Quantifying cannabis towards optimizing personal consumption

I love data. I’m a firm believer in quantification, I also believe that cannabis is a unique experience and how one flower affects one user may differ for another. I want to understand why that is and I want as many data points as I can manage to get to forwarding that understanding.

All indications look to be I’ll be setup in a space and testing soon, with a focus on the home growers market I expect to be testing a lot of plants. I also expect to be hearing all about it from tons of people. I’m really excited about the possibly of having data on cannabinoids and terpenes, maybe NPK.

I’ve been thinking about what all that data could be useful for in the grander scheme, what might be the best way to make this data available to people so we may better empower personalized cannabis? Does it exist already?

1 Like

You said data a lot but I’m not clear on what you’re talking about doing exactly? What data are you collecting?

Correlating user experience with flower quantification?

4 Likes

Focusing on terpenes and cannabinoids generally speaking. I think those would be the primary data points to be collected and referenced back to the strain/genetics. I’ll be looking to perform things like plant nutrient testing and the like. I’m not certain how accurate these numbers will be so I am hesitant to want to include them at the moment.

Maybe, but not at a ‘flower’ level ideally. I envision, a user might report, I smoked this batch of flower and noticed these effects, but the goal would be to correlate the effects observed with the individual compounds to build a profile that isn’t flower specific and can be identified in flowers or concentrates. Maybe with this data we can build profiles of useful effects to the individual, and suggest strains based on previous testing history. Kind of like the user backed suggested content systems of netflix and others.

I’m curious to see how varied 3 - 10 batches of “jack herer” are from one another, grower to grower, batch to batch. I can say for myself that some grows and batches were on point effect wise and others not. What was the difference? :man_shrugging: “Jack herer” and THC% isn’t useful enough to medicate adequately IMO and I think we all deserve better. I want to know more about how/when my medicine best suits me. I want to guess less, so my choices feel less random or hit and miss, and I want to enable that for everyone who wants that for themselves.

5 Likes

Got it… this is a good idea.

4 Likes

Cool. How do we help?

1 Like

I’m not exactly sure. I think this general conversation is helpful in flushing out what all we may be able to do with the data, as well as how to make that data as meaningful as possible (such as restricting effects data to those who provided the sample, or otherwise have some kind of authentication towards that to keep the data as clean as possible)

The machines will provide data on cannabinoids and terpenes, whatever else we may glean from the data they provide. I think it would be helpful to brainstorm meaningful data points to collect when a sample is collected from the grower.

The obvious ones to me would be strain name/genetics/lineage information. As we should primarily be collecting from home growers, maybe ancillary information about grow medium, feeding, lights? Idk what really matters here but I do think the questions should be focused as to not make it overly complicated for the person to answer.

Number of days in flower along with details of lighting.

1 Like

Cannabis concentrates helped my wife kick Tramadol. It be great to have a better understanding of what constituents were the most effective. I suspect the more couch lock type noids.

2 Likes

You will definitely want to correlate user experience with terpene profiles, cannabinoid profiles, and concentrations.

Assembling the data set and running AI could get you info on entourage.

Protip: be prepared for multiple neurotypes which will result in different effects experienced for the same flower.

4 Likes

We need terpene allergen tests for this. This issue of people liking things that they’re slightly allergic to is important to this specific “personal taste” data.

1 Like

Do these tests exist today?

No idea. Statement is true regardless I would say, I don’t know for a fact that the endorphin-cortisol response pertains to inhalation as a means of ingestion, but I can assure that the information is pertinent.

1 Like

There’s just no prepping some folks for that shit :rofl:

Personally I’m surprised the damn thing (neural network in our heads) works at all.

If the claim that we got to this design by accident is to be believed, the concept that they don’t all work the same way seems like a given.

I’d also suggest listening to the fungi…they’ve been doing networks way longer.

4 Likes

1 Like

The data from the Releaf App / project may be of use. No idea on availability, just remember running into them at a cannabis industry event.

“Releaf is a patent pending app that enables you to anonymously track real-time and historic experiences with specific cannabis & CBD products. Easily track the product you’re using, where you purchased it from, why you are using it, relatively how much you used, as well as symptom relief, feelings, and side effects you experienced. The more product use and outcomes you share with Releaf, the more intelligent the reports are that Releaf will provide back to you.”

Crowd sourcing > AI in this case at this time…

2 Likes

I think this will be a good base to work from and possibly expand upon?

Although I don’t imagine many would participate in such depth, I was thinking about multi layered reporting. For example, any supplements you may be taking with or at the time of administration. As an example, I have noticed that taking inositol with hhc seems to potentiate its effects on me. Thats important data that isn’t readily available as far as I am aware.

I think limiting reviewers of effects to those having consumed specific batches that have been tested and quantified, my hope in this, is that correlation between effects and quantifiable attributes can be more clearly observed.

1 Like