Machine Learning for Healthcare Applications - Группа авторов

Скачать книгу

to finding false examples, however to likewise giving quicker methodologies with less computational cost when applied to tremendous measured datasets.

      1. Mason, E., Jain, S., Kendall, M., Mostashari, F., Blumenthal, D., The regional extension center program: Helping physicians meaningfully use health information technology. Ann. Intern. Med., 153, 666–670, 2010.

      2. Mossialos, E., Wenzl, M., Osborn, R., Sarnak, D., International Profiles of Healthcare Systems, The Commonwealth Fund, New York, NY, 2016.

      3. Parikh, R.B., Kakad, M., Bates, D.W., Integrating predictive analytics into high-value care: the dawn of precision delivery. JAMA, 315, 651–652, 2016.

      4. Goldstein, B.A., Navar, A.M., Pencina, M.J., Ioannidis, J.P., Opportunities and challenges in developing risk prediction models with electronic health records data: A systematic review. J. Am. Med. Inform. Assoc., 27, 1, 198–208, 2016.

      5. Jung, K., Covington, S., Sen, C.K., Januszyk, M., Kirsner, R.S., Gurtner, G.C. et al., Rapid identification of slow healing wounds. Wound Repair Regen., 24, 181–188, 2016.

      6. Pencina, M.J. and Peterson, E.D., Moving from clinical trials to precision medicine: The role for predictive modeling. JAMA, 315, 1713–1714, 2016.

      7. Chen, R. and Michael, S., Promise of personalized comics to precision medicine. Wiley Interdiscip. Rev. Syst. Biol. Med., 5.1, 77–82, 2013.

      8. Nickel, M. and Kiela, D., Poincare Embeddings for Learning Hierarchical Representations arXiv:170508039, abs/1705.08039, 1–10, 2017.

      9. Lin, C., Jain, S., Kim, H., Bar-Joseph, Z., Using neural networks for reducing the dimensions of single-cell RNA-Seq data. Nucleic Acids Res., 45, 17, e156–e156, 2017.

      10. Choi, E., Bahadori, M.T., Song, L., Stewart, W.F., Sun, J., GRAM: Graph-based attention model for healthcare representation learning. International Conference on Knowledge Discovery and Data Mining (KDD), ACM, pp. 787–795, 2017.

      11. Silver, M., Sakata, T., Su, H.C., Herman, C., Dolins, S.B., O’Shea, M.J., Case study: How to apply data mining techniques in a healthcare dataware house. Healthcare Inf. Manage., 15, 2, 155–164, 2001.

      12. NHCAA, https://www.nhcaa.org, 2020.

      13. FBI Reports and publications: Financial, https://www.fbi.gov/starts-services/publications, 2009.

      14. FBI Scams and Safety: Common fraud schemes, https://www.fbi.gov/, Scams-and-safety, 2011.

      15. Spencer, K. and Herbert, D., Mass Marketing of Property and Liability Insurance, Department of Transportation, United States of America, 1970.

      16. Li, J., Huang, K., Jin, J., Shi, J., A survey on statistical methods for healthcare fraud detection. Healthcare Manage. Sci., 11, 275–287, 2008.

      17. London: The Guardian. The mystery of John Darwin, https://www.theguardian.com, 2007.

      18. Relles, D., Ridgeway, G., Cater, G., Data mining and the implementation of a prospective payment system for inpatient rehabilitation. Health Serv. Outcomes Res. Method, 3, 3–4, 247–266, 2002.

      19. Koh, H. and Tan, G., Data mining applications in healthcare. J. Healthc Inf. Mgmt., 19, 2, 64–72, 2005.

      21. Jeffries, D., Zaidi, I., Jong, B., Holland, M., Miles, D., Analysis of flow cytometry data using an automatic processing tool. Cytometry Part A, 73A, 9, 857–867, 2008.

       * Corresponding author: [email protected]

Part 2 MACHINE LEARNING/DEEP LEARNING-BASED MODEL DEVELOPMENT

      A Framework for Health Status Estimation Based on Daily Life Activities Data Using Machine Learning Techniques

       Tene Ramakrishnudu*, T. Sai Prasen and V. Tharun Chakravarthy

       National Institute of Technology-Warangal, India

       Abstract

      In the current generation, it is very important to monitor our health. With the busy lives of people nowadays, many are experiencing health-related issues at an early age. Many of these issues arise because of our daily life activities. People are interested in many activities, but they hardly know the consequences of those activities. Hence it is very important to detect daily life activities that affect the health of a person and predict the diseases that may come in the future. However, there are existing methods for predicting a particular kind of disease like diabetes, tuberculosis, etc., based on electronic health records. The proposed system predicts the overall health status of a person using machine learning techniques. The overall health status includes how well a person is sleeping, eating, doing physical activity, etc. Also, the proposed system monitors the health of persons and alerts when they are deviating from a normal state. In this chapter, we will discuss the data collection approach, architecture of the system, overall health estimation models, implementation details, and the analysis of the result.

      Keywords: Healthcare data analysis, machine learning in healthcare, data analytics, health status estimation

      2.1.1 Health Status of an Individual

      The overall health status of a person is assessed by comparing the level of wellness with the level of illness. The health status can be estimated through many parameters. Some of the parameters are (i) Sleep status: the health level of a person is depending on his/her sleep timings, (ii) Screen status: the health level of a person is depending on the amount of time spent on screen, (iii) Drink status: the health level of a person is depending on his/her drinking habits, (iv) Smoke status: the health level of a person is depending on his/her smoking activities (v) Calories status: the health level of a person is depending on the calories consumed and physical activities.

      The things that an individual does daily can be referred to as activities. Some of the activities include sleeping, watching television, consuming alcohol, smoking cigarette, listening to the radio, reading books, etc. Measures of an Individual include physical measures like height, weight, and some other measures like age, gender, etc. Basically, many of the measures are permanent they will not change frequently, whereas the activities might change frequently.

      2.1.3 Traditional Approach to Predict Health Status

      In general, health status can be predicted by consultancy experts. If an individual wants to know about their sleep status (i.e. whether their sleep pattern is good? And whether they are taking the adequate amount of sleep?), they can consult an expert at sleep centers. If an individual wants to know about their calorie status (i.e. How much calories they need to consume to maintain/increase/decrease weight? How much exercise they need to do to maintain the calories in balance?), they can consult physicians.

      But what the experts do, they give some suggestions by considering the measures and activities mentioned previously. For this, the experts use some rules and conditions on the measures

Скачать книгу