Women Healthcare Mobile App-An Approach to Predict Early Stage of Cervical Cancer

Abstract

Most of the woman nowadays is ending up their life at middle age between 35-50 years, reason they are suffering from Cervical related cancer tumours. Many women are unaware of having cervical related cyst in the early stages. A survey was conducted on classifiers such as Decision Tree, Multilayer Perceptron, and Nave Bayes, with True Positive Rate, False Positive Rate, Precision, and Recall being measured, and an Android Mobile App was built to forecast the risk of having a Cervical associated cyst in its early stages. For training and testing the classifiers, we have used cervical dataset from University of California at Irvine Machine Learning Repository. Cervical Dataset consists of 858 records containing 32 attribute values and 4 diagnosis class value. For Cervical cancer prediction only 21 attributes and 1 biopsy class value is considered. The proposed Android Mobile App capable of predicting risk of a woman is affected by cervical related cyst in early stages. A normal woman can identify chances of having Cervical Cancer at finger tips with the proposed Android Mobile App.

Digital health interventions for cervical cancer care: A systematic review and future research opportunities

“…Inclusion of more machine learning based studies. The data analysis in this study showed that a number of machine learning algorithms were incorporated for cervical cancer screening purposes [32][33][34]36,44]. Nevertheless, very few machine learning algorithms have been used for this purpose.…”

“…Training supervised ML models and comparative evaluation of models [36][37][38] Establishing an ensemble prediction model based on the influencing factors for cervical cancer [39] Building A machine learning model to predict the recurrenceproneness for cervical cancer [40] Predicting the risk of cervical cancer by proposing a novel ensemble approach. A gene-assistance module is also included as an optional strategy to enhance the robustness of the prediction.…”

“…As per the plans for their future work, the inclusion of more datasets will not only improve the performance of the model but also may result in the selection of new features. Again, Naveen et al [36] In another study [34], author focused to the screening image evaluation where they designed and developed a deep Learning based image evaluation system for cervical prescreening using smartphones. The model was trained using transfer learning where MSCOCO dataset weights were used.…”

“…Use of data from different stages of cervical cancer. The analysis results showed that most of the researchers were concerned about the early stages of cervical cancer [32][33][34]36,44]. Though this is a praiseworthy approach which can aid in early prevention of cervical cancer and its associated complications, there are necessities for conducting research with data from the later stages of the cancer as well.…”