Backpropagation neural network-based survival analysis for breast cancer patients.

Autor: Pei, J. F., Zhang, J., Jin, D. C., Miao, B. L.
Předmět:
Zdroj: International Journal of Radiation Research; Jan2024, Vol. 22 Issue 1, p163-169, 7p
Abstrakt: Background: A substantial number of women throughout the world are affected with breast cancer, a dangerous and sometimes deadly condition. The creation of precise prediction models to determine the chance of survival in breast cancer patients has drawn increasing attention in recent years. The use of backpropagation neural networks (BPNNs) to forecast breast cancer patient survival is investigated in this study. Materials and Methods: A total of 198 patients with early breast cancer who were treated in our hospital were selected The control group received traditional breast cancer radical mastectomy and radiotherapy, and the experimental group received mastoscopy Adjuvant nipple-areola complex (NAC) modified radical mastectomy combined with prosthesis implantation and radiotherapy was used to compare the surgical conditions, postoperative complications, patient satisfaction and living standards in two groups. Results: The range of change was small, and the difference was statistically significant (P<0.05) ;in the laboratory group, patient dissatisfaction was noticeably raised over that in the standard group, and the discrepancy was politically sensitive (P<0.05); over that in the control group, the scores of all dimensions of survival quality were appreciably raised over that in the laboratory group, and the correlation was striking (P<0.05). Survival quality was greatly expanded in the experimental and postoperative groups at 3 and 6 months postoperatively before surgery, and the difference was statistically significant (P<0.05); Conclusion: The study demonstrates that BPNN-based predictive models can be useful tools for improving the accuracy of breast cancer survival rate prediction, thus aiding in more effective treatment planning and decision-making for breast cancer patients. [ABSTRACT FROM AUTHOR]
Databáze: Complementary Index