AardRock Wiki - User contributions [en]

Glycemic Effect

2006-05-09T15:32:55Z

Dpkingma: added some data

The software needs to give an estimation of the temporal effect of food.

Inspiritional research/papers:
* http://www.fao.org/docrep/w8079e/w8079e00.htm#Contents
* http://www.ajcn.org/cgi/content/full/76/1/5

Glycemic Index (GI) / Glycemic Load data:
* http://www.naturalstrength.com/nutrition/detail.asp?ArticleID=220
* Big: http://www.ajcn.org/cgi/content/full/76/1/5/T1

Glycemic Effect

2006-05-09T15:29:02Z

Dpkingma:

The software needs to give an estimation of the temporal effect of food.

Inspiritional research/papers:

Glycemic Index (GI) / Glycemic Load data:

File:Statisticallearningproblem.png

2006-05-08T10:10:43Z

Dpkingma: Image describing the learning problem domain.

Image describing the learning problem domain.

User:Dpkingma

2006-05-02T10:14:12Z

Dpkingma:

See: [[User:DurkKingma]]

Advisory System

2006-05-02T10:13:03Z

Dpkingma: /* The Glucose Model */

=Cheetah advisory system=

===Prelude===
Cheetah will generatie advice about insulin dosis or food intake. Several software packages offer this kind of functionality, but none of them an advice as personal as Cheetah will. The system learns from the user and learns how its body reacts on events such as insulin usage, food comsumption and activitivies like sports. The generated KB about the user is then shared (anonymously) with other users over the internet. Likewise, the user receives knowledge from other users. And this is exactly what makes cheetah unique: it will be the first diabetes software that doesnt just learn from one isolated user, but learns from all. This combined with a smart knowledge inferencing system will add up to a complete diabetes health system.

===Advice===
Three types of advice can hypothetically be generated. The first one is short-term, and gives advice about today. The second one tackles long-term problems and generates modifications to the current food/insulin diet. The third kind of advice generates a complete diet for you.
[Type A is a must for Cheetah, while types B and C are optional at the moment]

====Type A Advice====
Examples:
* Ann wants to eat grandma's pie. Give her current glucose level, she wants insulin advice to prevent a hyper.
* Ann will exercise in 2 ours, and wants food advice to prevent a hypo.
Advice Features:
* Advice is generated about insulin dosis or food intake about today (max. 24h ahead).
Steps:
* The user asks for short-term advice
* The system presents a scheme on which the user fills in today's future insulin or food intake. (Needs a intuitive and easy UI which allows the user to choose default values). The user can fill in "?" which are events to be advised.
* Cheetah puts these things into his model and sees how to fill in the ?'s to make the glucose prediction as stable as possible.
* Cheetah presents the advise.

====Type B Advice====
Examples:
* Ann always has a hypo at night. She wants advice to prevent this to happen again.
* Ann always has a hyper in the evening. She wants advice to prevent his to happen again.
Advice Features:
* Advice is generated not only for today, but is more general. Unlike type A, the (hypo/hyper) problem is recurrent.
Steps:
* The user (and cheetah) spots a hypo/hyper problem and wants a solution.
* Cheetah searches what causes the recurrent problem.
* Cheetah presents the cause and gives advice how to solve this.

====Type C Advice====
Examples:
* Pete has very much hypo's and hypers and wants the systeem to make a compleet food diet for him.
* Pete goes on a food diet and asks the system to make a new insulin advise for him.
Advice Features:
* Advice consists of a complete new personal diet, generated from scratch.
Steps:
* [This is a stub] System makes everything a variable and optimizes the curve.

===The Glucose Model===
Lets begin with thinking about Type A Advice and forget type B and C advice. Regardless of the precise implementation, the system needs a way to (trustably) predict if the user will be o.k. given some conditions, and will not be o.k. given some other contitions. For now, lets forget on how this system can be build/learned, and just look at the properties of such system.
In such system, the "o.k.-ness" is directly linked with the blood sugar level: these should between some variables, say 5 and 15. Also, the system should be able to say if a person will be o.k. in 1 hour, in 2 hours, in 3 hours etc. More concretely speaking, there should be a ''model'' mapping conditions to glucose levels over time. For each certain interval, say every hour, the model looks at the conditions and predicts blood sugar levels. If, given certain conditions, future glucose levels are optimal then the system knows the conditions are also optimal.
Conditions are things like insulin or food intake, activities, current time, current glucose levels, health status etc.
Conditions come in three kinds.
1. The first ones are certain conditions, from which the effect is certain. Such conditions are current time and current glucose level (if measured). These could also be hard-coded events which effects are certain a priori.
2. The second ones are predicted uncertain conditions. Such conditions have been seen before, and the effect has been estimated (with some certainty). Ann (or someone else in the network) has eaten grandma's applepie before, and the system has estimated the amount of effectivly absorbed carbs. Or: Ann has done a hockey training before, and the system has estimated (with some certainty) the amount of used glucose over the hours.
3. The third ones are conditions which effect are yet unnknown. For example, you're eating something totally exotic and the system cant give an estimation. The effects of such a condition are yet to be estimated.

Advisory System

2006-05-02T10:11:24Z

Dpkingma:

=Cheetah advisory system=

===Prelude===
Cheetah will generatie advice about insulin dosis or food intake. Several software packages offer this kind of functionality, but none of them an advice as personal as Cheetah will. The system learns from the user and learns how its body reacts on events such as insulin usage, food comsumption and activitivies like sports. The generated KB about the user is then shared (anonymously) with other users over the internet. Likewise, the user receives knowledge from other users. And this is exactly what makes cheetah unique: it will be the first diabetes software that doesnt just learn from one isolated user, but learns from all. This combined with a smart knowledge inferencing system will add up to a complete diabetes health system.

===Advice===
Three types of advice can hypothetically be generated. The first one is short-term, and gives advice about today. The second one tackles long-term problems and generates modifications to the current food/insulin diet. The third kind of advice generates a complete diet for you.
[Type A is a must for Cheetah, while types B and C are optional at the moment]

====Type A Advice====
Examples:
* Ann wants to eat grandma's pie. Give her current glucose level, she wants insulin advice to prevent a hyper.
* Ann will exercise in 2 ours, and wants food advice to prevent a hypo.
Advice Features:
* Advice is generated about insulin dosis or food intake about today (max. 24h ahead).
Steps:
* The user asks for short-term advice
* The system presents a scheme on which the user fills in today's future insulin or food intake. (Needs a intuitive and easy UI which allows the user to choose default values). The user can fill in "?" which are events to be advised.
* Cheetah puts these things into his model and sees how to fill in the ?'s to make the glucose prediction as stable as possible.
* Cheetah presents the advise.

====Type B Advice====
Examples:
* Ann always has a hypo at night. She wants advice to prevent this to happen again.
* Ann always has a hyper in the evening. She wants advice to prevent his to happen again.
Advice Features:
* Advice is generated not only for today, but is more general. Unlike type A, the (hypo/hyper) problem is recurrent.
Steps:
* The user (and cheetah) spots a hypo/hyper problem and wants a solution.
* Cheetah searches what causes the recurrent problem.
* Cheetah presents the cause and gives advice how to solve this.

====Type C Advice====
Examples:
* Pete has very much hypo's and hypers and wants the systeem to make a compleet food diet for him.
* Pete goes on a food diet and asks the system to make a new insulin advise for him.
Advice Features:
* Advice consists of a complete new personal diet, generated from scratch.
Steps:
* [This is a stub] System makes everything a variable and optimizes the curve.

===The Glucose Model===
Lets begin with thinking about Type A Advice and forget type B and C advice. Regardless of the precise implementation, the system needs a way to (trustably) predict if the user will be o.k. given some conditions, and will not be o.k. given some other contitions. For now, lets forget on how this model can be made, and look at the properties of such system.
In such system, the "o.k.-ness" is directly linked with the blood sugar level: these should between some variables, say 5 and 15. Also, the system should be able to say if a person will be o.k. in 1 hour, in 2 hours, in 3 hours etc. More concretely speaking, there should be a ''model'' mapping conditions to glucose levels over time. For each certain interval, say every hour, the model looks at the conditions and predicts blood sugar levels. If, given certain conditions, future glucose levels are optimal then the system knows the conditions are also optimal.
Conditions are things like insulin or food intake, activities, current time, current glucose levels, health status etc.
Conditions come in three kinds.
1. The first ones are certain conditions, from which the effect is certain. Such conditions are current time and current glucose level (if measured). These could also be hard-coded events which effects are certain a priori.
2. The second ones are predicted uncertain conditions. Such conditions have been seen before, and the effect has been estimated (with some certainty). Ann (or someone else in the network) has eaten grandma's applepie before, and the system has estimated the amount of effectivly absorbed carbs. Or: Ann has done a hockey training before, and the system has estimated (with some certainty) the amount of used glucose over the hours.
3. The third ones are conditions which effect are yet unnknown. For example, you're eating something totally exotic and the system cant give an estimation. The effects of such a condition are yet to be estimated.

Advisory System

2006-05-02T09:34:02Z

Dpkingma:

=Cheetah advisory system=

==Prelude==
Cheetah will generatie advice about insulin dosis or food intake. Several software packages offer this kind of functionality, but none of them an advice as personal as Cheetah will. The system learns from the user and learns how its body reacts on events such as insulin usage, food comsumption and activitivies like sports. The generated KB about the user is then shared (anonymously) with other users over the internet. Likewise, the user receives knowledge from other users. And this is exactly what makes cheetah unique: it will be the first diabetes software that doesnt just learn from one isolated user, but learns from all. This combined with a smart knowledge inferencing system will add up to a complete diabetes health system.

==Advice==
Three types of advice can hypothetically be generated. The first one is short-term, and gives advice about today. The second one tackles long-term problems and generates modifications to the current food/insulin diet. The third kind of advice generates a complete diet for you.
[Type A is a must for Cheetah, while types B and C are optional at the moment]

===Type A Advice===
Examples:
* Ann wants to eat grandma's pie. Give her current glucose level, she wants insulin advice to prevent a hyper.
* Ann will exercise in 2 ours, and wants food advice to prevent a hypo.
Advice Features:
* Advice is generated about insulin dosis or food intake about today (max. 24h ahead).
Steps:
* The user asks for short-term advice
* The system presents a scheme on which the user fills in today's future insulin or food intake. (Needs a intuitive and easy UI which allows the user to choose default values). The user can fill in "?" which are events to be advised.
* Cheetah puts these things into his model and sees how to fill in the ?'s to make the glucose prediction as stable as possible.
* Cheetah presents the advise.

===Type B Advice===
Examples:
* Ann always has a hypo at night. She wants advice to prevent this to happen again.
* Ann always has a hyper in the evening. She wants advice to prevent his to happen again.
Advice Features:
* Advice is generated not only for today, but is more general. Unlike type A, the (hypo/hyper) problem is recurrent.
Steps:
* The user (and cheetah) spots a hypo/hyper problem and wants a solution.
* Cheetah searches what causes the recurrent problem.
* Cheetah presents the cause and gives advice how to solve this.

===Type C Advice===
Examples:
* Pete has very much hypo's and hypers and wants the systeem to make a compleet food diet for him.
* Pete goes on a food diet and asks the system to make a new insulin advise for him.
Advice Features:
* Advice consists of a complete new personal diet, generated from scratch.
Steps:
* [This is a stub] System makes everything a variable and optimizes the curve.

DesignThoughts

2006-05-02T07:56:36Z

Dpkingma:

= Design Thoughts =

We can use this page to describe how we intend to implement certain ideas, and to discuss designs. Note that this is from the Developer's point of view, while the [wiki:Stories Stories] and CheetahUserManual are from the User's point of view.

* [[CheetahInstallation]]
* [[SoftwareUpdates]]
* [[CheetahApplicationLifecycle]]
* [[Internationalization]] (I18N)
* [[ExceptionLogging]]
* [[ProgressMonitoring]]
* [[UserPreferences]]
* [[DataModel|Data Model]]
* [[RegistrationProcedure| Registration Procedure]]
* [[LoginProcedure| Login procedure]]
* [[Wizard Rabbit Treasurer]]
* [[Advisory System]]

LogboekDurkKingma

2006-04-27T09:40:47Z

Dpkingma:

Datum Tijd Duur Cumulatief Activiteit
--------------- -------------- --------------- ------------- ----------------------------------------
8 februari 13-17 4 4 Eerste overleg en taakverdeling
11 februari 9-11 2 6 Inlezen op Agile Development
13 februari 12-14 2 8 Cruise Control onderzocht, nu pas door ziekte.
15-17 2 10 Vergadering
14 februari 15-17 2 12 XP Game
16 februari 9-12 3 15 WiredReach fundamenten uitgezocht
20 februari 15-17 2 17 vergadering
20-22 2 19 Notulen uitgewerkt / Op het web gezet
21 februari 9-15 6 25 WiredReach/Eclipse platform onderzoeken (met Martijn)
22 februari 9-10 1 26 Wiki logboek gebouwd
10-14 4 30 WiredReach/Eclipse platform onderzoeken
25 februari 12-18 6 36 Implementatie WiredReach proef
27 februari 9-15 6 42 JXTA/Andere protocollen vergeleken
15-17 2 44 Vergadering
28 februari 15-17 4 45 AardRock vergadering
5 maart 9-15 6 51 Onderzoek WiredReach/JXTA
15-17 2 53 Vergadering
6 maart 15-17 2 55 Vergadering AardRock
12-16 4 59 Lunch, vergaderd over implementatie P2P gedeelte
13 maart 15-17 2 61 Vergadering
16 maart 9-13 4 64 Uitwerking/Uitdenken P2P communicatie datamodel
21 maart 15-17 2 67 Vergadering AardRock
23 maart 9-15 6 73 De-Eclipsed WiredReach, aan RDF Black Box gewerkt, Jena uitgezocht.
27 maart 10-17 3 76 RDF Black Box uitgewerkt, vergadering
28 maart 15-17 2 78 Vergadering AardRock
18 april 14-17 3 81 Vergadering Aardrock / Proefpresentatie
19 april 15-17 2 83 Overleg adviessysteem met Just
25 april 9-15 6 89 Neuraal Netwerk vergadering / Adviessysteem cheetah
15-17 2 91 Vergadering Cheetah
27 april 9-17 8 99 Bayesiaans netwerk / neuraal netwerk / onderzoek algoritmes
28 april 9-11 2 101 Uitzoeken andere adviessoftware / dagboeken / mailen Harold de Valk
28 april 11-12 1 102 Update logboek

DurkKingma

2006-04-27T08:54:56Z

Dpkingma:

==Personal Data==

* Born May 30, 1983
* Adress: Weerdsingel Oostzijde 1bis, 3514AA, Utrecht
* Contact phone: +31641511577

==Project Logboek==
* [[LogboekDurkKingma]]

==Education==

* September 2002 until ...: Computer science at Utrecht University. Minors: Medical Technical Science

==Work Experience==

* (2004) KPN ADSL-helpdesk (technische/tweede lijn)
* (2005-...) Operator Groep Delft

==Skills==

* Java
* (X)HTML, CSS
* PHP, SQL
* Haskell, Z80 Assembly, other 'educational' languages

==Hobbies / Extracurricular stuff==

* Student life (USC/going out)
* Sports (Rowing, Hockey, Golf, Hiking)
* 2006: Organizer of the 123th Varsity (national student rowing championships)
* 2006: Organizer of the 1st Nacht van de Fooi (3rd world fundraising)
* Traveling

DurkKingma

2006-04-27T08:53:14Z

Dpkingma:

DurkKingma

2006-04-27T08:51:06Z

Dpkingma:

Personal Data

* Born May 30, 1983
* Adress: Weerdsingel Oostzijde 1bis, 3514AA, Utrecht
* Contact phone: +31641511577

Education

* September 2002 until ...: Computer science at Utrecht University. Minors: Medical Technical Science

Work Experience

* (2004) KPN ADSL-helpdesk (technische/tweede lijn)
* (2005-...) Operator Groep Delft

Skills

* Java
* (X)HTML, CSS
* PHP, SQL
* Haskell, Z80 Assembly, other 'educational' languages

Hobbies / Extracurricular stuff

* Student life (USC/going out)
* Sports (Rowing, Hockey, Golf, Hiking)
* 2006: Organizer of the 123th Varsity (national student rowing championships)
* 2006: Organizer of the 1st Nacht van de Fooi (3rd world fundraising)
* Traveling

Inspirational Diabetes Software

2006-04-25T15:03:24Z

Dpkingma: /* Interesting Related Projects */ Added link to AIDA

==Other diabetes care software==
Other software that aid in managing diabetes and offer serious input for Cheetah:
* [http://lifescan.com/products/booklets/ OneTouch Owner's Booklets] from [http://lifescan.com/ LifeScan]
**[http://lifescan.com/pdf/ultra_ob.pdf OneTouch® Ultra® Owner's Booklet] (US, 2005)
**[http://lifescan.com/pdf/ultrasmart_ob.pdf OneTouch® UltraSmart® Owner's Booklet] (US, 2003)
**[http://lifescan.com/pdf/netherlands/050-287-12A.pdf GLUCO TOUCH® Handleiding] (NL, 2001)
* [http://www.animascorp.com/products/userguide/manual/410-023-00%20rL.pdf ezManager User Guide] from [http://animascorp.com Animas]
**ezManager costs $129, may be billed to health insurance company
* [http://sourceforge.net/projects/glucotools/ GlucoTools]
**a collection of tools for Palm Pilot and Visor type PDAs that assist in managing diabetes]
**latest update in 2003
**aimed at individuals using insulin pumps (often referred to as "pumpers"
* [http://sourceforge.net/projects/insbuddy Insulin Buddy]
**doesn't seem very active
* [http://lists.debian.org/debian-med/2004/03/msg00006.html Open-Source Diabetes tools at sourceforge.net]

= Interesting Related Projects =

* [http://www.csc.liv.ac.uk/~floriana/PIPS/papers/FoodOntology.pdf An example of food ontology for diabetes control] (might come in handy)
* [http://2aida.org/aida/example.htm AIDA: am exentensive pancreas/blood sugar level simulator. I (Durk) am asking the author for his methods and source. Very usefull for creating glucose forecast and insuline/food recommendations.

= How is Cheetah different? =

Yet another software solution that helps managing diabetes? What's new? So let's test Cheetah against these IdeaKillers.

Use these idea killers to make Cheetah the strongest, most wanted piece of software. Ever!

==Cheetah:==
* puts you in control
* involves you to co-create Cheetah to fit their most urgent needs through a unique community process
* helps reducing "hypers" and "hypos" to an absolute minimum
* allows physicians, specialists and care takers transparent and controlled access to a wealth of diabetic information
* is always up to date with food nutrition information
* is the only diabetes care software that continuosly learns and adapts to your situation to suggest insulin doses and carb intake
* is fully distributed
* does not require any adminstration or backup
* is self-organizing
* is self-healing
* is Original Dutch Software™
* is Open Source Software
* runs on Mac, PC, PDA and mobile phones
* has a strong and dedicated development team evolving the software
* uses a unique business model that helps creating a better planet

Got something else that makes Cheetah unique? '''Add it to the list!'''

Notulen20060418

2006-04-18T15:52:49Z

Dpkingma: Notulen ingevoerd

== Notulen 18/04/2006 ==

''Wie heeft wat gedaan?''
Patrick: user registration/login gedeelte. Is bezig met scherm voor invoer persoonlijke gegevens. Is deze week af.
Chris: updates gedeelte: montly/montly/daily. Verbinding van dit met database, daarna met Maven aan de slag.
Just: voorbereiding tussenpresentatie. Met Marco Wiering het over Neurale Netwerken gehad. Gaat hier nog meer literatuur over vinden. Is bezig met story over diabetis-specifieke invoer.
Durk is klaar met de Varsity.
Martijn: doorontwikkeld aan de database.
Martien heeft een verhaal over The Wizard, the Rabbit and the Treasurer rondgestuurd. Het is een model over datareplicatie en datareputatie. Graag feedback van iedereen over dit model.
Martien zit donderdag met Giel op HKU om de tafel over de UI. Donderdagavond wordt een ontwerp gemaild.
Sjoerd: Usability.

''Nieuwe projectleider?''
Voor volgende week moet er een nieuwe projectleider gekozen zijn. Zaterdag is de deadline voor een onderbouwingsbrief (mailen naar sp-cheetah). Maandagavond, niet te laat, dient er gestemd te zijn.

''Advies-systeem''
Just heeft bij zijn afspraak met Marco Wiering advies gekregen over de te gebruiken AI-methodes. Het is een goed idee om een algemeen neuraal netwerk te hebben die wordt geëvolueerd uit de data van alle gebruikers. Daarnaast een persoonlijk neuraal netwerk op basis van de eigen gegevens. Een zg 'point cloud' algoritme kan ook gebruikt worden.

''Rondvraag''
Roderik is vaak ziek. Chris gaat bellen om te vragen wat er aan de hand is.
De wiki staat op aardrock.com. Oude wiki is half overgezet.
Durk wilt meer gegevens over datamodel krijgen voor het te maken neuraal netwerk model.

Vergaderingen

2006-04-18T15:26:36Z

Dpkingma: