Training Workshop on Machine Learning (ML) Application for Automated Early Detection of Diseases in Edible Plants.

Proposed By: Engr. Renaud Blanchard Makaka, Sr. Technical Manager

Department: COMSTECH Training and Research Center on AI and Emerging Technologies

Venue: Cheikh Anta Diop University (UCAD), Dakar, Senegal

Date: 3-5 November 2026

Application Form Link: Click Here

Flyer in English: Click Here

Flyer in French: Click Here

2. RATIONALE / ALIGNMENT WITH THE CORE MANDATE OF OIC-COMSTECH

Under the COMSTECH Training and Research Center on AI and Emerging Technologies, this initiative aligns with OIC-COMSTECH’s mandate to strengthen scientific capacity, promote technology transfer, foster innovation, and support sustainable development.

By integrating AI, digital agriculture, and plant health diagnostics, the initiative supports climate-smart agriculture, enhances food security, and strengthens agricultural resilience against climate change and emerging plant diseases.

3. BACKGROUND

Agriculture remains a critical pillar of food security, economic growth, and livelihoods across many developing countries, particularly within OIC Member States in Africa and Asia. However, plant pests and diseases continue to pose a major threat to agricultural productivity, food systems, and rural incomes. According to the Food and Agriculture Organization of the United Nations (FAO, 2024), plant pests and diseases are responsible for up to 40% of global crop losses annually, resulting in economic losses exceeding USD 220 billion each year. Climate change, increasing trade, and the emergence of new pathogens are further accelerating the spread and severity of plant diseases.

Cassava (Manihot esculenta) is not simply a food crop in Africa; it is a strategic staple for food security, rural livelihoods and household income, particularly across Sub-Saharan Africa. The crop is cultivated in around 40 African countries and has become deeply integrated into smallholder farming systems, especially in areas affected by drought, poor soils and climatic variability. According to the Food and Agriculture Organization of the United Nations (FAO), more than 70 million people in Africa depend on cassava as a primary source of food, while the International Institute of Tropical Agriculture (IITA) estimates that cassava supports the livelihoods of more than 300 million Africans. Cassava is particularly important to smallholder farmers because of its ability to produce reasonable yields under difficult agroecological conditions, its tolerance to drought and marginal soils, and its flexibility in harvesting and storage. Beyond household consumption, cassava provides an important source of cash income: smallholder farmers commonly sell part of their production, while processing and marketing activities create additional economic opportunities along the value chain. FAO further identifies cassava as simultaneously a major staple, a low-cost source of carbohydrates and a source of cash income for producing households, making the crop particularly relevant to poverty reduction and rural economic resilience.

This strategic importance, however, is increasingly threatened by Cassava Mosaic Disease (CMD), one of the most destructive and widespread viral diseases affecting cassava production in Africa. CMD is caused by a group of cassava mosaic begomoviruses, including the African cassava mosaic virus (ACMV) and East African cassava mosaic virus (EACMV), and is primarily transmitted by the whitefly Bemisia tabaci as well as through the use of infected planting material. The disease causes characteristic mosaic patterns, leaf deformation, chlorosis, stunted plant growth and, in severe cases, substantial reductions or complete loss of root yield. FAO estimates that at least 30% of Africa’s cassava crop approximately 45 million tonnes is lost annually to CMD, underlining the scale of its threat to food security and rural livelihoods.

Early detection and timely management of plant diseases are essential to minimizing crop losses and safeguarding food systems. Traditional disease surveillance methods, which rely largely on visual inspection by trained experts, are often time-consuming, costly, and difficult to scale, particularly in rural and resource-constrained settings. In contrast, recent advances in Emerging technologies, AI, Machine Learning (ML), ComputerVision, IoT, GIS etc and laboratory diagnostics have emerged as complementary tools, offering transformative opportunities to automate disease detection through the analysis of images captured using smartphones, drones, or field-based sensors. However, ML-based detection tools serve primarily as screening mechanisms and require laboratory confirmation. Molecular and serological diagnostic techniques, particularly Polymerase Chain Reaction (PCR), provide high sensitivity and specificity for confirming plant pathogens.

International plant health frameworks promoted by the International Plant Protection Convention (IPPC) recommend a layered diagnostic approach that combines field-level detection with confirmatory laboratory testing. In line with these frameworks, this training workshop adopts an integrated early detection approach, strengthening capacities across the full diagnostic continuum rather than treating digital and laboratory methods as isolated solutions. The workshop aims to build participants’ practical and applied capacities to understand, design, and deploy Machine Learning solutions and laboratory diagnostic techniques for plant disease detection, while drawing lessons applicable to other strategic crops across OIC Member States.

The workshop aims to demonstrate how existing Machine Learning (ML) and Artificial Intelligence (AI) algorithms can support plant scientists in the early detection and diagnosis of crop diseases. The hands-on component will not focus on developing new ML/AI algorithms or protocols from scratch. Instead, participants will be introduced to the practical use of existing AI-based tools and models, including basic steps such as image acquisition, data preparation, disease classification, model use, and interpretation of results.

ARTIFICIAL INTELLIGENCE (AI)/ ML TOOLS FOR DETECTING CROP DISEASES

The detection and identification of diseases on crops at early stages and during the growing stages combine computer vision techniques, machine learning algorithms, deep learning, use of drone technologies and satellite imagery data to identify plant diseases at an early and growing stages. The benefits of this technology are leading to higher crop yield and sustainable economy and agriculture and contempt global food shortage and crisis.

Figure 1 Shows a basic architecture how AI technology uses Algorithms and machine learning tools to identify a healthy crop. The healthy crop is classified and identified based on the features identified from the original healthy plants. The Tool or the algorithm is trained to classify the healthy crop based on supplied data. Figure 2 shows a section of digital image for plants captured to be used as an input in the Mobile AI application while Figure 3 is displaying drone technology using deep learning tools to identify and classify healthy crops.

Figure 1: Block Architecture showing how AI techniques and algorithms detect disease on crops

Figure 2 : A digital Image of a plant section captured to be inputted in AI algorithm for disease identification

Figure 3 : Drone technology configured with Machine and Deep Learning tools classifying healthy parts of crops

Table 1 shows major Artificial Intelligence tools and technologies that are efficiently used for crop diseases detection and classification of healthy plants from non-healthy ones.

Overview of Laboratory Techniques

The practical laboratory session will introduce participants to the molecular detection of Cassava Mosaic Disease (CMD). It will provide a practical overview of key diagnostic steps, including plant sample preparation, DNA extraction, PCR principles, and interpretation of results. The session will also highlight how molecular diagnostics can complement visual and AI/ML-based disease detection.

Standard laboratory biosafety procedures, including PPE, equipment and surface disinfection, appropriate sample handling, and biological waste management, will be observed.

Participant Selection Criteria

• Relevant academic background in plant sciences, agriculture, agronomy, biology, biotechnology, phytopathology, computer science, AI/ML, data science, or related fields;
• Professional or research experience related to crop health, plant disease management, agricultural technologies,
• Demonstrated interest in applying AI/ML tools to agricultural and plant health challenges;
• Relevance of the workshop to the participant’s current professional or research activities;
• Potential to apply and disseminate the knowledge acquired within their home institution;
• Priority may be given to participants from OIC Member States and neighboring West African countries, ensuring appropriate geographical and institutional diversity.

In addition, the practical session in Lab will involve 10 selected participants based on the following criteria:

• Academic background in biology, agronomy, phytopathology, biotechnology, plant sciences, or related fields;
• Demonstrated experience or interest in plant disease diagnosis;
• Basic laboratory knowledge and experience considered an advantage;
• Clear relevance of the training to the participant’s professional or research activities;
• Potential and commitment to replicate and share the knowledge acquired within their home institution.

4. OBJECTIVES

• Strengthen capacities in the early detection and diagnosis of crop diseases through the practical use of existing AI/ML tools, complemented by laboratory-based diagnostic methods.
• Promote the adoption of AI-driven agriculturalsolutions for crop health monitoring, early disease detection, improved productivity, and farmers’ livelihoods.
• Facilitate knowledge exchange and scientific collaboration among plant scientists, AI/ML specialists, laboratory experts, and agricultural professionals.

5. STRATEGIC SIGNIFICANCE

• Supports food security and agricultural resilience in West Africa.
• Promotes the use of emerging technologies in agriculture (smart agriculture.)
• Strengthens laboratory and research capacities for plant disease diagnostics.
• Encourages South-South scientific cooperation and technology transfer.

6. EXPECTED IMPACTS

1. Improved technical skills in AI-based crop disease detection and laboratory diagnostics.
2. Enhanced disease surveillance and early warning systems for major food crops.
3. Strengthened institutional capacities in plant health research and diagnostics.
4. Increased collaboration among agricultural research institutions across OIC Member States.

7. TARGET PARTICIPANTS

7.1 The training workshop targets:

• Plant protection researcher, data scientists and farmers;
• Phytopathologist, Specialized agronomist.

7.2 Number of participants: 40 in person; +100 online

8. METHODOLOGY AND TRAINING APPROACH

The workshop will adopt a hands-on, practice-oriented approach, combining:

• Expert-led lectures and technical presentations;
• Practical lab sessions and case studies;
• Demonstrations of existing AI-based agricultural tools;
• Interactive discussions and knowledge-sharing sessions.

9. EXPECTED OUTCOMES

1. Enhanced Technical Capacity in Disease Early Detection
2. Improved Application of Machine Learning, IoT, GIS system for Field-Based Screening
3. Strengthened Competence in Laboratory Diagnostics
4. Effective Integration of Field and Laboratory Detection Methods
5. Improved Disease Surveillance and Early Warning Systems
6. Enhanced Institutional Coordination and Collaboration
7. Action-Oriented National Follow-Up and Implementation

10. PARTNERS WITH THE FOCAL NODE

Lead Organization

• OIC-COMSTECH

Partners

• Islamic Organization for Food Security (IOFS)
• OIC GS

Host Institution

• Cheikh Anta Diop University (UCAD)

Other Partners

• International Institute of Tropical Agriculture (IITA)
• Central And West African Virus Epidemiology (Wave) TBC

11. PROPOSED FOLLOW-UP

• Establish a group of experts in ML for agriculture and plant disease diagnostics.
• Zoom recording of the training session.

12. CONTACTS / OUTREACH

All official communications and outreach will be made through OIC-COMSTECH’s official channels and shared on the partners’ pages (UCAD, OIC, IOFS), National Focal Points and Young Affiliates.

OIC-COMSTECH Secretariat

Focal Person
Engr. Renaud Blanchard Makaka
Senior Technical Manager, OIC-COMSTECH
renaudmakaka@comstech.org

IOFS
Focal Person :
Dr. Abdelaziz Hajjaji, Program Manager, IOFS,
abdelaziz@iofs.org

OIC GS
Mr. Arman Tynybek , Programme Officer
atynybek@oic-oci.org

Cheikh Anta Diop University of Dakar (UCAD)
Prof. Alioune Dior FALL
Head of the Pharmacognosy and Botany Laboratory
Faculty of Medicine and Pharmacy
Email: alioune.fall@ucad.edu.sn

13. MEDIUM OF TRAINING: French and English