چكيده به لاتين
Health funding in all countries is faced with increasing pressure, for this reason many scholars and statesmen tried to reduce healthcare costs and improve efficiency. In our country (Iran), according to existing sanctions, costs of materials and technologies in the field of health care have increased. Hence, umbilical cord blood stem cells banks as health-oriented institutions of this exception and the supply of laboratory materials as well as tanks for the storage of stem cells are under pressure. Due to the limited capacity of storage tanks of umbilical cord blood samples, cord blood banks priority is storing high quality samples. If the samples were not ideal in terms of quality at the time of transplantation cannot be used.
In this thesis, we tried to use data records in the Royan cord blood bank as the largest stem cell bank to identify the factors affecting quality parameters of umbilical cord blood from newborns and then considering the characteristics of each qualitative component, to determine with high probability and before carrying out experiments if the characteristics of applicant meets required characteristics or leads to termination of the contract. Finally, with regard to what was said to be a requirement for entry to the isolation, stem cells from umbilical cord blood sample will be qualify for storage in umbilical cord blood tanks. For this purpose, tools such as Artificial Neural Networks MLP and radial basis, Multivariate regression and decision tree C4.5 for predicting blood sample quality and Categories desirable samples for termination or frozen samples were used. The results show that using the MLP neural network model (The best performance among other models), caused decline termination and lost Royan cord blood bank samples collected opportunity costs last year amount to 35716240000 Rials.
Keywords: Prediction, Umbilical cord blood Banking, Data Mining, Electronic records
