scholarly journals Recognition of Pharmacological Bi-Heterocyclic Compounds by Using Terahertz Time Domain Spectroscopy and Chemometrics

Sensors ◽  
2019 ◽  
Vol 19 (15) ◽  
pp. 3349 ◽  
Author(s):  
Maciej Roman Nowak ◽  
Rafał Zdunek ◽  
Edward Pliński ◽  
Piotr Świątek ◽  
Małgorzata Strzelecka ◽  
...  
Keyword(s):  
Heterocyclic Compounds ◽  
Chemical Compounds ◽  
Functional System ◽  
Support Vector ◽  
Optimal Parameters ◽  
Terahertz Time Domain Spectroscopy ◽  
Kernel Support Vector Machine ◽  
Inflammatory Drugs ◽  
Living Organisms

In this study, we presented the concept and implementation of a fully functional system for the recognition of bi-heterocyclic compounds. We have conducted research into the application of machine learning methods to correctly recognize compounds based on THz spectra, and we have described the process of selecting optimal parameters for the kernel support vector machine (KSVM) with an additional `unknown’ class. The chemical compounds used in the study contain a target molecule, used in pharmacy to combat inflammatory states formed in living organisms. Ready-made medical products with similar properties are commonly referred to as non-steroidal anti-inflammatory drugs (NSAIDs) once authorised on the pharmaceutical market. It was crucial to clearly determine whether the tested sample is a chemical compound known to researchers or is a completely new structure which should be additionally tested using other spectrometric methods. Our approach allows us to achieve 100% accuracy of the classification of the tested chemical compounds in the time of several milliseconds counted for 30 samples of the test set. It fits perfectly into the concept of rapid recognition of bi-heterocyclic compounds without the need to analyse the percentage composition of compound components, assuming that the sample is classified in a known group. The method allows us to minimize testing costs and significant reduction of the time of analysis.

Classification of the fragrant styles and evaluation of the aromatic quality of flue-cured tobacco leaves by machine-learning methods

2016 ◽  
Vol 14 (06) ◽  
pp. 1650033 ◽  
Author(s):  
Li Gu ◽  
Lichun Xue ◽  
Qi Song ◽  
Fengji Wang ◽  
Huaqin He ◽  
...  
Keyword(s):  
Evaluation System ◽  
Chemical Compounds ◽  
Support Vector ◽  
Tobacco Leaves ◽  
Machine Learning Methods ◽  
Online Tools ◽  
Svm Algorithm ◽  
Assessment Performance

During commercial transactions, the quality of flue-cured tobacco leaves must be characterized efficiently, and the evaluation system should be easily transferable across different traders. However, there are over 3000 chemical compounds in flue-cured tobacco leaves; thus, it is impossible to evaluate the quality of flue-cured tobacco leaves using all the chemical compounds. In this paper, we used Support Vector Machine (SVM) algorithm together with 22 chemical compounds selected by ReliefF-Particle Swarm Optimization (R-PSO) to classify the fragrant style of flue-cured tobacco leaves, where the Accuracy (ACC) and Matthews Correlation Coefficient (MCC) were 90.95% and 0.80, respectively. SVM algorithm combined with 19 chemical compounds selected by R-PSO achieved the best assessment performance of the aromatic quality of tobacco leaves, where the PCC and MSE were 0.594 and 0.263, respectively. Finally, we constructed two online tools to classify the fragrant style and evaluate the aromatic quality of flue-cured tobacco leaf samples. These tools can be accessed at http://bioinformatics.fafu.edu.cn/tobacco .

scholarly journals Electromyographic Grasp Recognition for a Five Fingered Robotic Hand

2012 ◽  
Vol 2 (1) ◽  
pp. 1
Author(s):  
Nayan M. Kakoty ◽  
Mantoo Kaiborta ◽  
Shyamanta M. Hazarika
Keyword(s):  
Daily Living ◽  
Wavelet Decomposition ◽  
Recognition Rate ◽  
Support Vector ◽  
Robotic Hand ◽  
Radial Basis Function Kernel ◽  
Kernel Support Vector Machine ◽  
Average Recognition Rate ◽  
Electromyographic Signals

This paper presents classification of grasp types based on surface electromyographic signals. Classification is through radial basis function kernel support vector machine using sum of wavelet decomposition coefficients of the EMG signals. In a study involving six subjects, we achieved an average recognition rate of 86%. The electromyographic grasp recognition together with a 8-bit microcontroller has been employed to control a five<br />fingered robotic hand to emulate six grasp types used during 70% daily living activities.<br /><br />

Sign in / Sign up

Export Citation Format

Share Document