Cervical cancer is the fourth most prevalent disease in women globally as well as main cause of cancer- related mortality among women. The overall survival rates of cervical carcinoma patients are significantly impacted by the detection of recurrence. The disease is entirely treatable, though, if it is found in its early, precancerous levels. According to current figures, the death percentage a few decades ago was excessive. Women who have previously had treatment with CIN3 are most at riskof developing cervical cancer once more. The dataset includes information on the effects of various treatments on post- recurrence survival. The progression of a disease is detected in this research. Of the 4913 patients who were diagnosed withcervical cancer overall, 260 women with the disease suffered recurrences. The Data has been roughly categorized into 27 characteristics that are important for illness staging and therapy. Thevery important attributes selected are FIGO 2009, chemotherapy,radiotherapy, recurrence confirmed by a pathologist, etc. Five different models are trained to predict the progression. Logistic regression is finally predicted with a high level of accuracy of 96.3%. Two hundred and sixty patients with 27 attributes consistof a number of missing values. Research is implemented through the Orange platform. The missing values are balanced through Average and Most frequent data imputation method. Important attributes are selected through the ReliefF technique, and then disease progression is predicted by 5 different algorithms Logistic regression, Support vector machine, Naive Bayes, Stochastic gradient decent and neural network. Finally logistic regression predict the progression of a disease with highest accuracy of 96.3%.
Progression of Disease Forecasting in Recurrence Cervical Cancer Using Logistic Regression
2023-11-22
657126 byte
Conference paper
Electronic Resource
English
Discussion. Additive Logistic Regression
NTIS | 2000
|Urban Noise Analysis Using Multinomial Logistic Regression
British Library Online Contents | 2016
|Urban Noise Analysis Using Multinomial Logistic Regression
Online Contents | 2016
|