Downloads: 117 | Views: 218
Research Paper | Computer Science & Engineering | India | Volume 4 Issue 7, July 2015 | Rating: 7.1 / 10
Particle Swarm Optimization with Hidden Markov Model Prediction Approach for Anti-Cancer Drug Sensitivity
Daniel Nesa Kumar.C | Jayasree.P
Abstract: Identification of the best treatment methods for prediction of the anti drug becomes a core objective of correctness drug. Patient-specific examination facilitate finding of personality genomic characteristics designed for each patient, and thus be able to successfully forecast personality inherited hazard of infection and carry out adapted anti-cancer rehabilitation. Though all of the existing methods for patient-specific examination have been effectively applied and experimented for detection of tumor especially for anticancer drugs and it is performed based on the non-robust manners. To manage this problem a novel schema is introduced in this work to personalized cancer therapy to forecast and tumor to anticancer drugs. In the proposed predictive modeling method of tumor sensitivity to anti-cancer drugs has primarily focused on generating functions that map gene expressions and genetic mutation profiles to drug sensitivity. In the proposed PSO-HMM method based anti cancer drug sensitive prediction is performed based on the creation of the gene expression and gene profiles. In the proposed PSO-HMM method initially the tumor sympathy, gene expression point is primarily investigate all the way through calculation of distance function. The proposed work anti cancer drug sensitivity prediction is performed based on PSO-HMM with genomic characterizations. In order to solve the problem of C and D Particle Swarm Optimization with Hidden Markov model based prediction framework is trained on genomic and practical information with the intention of be able to appreciably progress the drug understanding forecast accurateness for collected samples from Cancer Cell Line Encyclopedia (CCLE) database. From the results it shows that the accuracy and prediction results of the proposed PSO-HMM method achieves higher tumor cancer detection results than the integrated approach, elastic net and random forest techniques.
Keywords: Drug sensitivity prediction, personalized cancer therapy, Particle Swarm Optimization PSO, Hidden Markov Models HMM
Edition: Volume 4 Issue 7, July 2015,
Pages: 1334 - 1338