Have a personal or library account? Click to login

Patient-centric health-care data processing using streams and asynchronous technology

Open Access
|Jan 2018

Figures & Tables

Figure 1

Cruncher algorithm: A simple representation framework based on minimum description length for automatically forming a set of concepts from attribute-value data sets.
Cruncher algorithm: A simple representation framework based on minimum description length for automatically forming a set of concepts from attribute-value data sets.

Figure 2

API algorithm.
API algorithm.

Figure 3

Portal algorithm.
Portal algorithm.

Figure 4

Portal home page.
Portal home page.

Figure 5

Portal patients’ page.
Portal patients’ page.

Figure 6

Portal pharmacies page.
Portal pharmacies page.

Figure 7

Execution time of functions when using the socket–connection approach.
Execution time of functions when using the socket–connection approach.

Figure 8

Execution time of functions when using the request–response approach.
Execution time of functions when using the request–response approach.

j_ijssis-2018-003_tab_006

NamePharmacy_counts
FieldsName (text) (primary key) events (text) (primary key) count (int) date (text) (primary key, clustering column).
UseThis is useful for viewing interactions that pharmacies have with patients through the app over time. It is good for identifying trends in interaction between pharmacies and their customers, e.g. if the number of appointments decrease, you can reach out to users to remind them they can do so through the app.
Comments

j_ijssis-2018-003_tab_005

NameAppointment_count
FieldsStatus (text) (primary key) count (int) date (text) (primary key, clustering column).
UseThis is useful for horizontal and vertical comparisons of appointment data. Useful for pie/line charts showing comparison of appointments booked through the app. It is also useful for targeting pharmacies that have a high number of mobile app customers to reward them or look at areas with low app usage and increase marketing there.
CommentsPotentially other uses besides marketing.

j_ijssis-2018-003_tab_008

NameRemedy_count
FieldsDose_type (text) (primary key) count (int) date (text) (primary key, clustering column).
UseThis is useful for horizontal and vertical comparisons of remedy data. Can be used in line or bar graph to show the comparison of different types of drugs users’ record in the app. This info can be used to create marketing segments/targeted marketing, e.g. if there is low usage of people recording injections, you can create a marketing campaign to attract this target group to the app.
CommentsPotentially other uses besides marketing.

j_ijssis-2018-003_tab_007

NamePharmacy_appointments
FieldsName (text) (primary key) city (text) (primary key) app_date (text) (primary key, clustering column) totals (object) details (object).
UseThis is useful for viewing which pharmacies are actively receiving appointments through the mobile app and which are not. This info can therefore be used to incentivize pharmacies to keep using the app to organize or communicate with patients.
Comments

j_ijssis-2018-003_tab_002

NameUser_locations
FieldsGender (text) (primary key) city (text) (primary key) date_created (text) (primary key, clustering column) last_login (text) (primary key, clustering column) postcode (text) data (object).
UseThis is useful for viewing user activity by location. You can view locations of active users within a given time period and determine areas that have high or low usage for marketing purposes. Adding a heatmap layer will also bring out high- and low-usage hotspots.
CommentsCreated index on postcode as well.

j_ijssis-2018-003_tab_001

NameResource_counts
FieldsType (text) (primary key) date (text) (primary key, clustering column) count (int).
UseThis is useful for counting the number of pharmacies, services, appointments, users, etc.
CommentsThis table is not suitable for graphing historical data points, but is useful for showing state at specific points in time, e.g. summation that the widgets in the portal need.

j_ijssis-2018-003_tab_004

NameAdherence_drug
FieldsMedicine_name (text) (primary key) totals (object) date (text) (primary key, clustering column).
UseThis is useful for determining which drugs get low or high adherence. Allows you to profile the performance of the app with respect to different drugs. For example: Does mobile app help ‘drug x’ patients more than ‘drug y’, then investigate the cause from there.
CommentsWill need to stream and map remedy ids to medicine name within this cruncher (note: for viewing all drugs within the time period, use adherence_count with “status IN” clause)

j_ijssis-2018-003_tab_003

NameAdherence_count
FieldsStatus (text) (primary key) count (int) date (text) (primary key, clustering column).
UseThis is useful for horizontal and vertical comparisons of adherence data. Useful for pie/line charts showing the percentage of adherence status, or line graph for how the taken status has been growing.
CommentsHelpful when identifying fluctuation in adherence so as to investigate further.
Language: English
Page range: 1 - 18
Published on: Jan 3, 2018
Published by: Professor Subhas Chandra Mukhopadhyay
In partnership with: Paradigm Publishing Services
Publication frequency: 1 times per year

© 2018 Kenneth Mbuthia, Jin Dai, Stavros Zavrakas, Jize Yan, published by Professor Subhas Chandra Mukhopadhyay
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.