Recently, the use of radioisotopes in medical applications, such as photon or positron emission tomography, has received considerable attention. In this research, TALYS code was used to predict the cross section of distinct nuclear reaction and to evaluate the excitation function of 86Sr(p,n)86Y reaction for production of Yttrium-86 (86Y) radioisotope. Based on the properties of the Karaj cyclotron, the optimal proton energy range from 12 to 16 MeV, the 86Sr target thickness and the 86Y production yield were estimated. Then, the acquired data were optimized via Artificial Neural Network (ANN) to investigate the sensitivity to projectile energy, time, beam energy and current. At 14 MeV, the ANN demonstrated a higher produced 86Y while reducing contamination compared to a poorly trained neural network under incorrect conditions. Maximum 86Y production with minimal contamination reduced the purification error. This technique can influence the control of radioactive contamination during the production of medical radioisotopes to reduce the absorbed dose imposed on the patient.   

Keywords: Yttrium-86 radioisotope production; Contamination; Nuclear cross section; Artificial neural network; Modified weights; Iteration numbers.

Dr. Abdallah Khorshidi is an expert in medical radiation physics and nuclear engineering and works as a freelance researcher to explore new items in the field of nuclear science. He has conducted numerous research projects for master’s students as part of their theses and has published various articles in international journals. These can be briefly viewed on the ORCID website:

 https://orcid.org/0000-0002-6674-8789 

In addition, he has served as a reviewer for numerous manuscripts in various specialist journals, thus supporting the editorial team of international journals. These can be concisely viewed on the Publons website:

http://www.researcherid.com/rid/P-2983-2017