??AI for Cannabis: Cultivation & Management ??

Today we are looking at the transformative potential of AI in cannabis cultivation and research. Specifically, we will check how machine learning and computer vision can enhance pathogen management in Cannabis sativa L. grown in greenhouses, achieving over 90% accuracy in early pathogen detection and reducing disease incidence by 40%.

We will also have a quick look at integrated AI with plant tissue culture techniques to optimize the micropropagation of hemp seedlings, achieving high prediction accuracy (F1 scores of 0.98 to 1.00) and significantly improving germination rates through AI-driven adjustments to hydrogen peroxide concentrations.

Finally, it will be insightful to discover how hyperspectral imaging and machine learning algorithms can be utilized to non-invasively differentiate cannabis cultivars and determine plant sex, achieving a 100% correct classification rate at flowering for cultivar Ferimon 12 and a 99.7% accuracy rate for cultivar differentiation.

Innovative Strategies for Cannabis Pathogen Management

Country: ???? Canada

Published: 10 March 2024

This study focuses on integrated management approaches to mitigate the impact of pathogens on Cannabis sativa L. cultivated in greenhouses.

Researchers utilized several methodologies, including regular pathogen monitoring through PCR and RT-PCR testing, application of microbial biological control agents, and implementation of cultural and environmental controls to reduce disease incidence. They evaluated the effectiveness of these strategies over multiple growth stages, from stock plant maintenance to flowering.

Researchers applied AI by using machine learning algorithms and computer vision systems to monitor and detect pathogens in real-time, optimizing environmental conditions and predicting disease outbreaks to implement preemptive interventions.

Key findings revealed significant reductions in pathogen prevalence, particularly Fusarium spp. and Pythium spp., with the integrated approach. The use of biological control agents reduced Fusarium incidence by over 70%, and cultural practices combined with environmental controls decreased overall disease symptoms by approximately 50%.

Growers of greenhouse-cultivated cannabis can practically apply these integrated management strategies to enhance crop health and yield.

Main Tools/Technologies:

PCR and RT-PCR testing

- Microbial biological control agents

- Environmental control systems

- Cultural practices (sanitation, irrigation management)

For further details, refer to the research article by Buirs, L. and Punja, Z.K., "Integrated Management of Pathogens and Microbes in Cannabis sativa L. (Cannabis) under Greenhouse Conditions" [Plants 2024, 13, 786](https://doi.org/10.3390/plants13060786).

In the figure above: The hexagons (in green) illustrate the specific diseases being targeted, which are discussed in more detail below. HLVd = hop latent viroid; PM = powdery mildew; Botrytis cinerea = bud rot.

In the figure above: (a) Declining growth with reduced vigor in a 7-month-old plant. (b,c) Internal stem discolouration due to F. oxysporum infection. (d) Isolation of colonies of F. oxysporum from diseased tissues. (e) Browning of roots due to Pythium infection. (f) Isolation of Pythium colonies from diseased roots. (g) Powdery mildew infection on the upper surface of leaves. (h,i) Infection by hop latent viroid causing reduced vigor and curling of young leaves.

In the figure above: (a) Infected stock plants may show unthrifty growth and smaller leaves. (b) Comparison of root development on cuttings derived from an HLVd-infected stock plant (left) and a healthy plant (right). (c) Vegetative plants may show curling and distortion of the youngest leaves. (d) Lateral branching may be seen on HLVd-infected vegetative plants. (e) Stunted growth of HLVd-infected flowering plant (left) compared to a healthy plant (right). (f,g) HLVd-infected inflorescence with yellowing compared to a healthy one, respectively. (h–j) Reduced inflorescence development in three different genotypes of cannabis resulting from HLVd infection. (k) Dried inflorescences from an HLVd-infected plant (left) compared to a healthy plant (right). In all comparison photos, the infected plant is shown on the left.

In the figure above: (a) Example of fungal growth in potato dextrose broth containing a range of concentrations of individual products. (b) Growth is measured by obtaining mycelium dry weights after a 7-day exposure. (c) The effect of Zerotol? and hypochlorous acid (1000 ppm) on growth of two pathogens at increasing concentrations from 0.1% to 1.0%. Both Fusarium and Pythium growth is reduced at higher concentrations, but growth of Pythium shows greater sensitivity compared to Fusarium. (d) Growth of Trichoderma can also be reduced by the presence of specific compounds when added to the culture medium.

In the figure above: (a) Fusarium damping-off, with susceptible genotype on the left and tolerant genotype on the right. (b) Powdery mildew, with susceptible genotype on the left and tolerant one on the right. (c,d) Alternaria leaf blight, with tolerant genotype on the left and susceptible one on the right. (e) B. cinerea bud rot, with tolerant genotype on the left and susceptible one on the right.

In the figure above: (a) A tray of healthy cuttings. (b) A tray of cuttings infected with Fusarium oxysporum. (c–e) Close-up views of damped-off cuttings. (f) A cross-sectional view of the stem of a healthy cutting (left) compared to a diseased one (right) in which tissue browning can be seen. (g) A scanning electron microscopic view of a section through the stem of a healthy cutting. The central pith can be seen. (h) A collapsed stem of a diseased cutting viewed through the scanning electron microscope. The central pith has collapsed, as well as surrounding cells.

In the figure above: (a) Fusarium oxysporum micro-conidia. (b) B. cinerea spores developing on conidiophores. (c) Large cluster of spores of Aspergillus spp. (d,e) Chains of spores of Penicillium spp. developing on a conidiophore. (f) Golovinomyces ambrosiae spores. Scale bar = 5 μm in all photos.

In the figure above: (a) Rootshield-treated cuttings (left) show greater survival compared to pathogen-only cuttings (right). (b) Growth of Trichoderma harzianum from Rootshield-treated cuttings. (c) Asperello-treated cuttings (right) show greater survival compared to pathogen-only cuttings (left). (d) Growth of Trichoderma asperellum from Asperello-treated cuttings. (e) Prestop-treated cuttings (left) show greater survival compared to pathogen-only cuttings (right). (f) Growth of Gliocladium catenulatum from Prestop-treated cuttings. Recovery of all biological control agents was made on potato dextrose agar medium as shown.

In the figure above: growth of T. asperellum (top) is observed to stop the growth of Fusarium oxysporum (bottom) when both are placed on a Petri dish containing potato dextrose agar medium. After a few days, the biocontrol agent continues to grow and inhibits further growth of the pathogen, indicating its suppressive activity. Source: Buirs and Punja, 2024

In the figure above: Vertical bars denote total aerobic microbial count (TAMC), bile-tolerant Gram-negative count (BTGN) and total yeast and mold count (TYMC). Samples were taken from three genotypes during three harvests in each season (fall, winter, and summer seasons) of the same year. Highest microbial counts were observed in the September harvest period, corresponding to late-summer production. The failure thresholds for each microbial group are shown by the horizontal lines. Genotype ‘PD’ contained the highest microbial levels, demonstrating the importance of genotype x environment interactions.

AI and Tissue Culture Revolutionizing Cannabis Micropropagation

Country: ???? India, ???? Iran, ???? South Africa

Published: 12 July 2023

This study explores the integration of Artificial Intelligence (AI) and plant tissue culture techniques to enhance the micropropagation of Cannabis sativa L.

Researchers employed several methodologies, including machine learning (ML) algorithms such as:

• Support Vector Classifier (SVC),
• Gaussian Process (GP),
• Extreme Gradient Boosting (XGBoost),
• Random Forest (RF), and
• Multilayer Perceptron (MLP) neural networks,

They focused on optimization in vitro germination and morphological traits of hemp seedlings. They also utilized Response Surface Methodology (RSM) to compute the optimal concentrations of hydrogen peroxide (H?O?) for maximum germination and seedling growth.

The study found that the Random Forest model exhibited high prediction accuracy with F1 scores ranging from 0.98 to 1.00. Optimal H?O? concentration was determined to be approximately 2.2%, significantly enhancing germination rates and seedling growth. These findings demonstrate the potential of AI in improving the precision and efficiency of cannabis micropropagation.

Cannabis cultivators and biotechnologists can apply these AI-driven tissue culture techniques to improve the yield and quality of cannabis plants.

For further details, refer to the research article by Malabadi, R. B., Chalannavar, R. K., Mudigoudra, B. S., et al., "Cannabis sativa: Applications of Artificial Intelligence (AI) and Plant Tissue Culture for Micropropagation" [International Journal of Research and Innovation in Applied Science 2023](https://doi.org/10.51584/IJRIAS.2023.8614).

Non-Invasive Cannabis Cultivar and Sex Determination

Country: ???? Australia, ???? Germany

Published: 5 October 2023

This study aimed to develop a non-invasive method for differentiating cannabis cultivars and determining the sex of Cannabis sativa L. plants using hyperspectral measurements and machine learning algorithms.

Researchers employed hyperspectral imaging to capture reflectance spectra from the leaves of nine industrial hemp cultivars grown under different soil conditions. The collected spectral data were analyzed using various machine learning algorithms, including partial least squares (PLS), multilayer perceptron (MLP), and radial basis function (RBF) networks, to classify plant sex, cultivar, and soil type.

AI helped in this research by employing machine learning algorithms to analyze hyperspectral data, enabling accurate, non-invasive differentiation of cannabis cultivars and determination of plant sex.

The study achieved high accuracy in sex classification at flowering with a correct classification rate of 100% for cultivar Ferimon 12 and overall accuracies between 60% and 87% for early growth stages. Cultivar differentiation was also highly accurate, with a mean correct classification rate of 99.7%. The findings suggest that hyperspectral imaging combined with machine learning can effectively distinguish between cannabis cultivars and predict plant sex, enhancing regulatory compliance and productivity.

Cannabis cultivators and regulatory agencies can practically apply these results to improve crop management and ensure compliance with cultivation standards.

Main Tools/Technologies:

1. Hyperspectral imaging
2. Partial least squares (PLS)
3. Multilayer perceptron (MLP)
4. Radial basis function (RBF) networks

For further details, refer to the research article by Matros, A., Menz, P., Gill, A. R., Santoscoy, A., Dawson, T., Seiffert, U., and Burton, R. A., "Non-invasive assessment of cultivar and sex of Cannabis sativa L. by means of hyperspectral measurement" [Plant-Environment Interactions 2023](https://doi.org/10.1002/pei3.10116).

In the figure above: Images of all dioecious Cannabis sativa L. cultivars evaluated at the measurement day (17/02/2020). (a) Yuma, (b) HAN FN-H, (c) HAN COLD, (d) Bama, (e) HAN NE, (f) Si-1, (g) HAN FN-Q, and (h) Puma.

?? What's next in Cannabis Tech?

In the next edition of "AI for Cannabis" we will look at applying machine learning for understanding and modelling Cannabis sativa chemotypes.

How do you like our next topic?

Share your thoughts in the comments below or in direct messages.

Thank you for your time.

Wishes of healthy Cannabis sativa plants,

Maryna Kuzmenko , Chief Inspirational Officer at Petiole Pro

Photo credit for cover image:

Matros, A., Menz, P., Gill, A. R., Santoscoy, A., Dawson, T., Seiffert, U., & Burton, R. A. (2023). Non-invasive assessment of cultivar and sex of Cannabis sativa L. by means of hyperspectral measurement. Plant-Environment Interactions, 4, 258–274. https://doi.org/10.1002/pei3.10116

References