Cervical cancer is the fourth most frequent cancer in women representing 7.9% of female cancers as stated by the World Health Organization. One of the leading causes of this has been the lack of effective early treatment. Therefore, early detection of cervical cancer is of utmost importance. The asymptomatic nature is the major challenge faced in the diagnosis of cervical cancer in the early stage. Machine learning has been of great help in many medical applications and can be used as a classifier in the early detection of the cancerous cells present in the cervix region of the uterus. In this paper, a survey and analysis of the various machine learning approaches that have been implemented for the diagnosis of cervical cancer is done. The survey paper draws a comparison of the various existing techniques for the prediction of cervical cancer using medical data and points out their advantages and shortcomings.

Survey of Cervical Cancer Prediction Using Machine Learning: A Comparative Approach

Tuberculosis is one of the top ten causes of death worldwide. Although this disease is curable and preventable, yet many new tuberculosis cases still occur especially in developing countries. Many low-income families cannot afford the medical diagnosis for tuberculosis. Therefore, this paper proposes an initial screening for tuberculosis infection using a data mining approach. In this paper, the initial screening is conducted using classification method developed from non-linear support vector machine and gradient evolution algorithm. Herein, the gradient evolution algorithm is able to find the best parameter setting for the support vector machine algorithm. The classification is performed based on some information which can be easily collected without medical test. The proposed algorithm is also compared with some other metaheuristic-based support vector machine algorithms. The experimental results show that the proposed algorithm has promising results shown by the small error rate. In addition, a C5.0 decision tree is employed to further analyze the rules of TB infection. The result reveals that people who are protected with vaccine still possible to be infected by tuberculosis, especially if there is direct contact with an active tuberculosis patient. Furthermore, people with low body mass index and low education has higher risk to get tuberculosis infection. The result of this study could help people conduct self-diagnosis for the tuberculosis infection before deciding to do a medical test.

An Initial Screening Method for Tuberculosis Diseases Using a Multi-objective Gradient Evolution-Based Support Vector Machine and C5.0 Decision Tree

Students' mobility and credit validation has been a concern for several years among higher education institutions in Ecuador, this process involves a huge amount of manual work due to the absence of an automatic system to measure the similarity between different course contents. In order to tackle this problem, we propose an approach to semantically compare the syllabi contents through text similarity methods. Such methods have been widely used in different domains, in this work we take the higher education institutions syllabi to the Text mining world and develop a method to compare their semantic contents. We propose an approach that uses pre-processing techniques, Latent Semantic Analysis for dimensionality reduction, text enrichment through the Wikipedia API and Google Engine, Support Vector Machine as classifier, and cosine similarity as similarity metric. Our results show that our method successfully measures similarity among higher education institutions syllabi and can be generalized to most Ecuadorian institutions.

A Text Mining Methodology to Discover Syllabi Similarities among Higher Education Institutions

Machine Learning, a form of Artificial Intelligence (AI) is infiltrating medical field and it is an era where machines can play an important role in health improvement. The idea behind AI in medical field is to enhance doctor's medical expertise. Medical Imaging is experiencing a drastic acceleration with rapid advancement in Machine Learning (ML) techniques. ML plays an important role in the medical imaging field including diagnosis, registration, segmentation and image database retrieval. ML can be applied to medical data repositories that are too large for the human brain to parse. ML promises to cut cost dramatically and deliver more accurate diagnosis than that of a trained physician. The patterns built from large clinical data warehouse can help researchers to draw conclusions and predict an event. This paper presents a comprehensive survey of the state-of-art work on Machine Learning in medical domain, which identifies the contributions of different methods, applications. In addition current issues and challenges are also discussed to identify promising areas of future research.

Machine Learning will Transfigure Medical Sector: A Survey

It has become a bottleneck to use Support Vector Machine(SVM) due to such problems as slow learning speed,large buffer memory requirement,low generalization performance and so on,which are caused by large-scale training sample set and outlier data immixed in the other class.Concerning these problems,this paper proposed a new reduction strategy for large-scale training sample set according to the analysis on the structure of the training sample set based on the point set theory.This new strategy gets the potential support vectors and removes the non-boundary outlier data immixed in the other class by using fuzzy clustering.That can greatly reduce the scale of the training sample set and improve the generalization performance by effectively avoiding over-learning caused by outlier data,and finally speed up learning rate without reducing the classification accuracy.

New reduction strategy of large-scale training sample set for SVM

Compared with ordinary data, medical data has its own characteristics. Such as mode of polymorphism, incomplete and longer timeliness. These characteristics brought a lot of difficulties on medical data of collection and processing, so the incremental learning method in the application of medical data is particularly critical. In this paper, Based on the support vector machines (SVM) proposed an incremental learning method that combined with fuzzy c-average and generalized KKT conditions. Through the filter of historical sample set and new sample that is invalid to reduce the training sample. So as to achieve rapid, incremental learning. Finally, the improved algorithm applied to the two standard medical database from UCI, which verify the improved algorithm advantage.

Application of support vector machines in medical data

The paper presents an improved text similarity algorithm and applies to the Clinical Decision Support System. It can improve the efficiency of decision making. After analyzing the disadvantages of the conventional TF-IDF algorithm and Cosine Similarity calculation, we propose an improvement to the text similarity algorithm. The proposed algorithm takes into full consideration the impact of semantically-similar keywords and textually-similar keywords on text similarity, therefore achieves improved accuracy. The function of this Clinical Decision Support System is to search for relevant biomedical articles and locate solutions.

Wenming Guo

Papers

Application of support vector machines in medical data

An improved text similarity algorithm research for clinical decision support system