Accuracy of a Smartphone-Based Object Detection Model, PlantVillage Nuru, in Identifying the Foliar Symptoms of the Viral Diseases of Cassava-CMD and CBSD.

Nuru is a deep learning object detection model for diagnosing plant diseases and pests developed as a public good by PlantVillage (Penn State University), FAO, IITA, CIMMYT, and others. It provides a simple, inexpensive and robust means of conducting in-field diagnosis without requiring an internet connection. Diagnostic tools that do not require the internet are critical for rural settings, especially in Africa where internet penetration is very low. An investigation was conducted in East Africa to evaluate the effectiveness of Nuru as a diagnostic tool by comparing the ability of Nuru, cassava experts (researchers trained on cassava pests and diseases), agricultural extension officers and farmers to correctly identify symptoms of cassava mosaic disease (CMD), cassava brown streak disease (CBSD) and the damage caused by cassava green mites (CGM). The diagnosis capability of Nuru and that of the assessed individuals was determined by inspecting cassava plants and by using the cassava symptom recognition assessment tool (CaSRAT) to score images of cassava leaves, based on the symptoms present. Nuru could diagnose symptoms of cassava diseases at a higher accuracy (65% in 2020) than the agricultural extension agents (40-58%) and farmers (18-31%). Nuru's accuracy in diagnosing cassava disease and pest symptoms, in the field, was enhanced significantly by increasing the number of leaves assessed to six leaves per plant (74-88%). Two weeks of Nuru practical use provided a slight increase in the diagnostic skill of extension workers, suggesting that a longer duration of field experience with Nuru might result in significant improvements. Overall, these findings suggest that Nuru can be an effective tool for in-field diagnosis of cassava diseases and has the potential to be a quick and cost-effective means of disseminating knowledge from researchers to agricultural extension agents and farmers, particularly on the identification of disease symptoms and their management practices.

A Mobile-Based Deep Learning Model for Cassava Disease Diagnosis.

Convolutional neural network (CNN) models have the potential to improve plant disease phenotyping where the standard approach is visual diagnostics requiring specialized training. In scenarios where a CNN is deployed on mobile devices, models are presented with new challenges due to lighting and orientation. It is essential for model assessment to be conducted in real world conditions if such models are to be reliably integrated with computer vision products for plant disease phenotyping. We train a CNN object detection model to identify foliar symptoms of diseases in cassava (Manihot esculenta Crantz). We then deploy the model in a mobile app and test its performance on mobile images and video of 720 diseased leaflets in an agricultural field in Tanzania. Within each disease category we test two levels of severity of symptoms-mild and pronounced, to assess the model performance for early detection of symptoms. In both severities we see a decrease in performance for real world images and video as measured with the F-1 score. The F-1 score dropped by 32% for pronounced symptoms in real world images (the closest data to the training data) due to a decrease in model recall. If the potential of mobile CNN models are to be realized our data suggest it is crucial to consider tuning recall in order to achieve the desired performance in real world settings. In addition, the varied performance related to different input data (image or video) is an important consideration for design in real world applications.

Community phytosanitation to manage cassava brown streak disease.

Cassava viruses are the major biotic constraint to cassava production in Africa. Community-wide action to manage them has not been attempted since a successful cassava mosaic disease control programme in the 1930s/40s in Uganda. A pilot initiative to investigate the effectiveness of community phytosanitation for managing cassava brown streak disease (CBSD) was implemented from 2013 to 2016 in two communities in coastal (Mkuranga) and north-western (Chato) Tanzania. CBSD incidence in local varieties at the outset was >90%, which was typical of severely affected regions of Tanzania. Following sensitization and monitoring by locally-recruited taskforces, there was effective community-wide compliance with the initial requirement to replace local CBSD-infected material with newly-introduced disease-free planting material of improved varieties. The transition was also supported by the free provision of additional seed sources, including maize, sweet potato, beans and cowpeas. Progress of the initiative was followed in randomly-selected monitoring fields in each of the two locations. Community phytosanitation in both target areas produced an area-wide reduction in CBSD incidence, which was sustained over the duration of the programme. In Chato, maximum CBSD incidence was 39.1% in the third season, in comparison with an incidence of >60% after a single season in a control community where disease-free planting material was introduced in the absence of community phytosanitation. Kriging and geospatial analysis demonstrated that inoculum pressure, which was a function of vector abundance and the number of CBSD-infected plants surrounding monitored fields, was a strong determinant of the pattern of CBSD development in monitored fields. In the first year, farmers achieved yield increases with the new varieties relative to the local variety baseline of 94% in Chato (north-west) and 124% in Mkuranga (coast). Yield benefits of the new material were retained up to the final season in each location. The new variety (Mkombozi) introduced under community phytosanitation conditions in Chato yielded 86% more than the same variety from the same source planted in the no-phytosanitation control location. Although there was an 81% reduction in CBSD incidence in the new variety Kiroba introduced under community phytosanitation compared to control conditions in Mkuranga, there was no concomitant yield increase. Variety Kiroba is known to be tolerant to the effects of CBSD, and tuberous roots of infected plants are frequently asymptomatic. Community phytosanitation has the potential to deliver area-wide and sustained reductions in the incidence of CBSD, which also provide significant productivity gains for growers, particularly where introduced varieties do not have high levels of resistant/tolerance to CBSD. The approach should therefore be considered as a potential component for integrated cassava virus management programmes, particularly where new cassava plantations are being established in areas severely affected by CBSD.