You are on page 1of 5

Machine learning in healthcare sector

Machine learning, since its invention in 1949, changed working culture of many technology
sector drastically. Machine learning require dataset to effectively implement algorithm and
sitting on goldmine of historical datasets, healthcare was one of the very early adopters and
benefited greatly from technology advancements. As per report of McKinsey, a comparative
report of technologies in healthcare shows the dominance and predicted growth of the
same in upcoming years. Figure shows market size of different technologies in healthcare
industry and forecast of its growth.

Market size (in bn $)


14
12
10
8
6
4
2
0
Machine learning NLP Context aware computer vision Qerying Method
computing

2019 2026

Pharma industry is classified in two category Diagnostic healthcare and Predictive


healthcare. This work talks about various machine learning techniques and their application
in prediction of various diseases.

Various machine learning techniques used:

Support Vector Machine (SVM): SVM algorithm was designed in 1990’s. It is simple and
prominent process. A collection of data is divided in different categories in this technique.
This method is use for classification and regression problem [1]. It has been shown to
produce low prediction error compared to the classifier based on other methods like ANN
when we consider a large number of parameters in our study [2].

Naïve Bayes Classification: This algorithm is prominently use for classification as it performs
one scanning of data. Statistical classifiers are the example for Bayesian classifier [3].

Decision Tree: Decision tree is mostly used technique for classification containing internal
node and one leaf node with a class label. The top node of decision tree are called as root
nodes. It is one of very simple in terms of construction and do not require any parameters
[4].
K-nearest neighbor (KNN): KNN is one of the very frequent use methods for classification of
sample. This method helps to calculate distance measure from N numbers of training data
set [5].

Fuzzy logic: It evolve from fuzzy set theory. Its value always lies between 0 to 1 and gives
more relevant results to real life application problem. [6]

CART: Classification and Regression Tree Methodology is known as the CART. In


classification and regression tree the target variable represented as categorical and
continuous. These variables are used to predict value in the tree [7].

Machine learning techniques used for prediction of various disease:

Diagnostic of heart disease: To increase the accuracy of prediction of the Heart disease
Machine Learning Techniques are used in healthcare sector. Dataset are taken from the UCI
Machine Learning Repository. “Parthiban and Srivatsa” solve problem by using Naive Bayes
algorithm for detection and, SVM for analysis of heart disease [8]. They found utilizing Naïve
Bayes algorithm provides 74% accuracy and SVM offer 94.60% is achieved “Otoom A.F.” has
performed Support Vector Machine, Bayes Net to predict coronary heart disease [9]. The
accuracy provided by SVM is 88.3%, Bayes Net provides 84%.

Diagnostic of diabetic diseases: To increase the accuracy of diagnostic of diabetic diseases


different machine learning techniques are used. Dataset considered from the UCI Machine
Learning Repository. “Iyer A.” used Naive Bayes and Decision trees machine learning
algorithm to predict diabetic disease. Naive Bayes gives 79.56% accuracy and decision tree
provides 76.95% accuracy [10]. “Dash S and Sen SK” use experiments to diagnose diabetes
disease. Logiboost, CART algorithms are used and Logiboost provides the accuracy of
77.48% [11].

Diagnostic of breast cancer using machine learning: It is one of the top cancers that occurs
in a woman and it is the second main leading reason for woman in the United States and in
Asia countries. Machine learning algorithms are used to detect breast cancer. The data
considered from WISCONSIN dataset, UCI machine learning repository. “Williams et al.”
used as J-48, Naive Bayes to identify breast cancer risks in the United States. The
experiment is performed through WEKA tool. They found J-48 is the best algorithm for the
prediction of breast cancer it gives 94.2% accuracy, and their work with Naive Bayes gives
82.6% [12]. To predict breast cancer “Senturk et al.” used several classification models like
Support Vector Machine (SVM), Naive Bayes (NB), K-nearest neighbor, and Decision tree
(DT). K-NN gives 95.15% accuracy and SVM gives 96.40% accuracy [13]. “Majali et al.” used
decision tree and Frequent Pattern in data mining to predict the breast cancer. They found
decision tree gives 94% accuracy [14].
Diagnostics of thyroids disorder: To detect thyroid diseases we use classification algorithms
that are support vector machines and Decision tree are used and datasets are considered
from UCI repository. “Papageorgiou EI” proposed advanced approaches for thyroid
diagnosing diseases using fuzzy map utilizing data mining algorithms [15].

The below table summarizes different machine learning techniques used for diagnosis of
various diseases.

S.No. Disease M.L. Technique Dataset Accuracy Reference


No.
1 Heart Disease J48/ SVM UCI 84.35%/85.03% 16
2 Heart Disease Naïve Bayes Diabetic research 86.41% 17
institute in Chennai
3 Diabetic Disease SVM UCI 78% 18
4 Diabetic type 2 Naïve Bayes Different part of 95% 19
India
5 Breast Cancer J48 UCI 98.14% 20
6 Breast Cancer Decision Tree Swami vivekanada 97% 21
diagnostic center
hospital
7 Breast Cancer “Decision Tree” UCI 91% 22
followed by
“SVM”
8 Breast Cancer CART/ Naïve University of wiscons 92.42%/97.42% 23
Bayes in hospitals
9 Thyroid Disease SVM UCI 98.62% 24
Reference:

1. Murphy, K. P. (2012). A Probabilistic Perspective.


2. Byvatov, E., & Schneider, G. (2003). Support vector machine applications in
bioinformatics. Applied bioinformatics, 2(2), 67-77.
3. Hazra, A., Mandal, S. K., & Gupta, A. (2016). Study and analysis of breast cancer cell
detection using Naive Bayes, SVM and ensemble algorithms. International Journal of
Computer Applications, 145(2), 0975-8887.
4. Sharma¹, P., & Bhatia, A. P. R. (2012). Implementation of decision tree algorithm to
analysis the performance.
5. Bishop, C. M. (1995). Bayesian methods for neural networks.
6. Zimmermann, H. J. (2011). Fuzzy set theory—and its applications. Springer Science &
Business Media.
7. Learning, M. (2017). Heart Disease Diagnosis and Prediction Using Machine Learning
and Data Mining Techniques: A Review. Advances in Computational Sciences and
Technology, 10(7), 2137-2159.
8. Parthiban, G., & Srivatsa, S. K. (2012). Applying machine learning methods in
diagnosing heart disease for diabetic patients. International Journal of Applied
Information Systems (IJAIS), 3(7), 25-30.
9. Otoom, A. F., Abdallah, E. E., Kilani, Y., Kefaye, A., & Ashour, M. (2015). Effective
diagnosis and monitoring of heart disease. International Journal of Software
Engineering and Its Applications, 9(1), 143-156.
10. Iyer, A., Jeyalatha, S., & Sumbaly, R. (2015). Diagnosis of diabetes using classification
mining techniques. arXiv preprint arXiv:1502.03774.
11. Shailaja, K., Seetharamulu, B., & Jabbar, M. A. (2018, March). Machine Learning in
Healthcare: A Review. In 2018 Second International Conference on Electronics,
Communication and Aerospace Technology (ICECA) (pp. 910-914). IEEE.
12. Williams, K., Idowu, P. A., Balogun, J. A., & Oluwaranti, A. I. (2015). Breast cancer risk
prediction using data mining classification techniques. Transactions on Networks and
Communications, 3(2), 01-01.
13. Senturk, Z. K., & Kara, R. (2014). Breast cancer diagnosis via data mining:
performance analysis of seven different algorithms. Computer Science &
Engineering, 4(1), 35.
14. Shailaja, K., Seetharamulu, B., & Jabbar, M. A. (2018, March). Machine Learning in
Healthcare: A Review. In 2018 Second International Conference on Electronics,
Communication and Aerospace Technology (ICECA) (pp. 910-914). IEEE.
15. Papa Georgiou, E. I., Papandreou’s, N. I., Apostol Poulos, D. J., & Vassilios, P. J. (2008,
June). Fuzzy cognitive map-based decision support system for thyroid diagnosis
management. In 2008 IEEE international conference on fuzzy systems (IEEE world
congress on computational intelligence) (pp. 1204-1211). IEEE.
16. Chaurasia, V., & Pal, S. (2014). Data mining approach to detect heart
diseases. International Journal of Advanced Computer Science and Information
Technology (IJACSIT) Vol, 2, 56-66.
17. Vembandasamy, K., Sasipriya, R., & Deepa, E. (2015). Heart diseases detection using
Naive Bayes algorithm. International Journal of Innovative Science, Engineering &
Technology, 2(9), 441-444.
18. Kumari, V. A., & Chitra, R. (2013). Classification of diabetes disease using support
vector machine. International Journal of Engineering Research and Applications, 3(2),
1797-1801.
19. Vijayan, V. V., & Anjali, C. (2015, December). Prediction and diagnosis of diabetes
mellitus—A machine learning approach. In 2015 IEEE Recent Advances in Intelligent
Computational Systems (RAICS) (pp. 122-127). IEEE.
20. Shrivastavat, S. S., Sant, A., & Aharwal, R. P. (2013). An overview on data mining
approach on breast cancer data. International Journal of Advanced Computer
Research, 3(4), 256.
21. Jhajharia, S., Verma, S., & Kumar, R. (2016, August). A cross-platform evaluation of
various decision tree algorithms for prognostic analysis of breast cancer data.
In 2016 International Conference on Inventive Computation Technologies
(ICICT) (Vol. 3, pp. 1-7). IEEE.
22. Sivakami, K., & Saraswathi, N. (2015). Mining big data: breast cancer prediction using
DT-SVM hybrid model. International Journal of Scientific Engineering and Applied
Science (IJSEAS), 1(5), 418-429.
23. Shajahaan, S. S., Shanthi, S., & ManoChitra, V. (2013). Application of data mining
techniques to model breast cancer data. International Journal of Emerging
Technology and Advanced Engineering, 3(11), 362-369.
24. Kalaimani, I. (2019). Analysis for the Prediction of Thyroid Disease by Using ICA and
Optimal Kernel SVM Approach. International Journal of Emerging Technology and
Innovative Engineering, 5(3).

You might also like