Helping medication users discover and solve nutritional deficiencies

Easton, Benjamin Michael (2016) Helping medication users discover and solve nutritional deficiencies. [Dissertation (University of Nottingham only)]

[img] PDF - Registered users only - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (4MB)

Abstract

As a population, we are highly reliant on medications. The benefits these drugs have brought in terms of saving lives are enormous. However, these gains are not without issues. Drugs can cause essential minerals and vitamins in the body to become greatly reduced, a process known as nutrient depletion. This can lead in turn to a whole host of other complications and diseases. This project uses the latest in chat bot style interaction to educate and inform consumers, explaining to them what nutrient depletion is and how they can solve their specific issues. By making the process entertaining and unique, it provides an avenue of information for groups of individuals who would not read the information on complicated medical sites. Furthermore, a proprietary dataset containing links between specific nutrients and medications allows for a personalised result. This dataset, provided by pharmacist Mike Wakeman, is a combination of data from the most recent, up to date medical studies. Users are taken through an interactive process. First, it explains to users about deficiencies caused by any medications they take, as well as why these nutrients matter. Following on from this demographic information, such as age and gender, allows for even more specific results. All this interaction takes place through a conversational bot. Finally, the bot provides recommendations of foods specifically to help prevent these nutritional deficiencies. This process of food selection is a complex one. Different options are examined, with a genetic algorithm implemented for the final solution. If foods are not sufficient, HealthBot provides a base for linking to appropriate supplements. A variety of technologies are used to implement this feature set, using the Ruby on Rails framework for website development, which is linked to a Heroku cloud hosted server and a PostgreSQL database. Heavy use of the Wit.ai bot platform enables the system to understand natural language, such as the phrase "Tell me about biotin". HealthBot provides a full solution for identifying and solving medication and lifestyle related nutritional deficiencies.

Item Type: Dissertation (University of Nottingham only)
Depositing User: Gonzalez-Orbegoso, Mrs Carolina
Date Deposited: 18 Jan 2017 12:45
Last Modified: 19 Oct 2017 17:35
URI: http://eprints.nottingham.ac.uk/id/eprint/39158

Actions (Archive Staff Only)

Edit View Edit View