Cancer treatment is extremely aggressive and, in addition to causing considerable discomfort, can lead to death. Therefore, identifying aspects related to treatment assertiveness may be efficient for reducing the mortality rate of cancer patients. This paper seeks to identify the prognosis of cancer treatment survival through hybrid techniques based on the autonomous fuzzification process and artificial neural networks. The public dataset on cancer mortality is the source for conducting treatment assertiveness rating tests. The hybrid model had its results compared to other models present in the pattern classification literature with superior accuracy and identification of people likely to survive treatment (90.46%), and the fuzzy rules obtained with the execution of the model corroborate the high assertiveness of the model, even surpassing state of the art for the theme.

Knowledge extraction about patients surviving breast cancer treatment through an autonomous fuzzy neural network

De Campos Souza, P. V.;
2020-01-01

Abstract

Cancer treatment is extremely aggressive and, in addition to causing considerable discomfort, can lead to death. Therefore, identifying aspects related to treatment assertiveness may be efficient for reducing the mortality rate of cancer patients. This paper seeks to identify the prognosis of cancer treatment survival through hybrid techniques based on the autonomous fuzzification process and artificial neural networks. The public dataset on cancer mortality is the source for conducting treatment assertiveness rating tests. The hybrid model had its results compared to other models present in the pattern classification literature with superior accuracy and identification of people likely to survive treatment (90.46%), and the fuzzy rules obtained with the execution of the model corroborate the high assertiveness of the model, even surpassing state of the art for the theme.
2020
978-1-7281-6932-3
File in questo prodotto:
Non ci sono file associati a questo prodotto.

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/11582/340988
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
social impact