ISSN: 2321-7782 (Online)

Volume 4, Issue 2, February 2016

International Journal of Advance Research in
Computer Science and Management Studies
Research Article / Survey Paper / Case Study
Available online at: www.ijarcsms.com

An Overview of Data Mining Applications in Healthcare

Ravleen Singh1 Dr. Tariq Hussain Sheikh2
Lecturer G.D.C. Poonch - India Lecturer G.D.C.Mendhar - India

Abstract: There is a wealth of data available within the healthcare systems. However, there is a lack of effective
psychoanalysis tools to discover hidden relationships and trends in data. Without data mining it is difficult to understand
clearly the full potential of data collected within healthcare organization as data under analysis is exceptionally large, highly
dimensional, distributed and not definite. The objective of this study is to explore new and emerging areas of data mining
techniques used in healthcare management. This paper aims to make a detailed study report of different types of data mining
applications in the healthcare area and to minimize the complexity of the study of the healthcare data understanding.

Keywords: Data Mining.

I. INTRODUCTION

Data mining provides the methodology and technology to transform massive amount of data into useful information for
decision making. It is defined as the process of data selection and exploration and building models using vast data stores to
[1]
uncover unknown patterns .The investigative objective of data mining is to organize the data, text and images of huge data
into knowledge based or information using dispensation of computers. Data mining algorithms applied in healthcare sector play
[2].
a significant role in prediction and diagnosis of the diseases The large numbers of data mining applications are found in the
medical related areas such as Medical device industry, Pharmaceutical Industry and Hospital Management. The research
objective of data mining in Health Care System is to generate an automated tool to perceive, ascertain voluminous data and
organize into useable information i.e. Knowledge discovery data (KDD).

Fig 1.Data Mining Architecture

The knowledge discovery is an interactive process, consisting by developing an understanding of the application domain,
[3].
selecting and creating a data set, preprocessing, data transformation Data mining tools answer the question that traditionally
[4].
was a time consuming and too complex to resolve They prepare databases for finding predictive information. Data mining
tasks are Association Rule, Patterns, Classification, Prediction and Clustering. Most common modeling objectives are
[5].
classification and prediction The branch of computer science which is more actively and efficiently involved in medical
sciences is Artificial Intelligence. Any computer program that helps experts in making healthcare decision comes under the
domain of healthcare decision support system.

2016, IJARCSMS All Rights Reserved 131 | P a g e

Ravleen et al., International Journal of Advance Research in Computer Science and Management Studies
Volume 4, Issue 2, February 2016 pg. 131-138

Fig 2.Data Mining Life Cycle

II. DATA MINING METHODOLOGY

[6],[7] .
Data mining procedure performs following maneuvers proposed by the Cross-Industry Standard Process The main
motto acknowledged by business understanding step lead towards the emergence success of data mining project. The data
understanding step furtherance into preliminary data collection, data description, data exploration and verification of data
quality. After the identification of data in data preparation step, it should be selected, clean, built in the desired format and
formatted. The modeling step diagnosis the actual data analysis into modeling stage. There are different data mining softwares
viz. cluster analysis, discriminant analysis, regression analysis, neural analysis network, decision tree, link analysis and
association analysis. The suitable model is applied to data types. Further the evaluation step compares the data model resulting
from any data mining model by using any common standard techniques such as lift chart, profit chart and diagnostic
classification chart. Finally deployment resulted which is related to the actual execution and operationalization of data mining
model.

III. PROCESS OF KNOWLEDGE DISCOVERY AND DATA MINING

Fig 2.Data mining model and task

C1.Predictive data modeling is an important data mining task to determine future data states on the basis of past and
current values. Predictions may be made on the basis of regression, time series analysis, or some other approaches.

Time series analysis is to identify the value or attributes over a time period usually at evenly spaced time intervals i.e.
The attribute value can be generated on a daily or hourly basis depending upon the state of the ailing which is used to foresee
the future analysis.

Regression is a method to map target data using some known type of function. It deals with estimation of an output value
based on input values.

Classification is the task of generalizing known structure to apply to new data. Classification classifies a data item into
one of several predefined classes. A set of classification rules is generated from the classification model, based on the features
of the data in the training set, which can be used to classify future data and develop a better understanding of each class in the
database.

C2. Visualization techniques are useful methods of discovering patterns in a medical data set. Scatter diagrams in a
Cartesian plane of two interesting medical attributes can be used to identify interesting subsets of medical data sets. For

2016, IJARCSMS All Rights Reserved ISSN: 2321-7782 (Online) 132 | P a g e

Ravleen et al., International Journal of Advance Research in Computer Science and Management Studies
Volume 4, Issue 2, February 2016 pg. 131-138
example, for heart patients interesting subsets can be found with respect to blood sugar (fasting). Once interesting subsets are
obtained, we may use other data mining techniques on these subsets to discover further knowledge.

Association rule (Dependency modeling) Searches for relationships between variables. For example a supermarket
might gather data on customer purchasing habits.

Clustering is the task of discovering groups and structures in the data that are in some way or another "similar", without
using known structures in the data.

Summarization providing a more compact representation of the data set, including visualization and report generation

Link analysis Form of network analysis that examines the associations between objects Link. Classification provides
category of an object, not just based on its features, but also on connections in which it takes part, and features of objects
connected with certain path [8].

IV. DATA MINING TECHNIQUES

There are two basic design of data mining: hypothesis testing and knowledge discovery [9]. Hypothesis testing is a top-
down approach that is used when a confirmation or a rejection of an already defined hypothesis is needed. The knowledge
discovery is a bottom-up approach and it is used when we want to find something that we do not know from searching available
data. In Data mining the most important technique which is used is Knowledge Discovery in Database(KDD).KDD has different
steps like Data cleaning, Data integration, Data selection, Data transformation, Data mining, Pattern evaluation, Knowledge
presentation etc. The different types of techniques used in Data mining project include Decision tree, Bayesian networks, Naive
bayes, Neural networks etc.

D1. Decision tree-It is the most frequently used techniques of data analysis. It is used to classify records to a proper class
and is applicable in both regression and associations tasks. C4.5 is used in classification problems and it is the widely used
algorithm for building DT. It is suitable for real world problems as it deals with numeric attributes and missing values.

D2. Naive Bayse- this is a modest probabilistic classifier, which is based on an assumption about mutual independency of
attributes. The probability which is applied in the Nave Bayes algorithm are calculated according to the Bayes Rule, The naive
Bayes classifier's attractiveness is in its simplicity, computational efficiency, and good classification performance. The nave
Bayes classifier requires a very large number of records to obtain good outcomes.

D3. Neural Networks -There are 3 layers in neural networks: input layer, hidden layer, output layer. Hidden layer is the
products of the input layer. The condition between neurons has weights which are assigned to them. Their values are calculated
with the use of back Propagation algorithm. In hidden layers there are some nonlinear features added to the network. The out
layer may have more than one output node which predict the different diseases.

D4. Genetic algorithms -are based on the standard of genetic modification, mutation and natural selection. These are
algorithmic optimization strategies motivated by the principles observed in natural evolution [10]. The genetic algorithm creates a
number of random solutions to the problem. All these solutions may not be good, a group of solutions can be skipped entirely,
and it can come down to the overlapping solutions. Poor solutions are discarded and the good ones retained. Good solutions are
then being hybridized and then the whole process is repeated. Finally, similar to the process of natural selection, only the best
solutions remain. So, from the set of potential solutions to the problems that compete with each other, the best solutions are
chosen and combined with each other in order to obtain a universal solution from the set of solutions that will become better and
better, similar to the process of evolution of organisms.

D5. Nearest neighbor -method is a method that is also used for data classification. Unlike other techniques, there is no
learning process to create a model. The data used for learning is in fact a model. When the new data shows up, the algorithm
analyzes all the data in the database to find a subset of instances that are the best fit and based on that it is able to predict the
2016, IJARCSMS All Rights Reserved ISSN: 2321-7782 (Online) 133 | P a g e
Ravleen et al., International Journal of Advance Research in Computer Science and Management Studies
Volume 4, Issue 2, February 2016 pg. 131-138
outcome. The study [11] conducted on the application of nearest neighbor method on standard data set to detect efficiency in the
diagnosis of heart diseases, produced the results that application of this method achieved an accuracy of 97.4% which is a
higher percentage than any other published study on the same set of data. More details on data mining techniques can be found
in Berry and Linoff.[12]

Intelligent decisions are similar to human decisions but are automated decisions. Classification and prediction in machine
[13].
learning are among the techniques that can produce intelligent decision So are researches still in progress. Decision tree
models are best suited for data mining. They are inexpensive to construct, easy to interpret, easy to integrate with database
system and they have comparable or better accuracy in many applications. There are many Decision tree algorithms such as
[14].
HUNTS algorithm (one of the earliest algorithm), CART, ID3, C4.5 (a later version ID3 algorithm), SLIQ, SPRINT
Artificial neural networks (ANN) provide a powerful tool to help doctors to analyze, model and make sense of complex clinical
data across a broad range of medical applications. A neural network has been successfully applied to various areas of medicine,
[15].
such as diagnostic aides, medicine, biochemical analysis, and image analysis and drug development

V. DATA MINING EVOLVING APPLICATIONS

There is vast potential for data mining applications in healthcare. Some of them can be grouped as under:

E.1 Infection Control in Hospital

Two million patients each year in the United States are affected by Nosocomial infections. Computer-assisted surveillance
research has focused on identifying high-risk patients, expert systems, and possible cases and detecting deviations in the
occurrence of predefined events [16]. The system uses association rules on culture and patient care data obtained from the
laboratory information management systems and generates monthly patterns that are reviewed by an expert in infection control.
An early warning of the global spread of SARS virus is an example of the usefulness of syndromes systems based on data
mining [17].

E2. High-Risk Patients Identification

American Health ways provides diabetes disease management services to hospitals and health plans designed to enhance
the quality and lower the cost of treatment of individuals with diabetes. A robust data mining and model building solution
identifies patients who are trending toward a high-risk condition. WEKA 3.6 is used as the data mining tool to implement the
Algorithms. The J48 classifier performs classification with 81.8% accuracy in predicting the HIV status [18].

2016, IJARCSMS All Rights Reserved ISSN: 2321-7782 (Online) 134 | P a g e

Ravleen et al., International Journal of Advance Research in Computer Science and Management Studies
Volume 4, Issue 2, February 2016 pg. 131-138
E3.Treatment effectiveness
[19].
Data mining applications can be developed to evaluate the effectiveness of medical treatments By comparing the
outcomes of patient groups treated with different drug regimens for the same disease or condition can be compared to determine
which treatments work best and are most cost-effective. Data mining could be particularly useful in medicine when there is no
dispositive evidence favoring a particular treatment option Based on patients profile, history, physical examination, diagnosis
and utilizing previous treatment patterns, new treatment plans can be effectively suggested for Examples: Onset, treatment and
[20].
management of depression Treatment Decision Support Tool for Patients with Uterine Fibroids[21]. HianChyeKoh and
Gerald Tan works on data mining and its applications with major areas like Treatment effectiveness, Management of healthcare,
Detection of fraud and abuse, Customer relationship management [22].

E4.Healthcare management

To aid healthcare management, data mining applications can be developed to well identify and track chronic disease states
and high-risk patients, design appropriate interventions, and reduce the number of hospital admissions and claims. Sierra Health
Services has used data mining comprehensively to identify areas for quality improvements, including treatment guidelines,
disease management groups, and cost management [23].

E5.Healthcare Resource Management

Effectively manage the resource allocation by identifying high risk areas and predicting the need and usage of various
resources. If the inpatient length of stay (LOS) can be predicted efficiently, the planning and management of hospital resources
can be greatly enhanced. Neural network system is used to predict the disposition in children presenting to the emergency room
with bronchiolitis.

E6.Customer relationship management

Customer interactions may occur through call centers, physicians offices, billing departments, inpatient settings, and
ambulatory care settings. The principles of applying of data mining for customer relationship management in the other
industries are also applicable to the healthcare industry. In many cases prediction of purchasing and usage behavior can help to
provide proactive initiatives to reduce the overall cost and increase customer satisfaction.

E7.Pharmaceutical Industry

The pharmaceutical firms manage their inventories and to develop new product and services. Pharmaceutical companies
can benefit from healthcare CRM and data mining by tracking which physicians prescribe which drugs and for what purposes.
Pharmaceutical companies can decide whom to target, show what the least expensive or most effective treatment is plan for an
ailment [24].

E8.Fraud and abuse

Data mining applications that attempt to detect fraud and abuse often establish norms and then identify unusual or abnormal
patterns of claims by physicians, laboratories, clinics, or others. Among other things, these applications can highlight
inappropriate prescriptions or referrals and fraudulent insurance and medical claims. A method based on naive Bayes that
effectively combines the advantages of boosting and the explanatory power of the weight of evidence scoring framework was
[25]
presented in Furthermore, the classification algorithm C4.5 was applied for fraud/abuse detection by using the discovered
temporal patterns as predictive features. A data mining framework that uses the concept of clinical pathways (or integrated care
pathways) was utilized for detecting unknown fraud and abusive cases in a real-world data set gathered from the National
Health Insurance (NHI) program in Taiwan [26].

2016, IJARCSMS All Rights Reserved ISSN: 2321-7782 (Online) 135 | P a g e

Ravleen et al., International Journal of Advance Research in Computer Science and Management Studies
Volume 4, Issue 2, February 2016 pg. 131-138
E9.Human Talent Predictions

We can discover an employee performance using classification and prediction techniques in DM. Since the construction of
decision trees does not require any expert knowledge or parameter setting, they remain popular and are considered for
exploratory knowledge discovery. Still the technique which is otherwise known as the divide-and-conquer rule is undergoing
researches.

E10.Talent Forecasting

Association rules are used to associate employees profiles to the most appropriate program or job and then associate the
employee attitude with performance. And the predictions used on classification to find out the percentage of accuracy in
employee performance, behavior and attitude, analyzing, forecasting and identifying the best profile for different employees

E11.System Biology

Biological databases contain a wide variety of data types, often with rich relational structure. Accordingly multi-relational
[27].
data mining techniques are frequently applied to biological data Systems biology is at least as demanding as, and perhaps
more demanding than, the genomic challenge that has fired international science and gained public attention.

VI. LIMITATION IN DATA MINING

[28].
Data mining applications can greatly benefit the healthcare industry but they were some limitations as well Healthcare data
mining can be limited by the accessibility of data, because the raw inputs for data mining often exist in different settings and systems
[29],
such as administration, clinics, laboratories and more. Data from heterogeneous sources present the challenges. Secondly, the
missing, corrupted, inconsistent or non-standardized data such as pieces of information recorded in different formats in different data
sources is a big challenge. Missing attribute values can impact the assessment of whether a particular combination of attribute-value
[30].
pairs is significant within a dataset Cleaning data from noise and outliers and handling missing values, and then finding the right
[31].
subset of data, prepares them for successful data mining Thirdly, there may be ethical, legal and social issues, such as data
ownership and privacy issues, related to healthcare data. Fourthly, the successful application of data mining requires knowledge of
the domain areas as well as in data mining methodology and tools otherwise the user may not be able to avoid the pitfalls of data
[32].
mining Fifthly a sufficiently exhaustive mining of data will certainly yield patterns of some kind that are a product of random
[33]
instabilities especially for large data sets with many variables. Hence, many interesting or significant patterns and relationships
found in data mining may not be useful. Finally, Data mining projects can fail for a variety of reasons, such as lack of management
support, unrealistic user expectations, poor project management, inadequate data mining expertise, and many more.

2016, IJARCSMS All Rights Reserved ISSN: 2321-7782 (Online) 136 | P a g e

Ravleen et al., International Journal of Advance Research in Computer Science and Management Studies
Volume 4, Issue 2, February 2016 pg. 131-138
VII. CONCLUSION AND FUTURE WORK

The key to successful data mining is to first define the business or clinical problem to be solved. New knowledge is not
discovered by the algorithms, but by the user. Today, there have been many efforts with the goal of successful application of
data mining in the healthcare institutions. Primary potential of this technique lies in the possibility for research of hidden
patterns in data sets in healthcare domain. These patterns can be used for clinical diagnosis. However, available raw medical
data are widely distributed, different and voluminous by nature. These data must be collected and stored in data warehouses in
organized forms, and they can be integrated in order to form hospital information system. There are extra research challenges
for the integration of such an ample data for information system. As healthcare data are not limited to just quantitative data,
such as physicians notes or clinical records it is also useful to look into how digital diagnostic images can be brought into
[34], [35]
healthcare data mining applications. Some progress has been made in these areas with the future development of
information communication technologies; data mining will achieve its full potential in the discovery of knowledge hidden in the
medical data which is concerned with research edifice.

References

1. Tan, P., Steinbach, M. and Kumar, V. Introduction to Data Mining, Addison-Wesley, Boston, 2006.
2. grawal, R. and Srikant, R., 1994. Fast algorithms for mining association rules. In Proceedings of the 20th International Conference on Very Large
Databases (VLDB 94), Santiago, Chile.
3. Agrawal, R. and Shim, K., 1996. Developintightly coupled data mining applications on a relational database system. In Proceedings of the 2nd
International Conference on Knowledge Discovery in Databases and Data Mining (KDD 96), Portland, Oregon, USA.
4. Berry, M. J. A., & Linnof, G, Data mining Techniques, New York: Wiley, (1997).
5. Milley, A. (2000). Healthcare and data mining. Health Management Technology, 21(8), 44-47.
6. CRoss Industry Standard Process for Data Mining[Online] Available:http://www.crisp-dm.org, 2003.
7. David L. Olson and Dursun Delen, Data Mining Process in Advanced Data Mining Technique, Springer, 2008.
8. Getoor, L. (2003). Link Mining: A New Data Mining Challenge. SIGKDD Explorations Volume 4, Issue 2.
9. Berry MJ, Linoff G. Data mining techniques: for marketing, sales and customer support. USA: Wiley, 1997.
10. 10. Gupta, S., Kumar, D., & Sharma, A. (2011). Data Mining Classification Techniques Applied For Breast Cancer Diagnosis And Prognosis. Indian
Journal of Computer Science and Engineering (IJCSE) 188-195.
11. Shouman, M., Turner, T., & Stocker, R. (2012). Applying k-Nearest Neighbour in Diagnosing Heart Disease Patients. 2012 International Conference on
Knowledge Discovery (ICKD 2012) IPCSIT Vol. XX. Singapore:IACSIT Press
12. Berry, M.J.A. & Linoff, G.S. (1997). Data Mining Techniques: For Marketing, Sales and Customer Support. New York: John Wiley & Sons Inc.
13. Jantan, H., Hamdan, A. R., & Othman, Z. A., Human Talent Prediction in HRM using C4.5 Classification Algorithm, (IJCSE) International Journal on
Computer Science and Engineering Vol. 02, No. 08, 2010, 2526-2534.
14. Han, J. and M. Kamber, 2001. Data Mining: Concepts and Techniques. San Francisco, Morgan Kauffmann Publishers.
15. Miller, A., 1993. The application of neural networks to imaging and signal processing in astronomy and medicine. Ph.D. Thesis, Faculty of Science,
Department of Physics, University of Southampton.
16. SE Brosette , AP Spragre, WT Jones and SA. Moser, A data mining system for infection control surveillance, Methods Inf Med , Vol. 39, pp. 303-310,
2000.
17. Brewin, B. (2003). New health data net may help in fight against SARS. Computerworld , 37(17), 1, 59.
18. EliasLemuye, Hiv Status Predictive Modeling Using Data Mining Technology.
19. Milley, A. (2000). Healthcare and data mining. Health Management Technology, 21(8), 44-47.
20. Maja Hadzic, Fedja Hadzic and Tharam Dillon et al. Mining of patient data: towards better treatment strategies for depression. International Journal of
Functional Informatics and Personalised Medicine, 2010
21. [Campbell2010] Kevin Campbell, N. Marcus Thygeson and Stuart Speedie. Exploration of
22. Classification Techniques as a Treatment Decision Support Tool for Patients with Uterine Fibroids; Proceedings of International Workshop on Data
Mining for HealthCare Management, PAKDD-2010.
23. HianChyeKoh and Gerald Tan, Data Mining Applications in Healthcare , journal of Healthcare Information Management Vol 19, No 2.
24. Berry MJ, Linoff G. Data mining techniques: for marketing, sales and customer support. USA: Wiley, 1997.
25. Brannigan, M. (1999). Quintiles seeks mother lode in health data mining. Wall Street Journal, March 2, 1.
26. S. Viaene, A. Richard and D. G. Dedene, A case study of applying boosting Naive Bayes to claim fraud diagnosis, IEEE Transactions on Knowledge
and Data Engineering, vol. 16, no. 5, pp. 612620, May 2004.
27. W. S. Yang and S. Y. Hwang, A process-mining framework for the detection of healthcare fraud and abuse, Expert Systems with Applications, Article
in Press, corrected proof, 2005.

2016, IJARCSMS All Rights Reserved ISSN: 2321-7782 (Online) 137 | P a g e

Ravleen et al., International Journal of Advance Research in Computer Science and Management Studies
Volume 4, Issue 2, February 2016 pg. 131-138
28. David Page and Mark Craven, Biological Applications of Multi Relational Data Mining.
29. Silver, M. Sakata, T. Su, H.C. Herman, C. Dolins, S.B. & OShea, M.J. (2001). Case study: how to apply data mining techniques in a healthcare data
warehouse. Journal of Healthcare Information Management, 15(2), 155-164.
30. Benko, A. & Wilson, B. (2003). Online decision support gives plans an edge. Managed Healthcare Executive, 13(5), 20.
31. Parameshvyas Laxminarayan, Sergio A. Alvarez, Carolina Ruiz, and Majaz Moonis, Mining Statistically Significant Associations for Exploratory
Analysis of Human Sleep Data, IEEE TRANSACTIONS ON INFORMATION TECHNOLOGY IN BIOMEDICINE, VOL. 10, NO. 3, JULY 2006
32. Amir R. Razavi & Hans Gill & Hans Ahlfeldt, Nosrat Shahsavar, Predicting Metastasis in Breast Cancer: Comparing a Decision Tree with Domain
Experts, J Med Syst (2007) 31:263273
33. McQueen, G. & Thorley, S. (1999). Mining fools gold. Financial Analysts Journal, 55(2), 61-72.
34. Hand, D.J. (1998). Data mining: statistics and more? The American Statistician, 52(2), 112-118.
35. Ceusters, W. (2001). Medical natural language understanding as a supporting technology for data mining in healthcare. In Medical Data Mining and
Knowledge Discovery, Cios, K. J. (Ed.), Physica- Verlag Heidelberg, New York, 41-69.
36. Megalooikonomou, V. & Herskovits, E.H. (2001). Mining structurefunction associations in a brain image database. In Medical Data Mining and
Knowledge Discovery, Cios, K. J. (Ed.), Physica-Verlag Heidelberg, New York, 153-180.

2016, IJARCSMS All Rights Reserved ISSN: 2321-7782 (Online) 138 | P a g e