Second Workshop on Data Mining for Healthcare Management

Held in conjunction with
The 15th Pacific-Asia Conference on Knowledge Discovery and Data Mining
24-27 May 2011 (Tue-Fri) - Shenzen, China

*** Invited Talk *** Dr. Vipin Gopal, Director, Clinical Analytics, Humana. Predictive Modeling for Care Management: Impact on Outcomes, Costs and Engagement
Abstract: Advanced analytical methods, including data mining and predictive modeling, are emerging as critical components in addressing various topics in the effective delivery of care. This talk will give an overview of the efforts at Humana, a Fortune 100 health benefits company, in developing advanced analytic solutions for positively impacting care management. The focus will primarily be on topics around effective identification and engagement of the .right members for the right care at the right time.. Among others, we will look at the readmissions problem, that of patients getting readmitted in the hospital within 30 days of discharge from an initial admission. A recent New England Journal of Medicine article (Jencks, Williams and Coleman, 2009) observed that about 1 in every 5 admissions in the traditional Medicare program (a health benefits program administered by the US Government) had readmissions, costing a total of about $17.4B in 2004. Predictive modeling solutions that help identify patients who are at high risk for readmissions, and subsequent interventions to reduce the overall readmissions rates, will be discussed. Application of advanced analytics to design preventive care solutions, enhanced management of chronic conditions and better engagement of members, resulting in improved clinical outcomes and lower costs, will also be discussed.
M Saravanan, S Shanthi and S Shalini. Usage of Mobile phones for Personalized Healthcare Solutions
Abstract: One of the greatest hurdles in providing the appropriate healthcare is the availability of proper information at the point of individual.s care.  Mobile phone-based health solutions can bridge this gap and can support with right information at the right time. In order to overcome some of the prevalent issues and hence provide the necessary healthcare, we here introduce one such mobile based system known as the Personalized Mobile Health Service System for Individual.s Healthcare which caters to the specific needs of the user without the constraint on mobility. This system helps in guiding the user with regard to the food they consume, the precautionary measures to be taken in case of any ailments, and when they travel to a new location etc. This system also proves to be supportive in situations when the user is in a traumatic condition suffering alone. The main advantage of the system is that it will keep updating the details to the user on a regular basis.
Boyu Wang and Feng Wan. Robust Learning of Mixture Models and Its Application on Trial Pruning for EEG Signal Analysis
Abstract: This paper presents a novel method based on deterministic annealing to circumvent the problem of the sensitivity to atypical observations associated with the maximum likelihood (ML) estimator via conventional EM algorithm for mixture models. In order to learning the mixture models in a robust way, the parameters of mixture model are learned by trimmed likelihood estimator (TLE), and the learning process is controlled by temperature based on the principle of maximum entropy. Moreover, we apply the proposed method to the single-trial (electroencephalography) EEG classification task. The motivation of this work is to eliminate the negative effects of artifacts in EEG data, which usually exist in real-life environments, and the experimental results demonstrate that compared with conventional EM algorithm, the proposed method can successfully detect the outliers and therefore achieve more reliable result.
Wei Gu, Baijie Wang and Xin Wang. An Integrated Approach to Multi-Criteria-based Health Care Facility Location Planning
Abstract: Optimal location of health care facilities is critical to the success of health care services. Given its importance, this is an active research topic in health informatics, operational research and GIS. This paper presents an integrated approach to health care facility planning whereby the methods from three research topics are combined. The integrated approach is applied in order to solve preventive health care facility location planning problems. In this approach, a new health accessibility estimation method is developed in order to capture the current characteristics of preventive health care services. Based on this, the preventive health care facility location planning problem is formalized as a multi-criteria facility location model. A new algorithm is proposed in order to solve the model. Experiments on synthetic datasets and on the Alberta breast cancer screening program data are conducted and the results support our analysis.
Chaveevan Pechsiri, Sumran Painual and Uraiwan Janviriyasopak. Medicinal Property knowledge Extraction from Herbal Documents for Supporting Question Answering System
Abstract: The aim of this paper is to automatically extract the medicinal properties of an object, especially an herb, from technical documents as knowledge sources for health-care problem solving through the question-answering system, especially What-Question, for disease treatment. The extracted medicinal property knowledge is based on multiple simple sentence or EDUs (Elementary Discourse Units). There are three problems of extracting the medicinal property knowledge: the herbal object identification problem, the medicinal property identification problem for each object and the medicinal property boundary determination problem. We propose using NLP (Natural Language Processing) with statistical based approach to identify the medicinal property and also with machine learning technique as Na´ve Bayes with verb features for solving the boundary problem. The result shows successfully the medicinal property extraction of the precision and recall of 86% and 77%, respectively, along with 87% correctness of the boundary determination.