine Learning Methods ◽

Dimensional Property ◽

Model Protein ◽

Perturbation Maps ◽

Specific Concept ◽

Perturbation Map

AbstractAllosteric regulation is a well-established phenomenon classically defined as conformational or dynamical change of a small number of allosteric residues of the protein upon allosteric effector binding at a distance. Here, we developed a novel approach to delineate allosteric effects in proteins. In this approach, we applied robust machine learning methods, including Deep Neural Network and Random Forest, on extensive molecular dynamics (MD) simulations to distinguish otherwise similar allosteric states of proteins. Using PDZ3 domain of PDS-95 as a model protein, we demonstrated that the allosteric effects could be represented as residue-specific properties through two-dimensional property-residue maps, which we refer as “residue perturbation maps”. These maps were constructed through two machine learning methods and could accurately describe how different properties of various residues are affected upon allosteric perturbation on protein. Based on the “residue perturbation maps”, we propose allostery as a residue-specific concept, suggesting all residues could be considered as allosteric residues because each residue “senses” the allosteric events through perturbation of its one or multiple attributes in a quantitatively unique way. The “residue perturbation maps” could be used to fingerprint a protein based on the unique patterns of residue perturbations upon binding events, providing a novel way to systematically describe the protein allosteric effects of each residue upon perturbation.Author SummaryAllostery is protein regulation at distance. A perturbation at one site of the protein could distantly affect another site. The residues involved in these sites are considered as allosteric residues. The allostery concept has been widely used to understand protein mechanisms and to design allosteric drugs. It is long believed only a small number of residues are allosteric residues. Here, we argue that all residues in a protein are allosteric residues. Upon the perturbation of the allosteric events, the different properties of each residue are affected at the distinct extend. We used hybrid models including molecular dynamics simulations and machine learning components to reveal that not only many properties of residues are affected upon ligand binding, but also each residue is affected through perturbation of its various properties, which makes the residue distinguishable from other residues. According to our findings in a model protein, we defined a “residue perturbation map” as a two-dimensional map that fingerprint a protein based on the extent of perturbation in different properties of all its residues in a quantitative fashion. This “residue perturbation map” provides a novel way to systematically describe the protein allosteric effects of each residue upon perturbation.

Download Full-text

Machine Learning Methods for Knee Feature Extraction from MR Images

2020 International Conference on Smart Electronics and Communication (ICOSEC) ◽

10.1109/icosec49089.2020.9215328 ◽

2020 ◽

Author(s):

Sujeet More ◽

Jimmy Singla

Keyword(s):

Machine Learning ◽

Feature Extraction ◽

Mr Images ◽

Learning Methods ◽

Machine Learning Methods

Download Full-text

Automatic diagnosis of melanoma using machine learning methods on a spectroscopic system

BMC Medical Imaging ◽

10.1186/1471-2342-14-36 ◽

2014 ◽

Vol 14 (1) ◽

Cited By ~ 13

Author(s):

Lin Li ◽

Qizhi Zhang ◽

Yihua Ding ◽

Huabei Jiang ◽

Bruce H Thiers ◽

...

Keyword(s):

Machine Learning ◽

Learning Methods ◽

Automatic Diagnosis ◽

Machine Learning Methods

Download Full-text

Knee osteoarthritis prediction on MR images using cartilage damage index and machine learning methods

2017 IEEE International Conference on Bioinformatics and Biomedicine (BIBM) ◽

10.1109/bibm.2017.8217734 ◽

2017 ◽

Cited By ~ 2

Author(s):

Yaodong Du ◽

Juan Shan ◽

Ming Zhang

Keyword(s):

Machine Learning ◽

Knee Osteoarthritis ◽

Cartilage Damage ◽

Damage Index ◽

Mr Images ◽

Learning Methods ◽

Machine Learning Methods

Download Full-text

Analysis and object markup of hyperspectral images for machine learning methods

Information Technology and Nanotechnology ◽

10.18287/1613-0073-2019-2391-309-317 ◽

2019 ◽

pp. 309-317

Author(s):

V P Gromov ◽

L I Lebedev ◽

V E Turlapov

Keyword(s):

Machine Learning ◽

Pearson Correlation ◽

Principal Component ◽

Hyperspectral Images ◽

Spectral Features ◽

Two Dimensional ◽

Learning Methods ◽

Machine Learning Methods ◽

Reference Signature ◽

Selection Of

The development of the nominal sequence of steps for analyzing the HSI proposed by Landgrebe, which is necessary in the context of the appearance of reference signature libraries for environmental monitoring, is discussed. The approach is based on considering the HSI pixel as a signature that stores all spectral features of an object and its states, and the HSI as a whole - as a two-dimensional signature field. As a first step of the analysis, a procedure is proposed for detecting a linear dependence of signatures by the magnitude of the Pearson correlation coefficient. The main apparatus of analysis, as in Landgrebe sequence, is the method of principal component analysis, but it is no longer used to build classes and is applied to investigate the presence in the class of subclasses essential for the applied area. The experimental material includes such objects as water, swamps, soil, vegetation, concrete, pollution. Selection of object samples on the image is made by the user. From the studied images of HSI objects, a base of reference signatures for classes (subclasses) of objects is formed, which in turn can be used to automate HSI markup with the aim of applying machine learning methods to recognize HSI objects and their states.

Download Full-text