1. V3
1.1. Recommendation
1.1.1. Meal Plan
1.1.2. Fulfillment and Ordering
1.2. API
1.3. Queries
1.3.1. Tell me 5 things i should eat to make me healthier
1.3.2. If a person eats xyz foods, what supplements should he take
1.3.3. If person X has abc health goals and current blood reports is..../ what supplement/ food should
1.3.4. If person X has XYZ disease and BMI what food suppplemets should he take
2. Data and Data Processing
2.1. V2 additional datasets
2.1.1. V2 - Microbiome Data
2.1.1.1. Pathways dataset
2.1.1.2. probiotics dataset
2.1.2. Medication Data
2.1.3. Survey
2.1.4. UX Data
2.1.4.1. Current food searched for
2.1.5. histamine dataset
2.2. User Provided Data
2.2.1. Query data
2.2.1.1. breakfast +drink
2.2.1.2. 7 fruits
2.3. connect datasets
2.3.1. connect patient with disease Profile
2.3.2. connect recipe and (food) charmee taxonomy
2.3.3. connect USDA ingredient and recipe
2.3.4. User profile Allergy with List of Allergy with Ingredients
2.3.5. RDA data with patient and FDA profile
2.4. Pipeline Data Quality
2.4.1. errors and issues
3. V2 - what we should do
3.1. Team
3.1.1. Stack competency across the team
3.1.2. Regular clinical input
3.1.3. popular foods
3.1.4. Metabolic syndrome disorders
3.1.5. Team Structure
3.1.5.1. Data and Algorithms
3.1.5.1.1. Sourcing and Data Gaps
3.1.5.1.2. algorithms
3.1.5.1.3. Cleaning
3.1.5.1.4. query
3.1.5.1.5. documentation
3.1.5.1.6. QC and boundary conditions
3.1.5.2. Data Engineering Team
3.1.5.2.1. UI
3.1.5.2.2. aws
3.1.5.2.3. security
3.1.5.2.4. tools
3.1.5.2.5. documentation
3.1.5.2.6. QA
3.2. Process
3.2.1. Sprint planning
3.3. Questions
3.3.1. RDA replaced by?
3.4. evaluate fndds and legacy for completion
4. Plan for KG
4.1. Jeremy View
4.1.1. Plan to get to SAMD in 2-3 years
4.1.2. Current Version : Food recommender
5. Infrastructure and Security
5.1. UPDATED dataset
5.2. stress test to review performance
5.3. obligatory .authentication
5.4. daily backups
5.5. neo4j bloom enterprise license ?
5.6. code cicd
6. Explainer and Reasoning System ( V2 and V3)
6.1. based on RDA
6.2. Disease Reco. agency specifications
6.3. Allergies
6.4. Intolerances
6.5. Food Score
6.6. Why is the system recommending this FOOD to me
6.7. Datasources
6.7.1. based on blood report
6.7.1.1. High cholestrol
6.7.1.2. high liver enzymes
6.8. v2
6.8.1. how will it benefit me to eat this
6.8.2. how much should i eat
6.9. Food Preferences
6.10. USDA List
7. Timelines
7.1. July
7.1.1. Improve Security
7.1.2. Complete food Matching for recipe and FDA
7.1.3. Data: Begin Loading Microbiome data and connecting with NEO4J and Symptoms data
7.1.4. Begin Explainer Design Development
7.1.5. develop tagging plan
7.1.6. develop a Data quality and assessment plan
7.1.7. complete taxonomy and automated genration and linkage of taonomy
7.1.8. Validate matching for
7.1.8.1. disease rules
7.1.8.2. scoring generation
7.2. August
7.2.1. complete and automate scoring
7.2.2. assess and expand food database and complete development
7.2.3. build further competence in KNIME and NEO4J
7.2.4. release explainer version 1
7.2.5. release food scoring with diseases version 1
7.2.6. infra planning for version 2
7.3. September
7.3.1. completion of UX development
7.3.2. Explainer version 2
7.3.3. scroring version 2
7.3.4. Begin API building
7.3.4.1. Integrate with BIO+ME
7.3.5. QA
7.3.5.1. Pipeline
7.3.5.2. Query quality
7.3.5.3. is all the data getting loaded
7.3.5.4. are we missing key data
7.3.5.5. how do we measure prediction quality
7.4. October
7.4.1. Testing / QA and refinement
7.5. November 1st
7.6. demo validation points for each month
8. Feedback from presentation
8.1. No useful answers returned , how to manage
8.2. Can we turn a unhealthy recipe into a healthy recipe
8.3. revisit food ingredient matching
8.3.1. preparation methods
8.4. each recipe needs a micro and macro nutrient calc
8.5. for further research
8.5.1. ingredient substitution
8.5.2. RDA
8.5.2.1. WHAT IS THE RIGHT REF DATASET and right approach
8.5.3. meal and nutrition planning
8.5.3.1. shreyas
8.5.4. food scoring to include diseases