Cannabis Inflorescence Classification Using FT-NIR and Machine Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for cannabis inflorescence classification, particularly for medicinal cannabis cultivars, are laborious, expensive, and time-consuming, and fail to accurately predict the concentrations of cannabinoids and terpenes, often requiring hazardous solvents and skilled personnel, while neglecting terpene content.
Innovation Solution
A method utilizing Fourier Transform Infrared (FT-NIR) spectroscopy combined with chemometrics and machine learning is developed to classify cannabis cultivars by processing ground inflorescence spectrograms, enabling rapid and accurate prediction of cannabinoids and terpenes without extensive sample preparation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If chromatographic methods (HPLC-PDA, GC-MS) are used for cannabis classification, then measurement precision of cannabinoid concentrations is improved, but loss of time and operational complexity increase significantly
Solution Approach 1:
The patent replaces the mechanical chromatographic separation system (HPLC, GC) with an optical spectroscopic system (FT-NIR). Instead of physically separating and identifying compounds through chromatography, the system uses near-infrared spectroscopy to detect chemical bonds and functional groups directly in the cannabis inflorescence, eliminating the time-consuming separation process while maintaining identification capability.
Solution Approach 2:
The patent extracts and focuses on specific diagnostic spectral regions (1000-2500 nm) that contain information about cannabinoid and terpene chemical bonds. By isolating these critical spectral ranges and using them for classification, the system avoids the need to analyze the entire complex chemical mixture through time-consuming chromatographic separation.
2Measurement precision
If chromatographic methods are used for cannabis classification, then measurement precision of cannabinoid concentrations is improved, but device complexity and operational difficulty increase
Solution Approach 1:
The patent replaces complex mechanical instrumentation (chromatographs requiring pumps, columns, detectors, and extensive sample preparation) with a simpler optical system consisting of an FT-NIR spectrometer that directly measures the sample. This substitution dramatically reduces device complexity while maintaining analytical capability through optical detection of chemical bonds.
3Measurement precision
If conventional classification methods are used, then classification accuracy based on cannabinoid ratio is improved, but terpene content characterization is neglected
Solution Approach 1:
The patent creates a universal FT-NIR spectroscopic method that simultaneously characterizes multiple compound classes (cannabinoids, terpenes, and other secondary metabolites) in a single measurement. The system uses multivariate analysis to extract information about all these compounds from their combined spectral signatures, providing comprehensive chemical characterization without requiring separate analyses for each compound type.
Solution Approach 2:
The patent merges the detection of different chemical compounds (cannabinoids and terpenes) into a single integrated spectroscopic measurement. Instead of analyzing compounds separately through multiple chromatographic runs, the system combines their spectral signals and uses chemometric models to deconvolute and quantify all components simultaneously, preserving terpene information that would otherwise be lost.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach provides a cost-effective, rapid, and accurate technique for cannabis inflorescence classification, replacing laborious chromatographic methods, and allows real-time evaluation of cannabinoid and terpene concentrations, suitable for both consumers and farmers.
Implementation Method 1
The spectrogram is then processed using suitably trained one or more machine learning modules to provide output data on a plurality of cannabinoids and terpenes in the sample
Implementation Method 2
Fourier transform near-infrared spectroscopy (FT-NIR) method uses the near-infrared (i.e., NIR; 700-1100 nm) and short-wave infrared (i.e., SWIR; 1100-2500 nm) regions of the electromagnetic spectrum
Data Source
AI summary
A method and respective system are described. The method provides classification of cannabis inflorescence, and comprising: grinding said cannabis inflorescence; and determining a spectrogram of ground cannabis inflorescence; and providing data indicative of said spectrogram to trained machine learning system, pretrained on classification of material composition of cannabis inflorescence, to thereby obtain output data indicative of at least one of composition of selected cannabinoids and terpenes in said cannabis inflorescence, and varieties of said cannabis inflorescence.