Nit6130 Data Collection And Storage Assessment Answer

Data 3

Other healthcare centers

Application of artificial intelligence in other healthcare centers

txt

Free

Yes 

1.3 Data storage

After collecting all the required data, another table is created to store the raw data collected from the data sources. Storage of data is very important as it makes sure the data is safe and can be used in the future when required (Lu et al., 2015). The data storage table is shown below: 

Table 2: Data storage table

Data source name

Date of collection

Saved file location

Saved file name

Saved file format

Number of records

Survey from major hospitals

22/4/2018

//raw data/

Survey.txt1

txt

50

Survey from clinics

25/4/208

//raw data/

Survey.txt2

txt

80

Survey from other healthcare centers

29/4/2018

//raw data/

Survey.txt3

txt

150

2 Design and implementation

After the collection and the storage of the data, the data needs to undergo data pre-processing and feature selection or the dimension reduction before it can be analyzed and implemented as required to obtain the desired results of the research.

2.1 Data pre-processing

Data pre-processing is any form of processing done on the raw data to prepare it to be fit to be used in an experiment or research. Data pre-processing is a common practice in the data mining process where it is done to transform the data into a format which will be easily and effectively used by the users (Ramírez-Gallego et al., 2017, pp.39-57). Like in data mining, in our case data preprocessing is done to transform the raw data into a more favorable data format which will be easily understood and analyzed to obtain the desired results of our research.

We have many processes involved in data pre-processing where some of the major processes include data cleaning, data integration, data transformation, data reduction, data discretization, among other processes (García, Luengo, and Herrera, 2016). Data pre-processing can be represented diagrammatically by the figure shown below:

Data integration is the process of combining the data from different sources to obtain one set of data which will be valuable and relevant to be used in the analysis. In our case, the data from the major hospitals, the clinics, and other healthcare centers is combined to obtain one data set which will be analyzed easily to understand how artificial intelligence affects the field of healthcare (Cudré-Mauroux, 2017, pp.5-6).

Data transformation is the process of converting all the integrated data into the format which is required during the analysis of the data (Heer, Hellerstein, and Kandel, 2015).

Data reduction is the process of transforming data into a correct, simpler, and well-organized and well-ordered data which can be manipulated or analyzed with much ease to obtain the desired results (Rehman et al., 2016, pp.917-928).

Data discretization is the technique of converting large and complex data sets into smaller, finite, and simpler data sets which can be easily understood and analyzed with much ease to obtain the desired results (Ramírez?Gallego et al., 2016, pp.5-21).

In our case, the collected raw data about the effects of artificial intelligence on the healthcare field undergo the whole process of data pre-processing to get the most suitable data which will be used in the analysis.

2.2 Feature selection or dimension reduction

After data pre-processing, features selection or dimension reduction is done to select the most appropriate features and do a further reduction to remove all the unnecessary data to make sure we’ll be left with only the data to be used in the analysis (Hira and Gillies, 2015). A new table shown below is prepared to record the data after feature selection and dimension reduction. 

Table 3: Feature selection and data reduction table

Date

Data source name

Purpose of pre-processing

Method of pre-processing

Original data records

Results data records

New data file name

2/5/2018

Data 1

Avoid duplicity

Data reduction(cleaning)

50

35

Survey.txt11

2/5/2018

Data 2

Feature selection

Data integration

80

64

Survey.txt22

2/5/2018

Data 3

Filter the data

Data reduction

150

127

Survey.txt33 

2.3 Experiment designing

This section explains how the research was conducted to obtain the desired results of how artificial intelligence is affecting healthcare.

2.3.1 Detailed design steps

To carry out our research successfully, we used hybrid methodology or the mixed research methodology which made our research easy since we collected and analyzed both numerical and non-numerical forms of data (Creswell and Clark, 2017). We used interviews and questionnaires as the main methods of data collection to obtain our desired data from the major hospitals, clinics, and the other healthcare centers in our societies (Flick, 2017). A simplified table showing some of the main questionnaire questions used in data collection is shown below: 

Table 4: A table of questionnaire questions

Question 1

What’s the name of your organization?

Question 2

Do you use artificial intelligence in your medical operations?

Question 3

If yes, please give some of the major operations where you use artificial intelligence in your organization

Question 4

What are the major benefits of artificial intelligence in your organization?

Question 5

What are the major challenges facing artificial intelligence in your organization?

Question 6

In your own views, has artificial intelligence helped to improve the quality of services offered in your organization and do you support the use of artificial intelligence in your organization or it should be ended? 

2.3.2 The results obtained

After visiting various major hospitals, clinics, and other healthcare centers where we interviewed various healthcare personnel and gave various questionnaire forms with some questions about the effects of artificial intelligence on their performance, we obtained the following simplified results:

Table 5: A table showing the organizations used in data collection

Data source name

Total number of organizations used in data collection

Number of organizations using artificial intelligence in their operations

Number of organizations not using artificial intelligence in their operations

Major hospitals

50

46

4

Clinics

80

67

13

Other healthcare centers

150

112

38 

Table 6: A table showing the organizations which supported the use of artificial intelligence in their operations

Data source name

Total number of organizations using artificial intelligence

Number of organizations supporting the use of artificial intelligence in their operations

Number of organizations who don’t support the use of artificial intelligence in their operations

Major hospitals

46

45

1

Clinics

67

65

2

Other healthcare organizations

112

108

4 

2.4 Implementation

After obtaining all the required data and doing all the required modifications, the implementation stage is undertaken.

2.4.1 The software and tools used in the analysis

After collecting, modifying, and recording of the required data, the data is usually analyzed using the appropriate software and tools. In our case, the main software to be used in the analysis of data are Ms. Word and Ms. Excel which will be used to analyze the obtained data and tabulate them in tables and charts which will help to enhance their visualization and their understanding to the other people who may be interested in the results of the research (Ward, Grinstein, and Keim, 2015).

2.4.2 The actual analysis and tabulation of results

After doing some analysis, a table can be drawn to represent the results of the number of organizations using artificial intelligence in percentages.

Table 7: A table showing the numbers and the percentages of the organizations using artificial intelligence in their operations

Data source name

Total number of organizations used in data collection

Number and percentages of organizations using artificial intelligence in their operations

Number and percentages of organizations not using artificial intelligence in their operations

Major hospitals

50

46 (92%)

4 (8%)

Clinics

80

67 (83.75%)

13 (16.25%)

Other healthcare centers

150

112 (74.67%)

38 (25.33%)

The data shown in the table above can be represented by the pie charts shown below:

Figure 5: A bar graph showing the numbers of the data sources

Table 8: A table showing the numbers and the percentages of the organizations supporting the use of artificial intelligence in their operations

Data source name

Total number of organizations using artificial intelligence

Number of organizations supporting the use of artificial intelligence in their operations

Number of organizations who don’t support the use of artificial intelligence in their operations

Major hospitals

46

45 (97.83%)

1 (2.17%)

Clinics

67

65 (97.01%)

2 (2.99%)

Other healthcare organizations

112

108 (96.43%)

4 (3.57%)

From the table above, we can get the pie charts shown below:

This section will analyze the estimated results and the actual results obtained in the whole research process.

3.1 Results estimation

Before carrying out any research, it’s good to have a rough idea of the results you expect to get in the research. The rough idea helps the researchers to make some estimations of the expected results of the research. In our case, we expected that we have many major hospitals, clinics, and other healthcare centers using artificial intelligence in their operations and we also expected that artificial intelligence has helped to improve the performance of the healthcare organizations and so most of the healthcare organizations support artificial intelligence in their operations (Iyengar, Kundu, and Pallis, 2018, pp.29-31).

We also expected to find that there are some challenges facing the application of artificial intelligence in the field of healthcare and the healthcare organizations are undertaking some measures to help in addressing some of the major challenges affecting the use of artificial intelligence in their operations (Price and Nicholson, 2017). The expectations we have explained can be seen as some of the major estimations of the results which we made before carrying out the main research.

3.2 Results summary

From the analysis done above, we can make the following summary of the results:

92% of the major hospitals use artificial intelligence in their operations.

83.75% of the clinics use artificial intelligence in their operations.

74.67% of the other healthcare centers use artificial intelligence in their operations.

From the same data analysis, we can also say that of all the organizations which use artificial intelligence in their operations:

97.83% of the major hospitals support the use of artificial intelligence in their operations.

97.01% of the clinics support the use of artificial intelligence in their operations.

96.43% of the other healthcare centers support the use of artificial intelligence in their operations.

The summary of the results given above clearly shows that artificial intelligence has affected the healthcare sector positively, and that’s the main reason it’s supported by a very large percentage of the medical organizations using it (Hamet and Tremblay, 2017, pp.36-40). Therefore, we can end our research by saying that medical organizations should embrace the use of artificial intelligence in their operations as it has very many benefits. Those medical organizations which have not yet incorporated the use of artificial intelligence in their operations should strive to do so as fast as possible for them to enjoy the many benefits. Lastly, we can say that the use of artificial intelligence in the medical organizations has some few challenges and the organization using it should look for some appropriate measures to address these challenges for them to enjoy the many benefits with few or no challenges.

4 Outline of the research and result analysis

• Data collection and storage
  • Data sources
  • Data collection
  • Data storage
• Design and implementation
  • Data pre-processing
  • Feature selection or dimension reduction
  • Experiment designing
    • Detailed design steps
    • The results obtained
  • Implementation
    • The software and tools used
    • The actual analysis and tabulation of results
  • Result analysis
    • Results estimation
    • Results summary 

References

Beam, A.L., Kompa, B., Fried, I., Palmer, N.P., Shi, X., Cai, T. and Kohane, I.S., 2018. Clinical Concept Embeddings Learned from Massive Sources of Medical Data. arXiv preprint arXiv:1804.01486.

Cody, R., 2017. Cody's data cleaning techniques using SAS. SAS Institute.

Creswell, J.W. and Clark, V.L.P., 2017. Designing and conducting mixed methods research. California: Sage publications.

Cudré-Mauroux, P., 2017, June. Big Data Integration. In Telecommunications (ConTEL), 2017 14th International Conference on (pp. 5-6). IEEE.

Flick, U. 2017. The Sage Handbook of Qualitative Data Collection. California: SAGE.

García, S., Luengo, J. and Herrera, F., 2016. Data preprocessing in data mining. New York: Springer.

Hamet, P. and Tremblay, J., 2017. Artificial intelligence in medicine. Metabolism-Clinical and Experimental, 69, pp.36-40.

Heer, J., Hellerstein, J.M., and Kandel, S., 2015. Predictive Interaction for Data Transformation. In CIDR.

Hira, Z.M., and Gillies, D.F., 2015. A review of feature selection and feature extraction methods applied to microarray data. Advances in bioinformatics, 2015.

Iyengar, A., Kundu, A. and Pallis, G., 2018. Healthcare Informatics and Privacy. IEEE Internet Computing, 22(2), pp.29-31.

Lu, G., Ho, L., Danilak, R., Mullendore, R.N., Jones, J. and Tomlin, A.J., Western Digital Technologies Inc and Skyera LLC, 2015. Data reliability schemes for data storage systems. U.S. Patent 9,021,339.

Price, I.I. and Nicholson, W., 2017. Artificial Intelligence in Health Care: Applications and Legal Implications.

Ramírez?Gallego, S., García, S., Mouriño?Talín, H., Martínez?Rego, D., Bolón?Canedo, V., Alonso?Betanzos, A., Benítez, J.M. and Herrera, F., 2016. Data discretization: taxonomy and big data challenge. Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, 6(1), pp.5-21.

Ramírez-Gallego, S., Krawczyk, B., García, S., Wo?niak, M. and Herrera, F., 2017. A survey on data preprocessing for data stream mining: current status and future directions. Neurocomputing, 239, pp.39-57.

Rehman, M.H., Chang, V., Batool, A. and Wah, T.Y., 2016. Big data reduction framework for value creation in sustainable enterprises. International Journal of Information Management, 36(6), pp.917-928.

Ward, M.O., Grinstein, G. and Keim, D., 2015. Interactive data visualization: foundations, techniques, and applications. AK Peters/CRC Press.