Order from us for quality, customized work in due time of your choice.
Dr. Mark Khatchuturian makes a presentation about predictive analytics in healthcare to provide a brief but extensive overview of the topic. The presentation begins with a definition, which is a reasonable approach. The listeners learn that predictive analysis uses technologies to predict future health-related events with the help of data analysis by defining patterns discernible in data sets. As demonstrated by Dr. Khatchuturian, the field of biomedical imaging provides healthcare with big data, which makes its study a very time-consuming activity. However, computers have superior analytic abilities that tend to increase as technology develops, which makes predictive data analytics possible (Belle et al., 2015).
Apart from that, Dr. Khatchuturian discusses the process of predictive analysis. In particular, he mentions the need for establishing the baseline and singling out the spatial features, which include the position, size, or the number of studied elements, and the temporal ones, which include the changes of the elements that occur with time. The features and their specifics allow determining the mathematical approach that applies to a particular case. Dr. Khatchuturian mentions several methods with varying levels of automatization; for example, the seed point method requires manual selection of a region, but the rest of the analysis is carried out automatically.
The field of predictive analysis is developing, and the number of validated, applicable tools and methods is growing (Belle et al., 2015). For example, Dr. Khatchuturian mentions the method that is not yet clinically approved but has the potential of substituting the need for a baseline with its morphology algorithm. This method is called anatomical labeling, and it requires deforming the brain and placing it in a common template, which can be created through the analysis of multiple actual brains of varied forms. It is also a form of applying predictive analysis to magnetic resonance imaging (MRI). Other MRI applications that Dr. Khatchuturian highlights are functional MRI (using visual stimuli on patients to change MRI signals) and the temporal study of ischemia images that investigate the patterns of brain changes after stroke. The analysis of the images and signals has a great potential in enabling and improving diagnosing and treatment determination (Padfield, Mendonca, & Gupta, 2015; Patil, Langoju, Joel, Patil, & Genc, 2015).
Moreover, Dr. Khatchuturian considers an example of applying predictive analysis to positron emission tomography (PET) which is used to investigate the successfulness of tumor treatment with the help of temporal data. Dr. Khatchuturian highlights the fact that it is not the most accurate predictor, but since it is the one that is available, it is used in practice. Since tools and methods proceed to develop (Sow, Turaga, Turaga, & Schmidt, 2015), it is entirely possible that more accurate approaches can be found or this one can be improved in the future (Belle et al., 2015). It is noteworthy that MRI and PET are two of the most widely used imaging modalities in healthcare (Padfield et al., 2015), which justifies Dr. Khatchuturians choice of examples.
Dr. Khatchuturian finishes with a discussion of the potential and the issues of the field. The potential of predictive analysis (the prediction of harmful events and their prevention) is valuable for healthcare (Dhar, 2014; Raghupathi & Raghupathi, 2014). The related issues include the liability dilemma, which is unresolved yet, and the challenge of validation, which currently lacks clear guidelines. Also, there is the problem of accuracy. Since the technology and the methods of predictive analysis are currently developing (Belle et al., 2015), this issue may find solutions in the future.
To sum up, the presentation of Dr. Khatchuturian provides an overview of an acute topic in healthcare that is supplied with relevant examples, all of which are briefly explained engagingly.
References
Belle, A., Thiagarajan, R., Soroushmehr, S., Navidi, F., Beard, D., & Najarian, K. (2015). Big Data Analytics in Healthcare. Biomed Research International, 2015, 1-16.
Dhar, V. (2014). Big data and predictive analytics in health care. Big Data, 2(3), 113-116.
Padfield, D., Mendonca, P., & Gupta, S. (2015). Biomedical image analysis. In C. K. Reddy & C. C. Aggarwal (Eds.), Healthcare data analytics (pp. 61-90). New York, NY: CRC Press.
Patil, A., Langoju, R., Joel, S., Patil, B., & Genc, S. (2015). Biomedical signal analysis. In C. K. Reddy & C. C. Aggarwal (Eds.), Healthcare data analytics (pp. 127-187). New York, NY: CRC Press.
Raghupathi, W. & Raghupathi, V. (2014). Big data analytics in healthcare: promise and potential. Health Information Science and Systems, 2(1), 3.
Sow, D., Turaga, K., Turaga, D., & Schmidt, M. (2015). Mining of sensor data in healthcare: A survey. In C. K. Reddy & C. C. Aggarwal (Eds.), Healthcare data analytics (pp. 91-126). New York, NY: CRC Press.
Order from us for quality, customized work in due time of your choice.