The developing of novel prophylactic and therapeutic vaccine candidates in the field of cancer immunology brought to very promising results against tumors, entitling full protection with reduced amount of the typical side effects of the actual conventional treatments. However, such treatments required a constant, life-long, administration procedure to keep protection. As both the period of protection and the relative number of administrations grow, the problem of finding the best administration protocol, in time and dosage, becomes more and more complex. Such a problem cannot be usually solved in in vivo experiments, as the costs in terms of time, money, and people would be prohibitive. We propose a hybrid approach that integrates machine learning and parallel genetic algorithms to enhance the research in silico of optimal administration protocols for a cancer vaccine. A neural network is used to improve both crossover and mutation operators. Preliminary results suggest that the use of such could bring to better administration protocols using a similar computational effort.

Combining Parallel Genetic Algorithms and Machine Learning to Improve the Research of Optimal Vaccination Protocols

Pennisi, Marzio;Russo, Giulia;Pappalardo, Francesco
Supervision
2018-01-01

Abstract

The developing of novel prophylactic and therapeutic vaccine candidates in the field of cancer immunology brought to very promising results against tumors, entitling full protection with reduced amount of the typical side effects of the actual conventional treatments. However, such treatments required a constant, life-long, administration procedure to keep protection. As both the period of protection and the relative number of administrations grow, the problem of finding the best administration protocol, in time and dosage, becomes more and more complex. Such a problem cannot be usually solved in in vivo experiments, as the costs in terms of time, money, and people would be prohibitive. We propose a hybrid approach that integrates machine learning and parallel genetic algorithms to enhance the research in silico of optimal administration protocols for a cancer vaccine. A neural network is used to improve both crossover and mutation operators. Preliminary results suggest that the use of such could bring to better administration protocols using a similar computational effort.
2018
9781538649756
agent based models; cancer; Genetic algorithms; machine learning; neural networks; optimal vaccine protocol; Computer Networks and Communications; Hardware and Architecture
File in questo prodotto:
File Dimensione Formato  
Combining Parallel Genetic Algorithms.pdf

accesso aperto

Tipologia: Versione Editoriale (PDF)
Dimensione 312.42 kB
Formato Adobe PDF
312.42 kB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.11769/334963
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 3
  • ???jsp.display-item.citation.isi??? 1
social impact