Browse Prior Art Database

Personal Dietary Evaluation Tool

IP.com Disclosure Number: IPCOM000245457D
Publication Date: 2016-Mar-10
Document File: 5 page(s) / 118K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is a system and model that present an up-to-date simulation of a user’s body status, which combines both structured and unstructured data into a seamless composite, providing real time medical data and advice to mitigate potentially harmful dietary situations. The learning machine helps a user determine, at any time and in location, whether a food is safe to consume.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 46% of the total text.

Page 01 of 5

Personal Dietary Evaluation Tool

People have different reasons (e.g., to lose weight, health concerns, allergies, religion, etc.) for implementing restrictive diets. Removing one type of staple food (e.g., wheat) can be very challenging. If the person needs to start adding more restrictions to those that already exist, that difficulty exponentially increases, as does the probability of making a potentially harmful mistake.

Current methods for keeping medical and dietary records for an individual and matching to a medical model are lacking a cognitive/learning machine component to assist people with dietary needs and restrictions.

A model is needed that helps a user determine, at any time and place, whether a food

is safe to consume.

The novel contribution is a system and model that presents an up-to-date simulation of a user's body status, which combines both structured and unstructured data into a

seamless composite, providing real time medical data and advice to mitigate potentially harmful dietary situations. The learning machine has a dual purpose: to keep the model as current as possible and to function as an on-site diagnostician.

The solution has three components:


• A network-based back-end (i.e., cloud); a solid network that manages and stores information from multiple sources such as, but not limited to:

- Medical practitioners - The Internet (e.g., restaurant menus, chemical compositions, food preparation descriptions, etc.)


- User input (from a mobile or stationary device)

- Wearable devices
- Tools that process biological material (e.g., diabetes monitor)


• A learning machine that can sort information and provide insights


• One or more portable devices that can accept various forms of biological and/or data-based input

The system comprises a two-tier hierarchical setup for organizing raw data, starting with the most accurate sources (i.e., the least likely to be changed) at the bottom and the less accurate sources (which can be frequently) modified at the top.

The foundation of this structure is the individual's biological data, obtained from visits to medical practitioners, ranging from weight, height, heart rate, personal medical history, the disorder's general trends/progression traits, to more specific information, such as radioallergosorbent test (RAST) results, immunoglobulin E (IgE) results, fluorescence enzyme-labeled assays, oral glucose tolerance test results (OGTT), and hydrogen breath test results. This section of data also contains the individual's current list of substances to avoid, the chemical compositions of food items, and medications (both prescribed and over-the-counter).

1


Page 02 of 5

The top layer includes statistics obtained from other sources that are not supervised by medical practitioners; this is data obtained by wearables, mobile sources (i.e. lab-on-a-chip devices such as the diabetes monitor), or flags of potential problems that the user discovers over time. Wearable devices are in...