David Hughes

David Hughes


Research Focus & Metrics

Academic Career Length

21Years (2006-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
27 October 2017 in Frontiers in Plant Science
Verified
Journal Article
Open access
ARAmanda Ramcharan
Amanda Ramcharan
KBKelsee Baranowski
Kelsee Baranowski
PMPeter McCloskey
Peter McCloskey
BABabuali Ahmed
Babuali Ahmed
JLJames Legg
James Legg
DHDavid Hughes
David Hughes
Cassava is the third largest source of carbohydrates for human food in the world but is vulnerable to virus diseases, which threaten to destabilize food security in sub-Saharan Africa. Novel methods of cassava disease detection are needed to support improved control which will prevent this crisis. Image recognition offers both a cost effective and scalable technology for disease detection. New deep learning models offer an avenue for this technology to be easily deployed on mobile devices. Using a dataset of cassava disease images taken in the field in Tanzania, we applied transfer learning to train a deep convolutional neural network to identify three diseases and two types of pest damage (or lack thereof). The best trained model accuracies were 98% for brown leaf spot (BLS), 96% for red mite damage (RMD), 95% for green mite damage (GMD), 98% for cassava brown streak disease (CBSD), and 96% for cassava mosaic disease (CMD). The best model achieved an overall accuracy of 93% for data not used in the training process. Our results show that the transfer learning approach for image recognition of field images offers a fast, affordable, and easily deployable strategy for digital plant disease detection.

20 March 2019 in Frontiers in Plant Science
Verified
Journal Article
Open access
PDF file
ARAmanda Ramcharan
Amanda Ramcharan
PMPeter McCloskey
Peter McCloskey
KBKelsee Baranowski
Kelsee Baranowski
NMNeema Mbilinyi
Neema Mbilinyi
LMLatifa Mrisho
Latifa Mrisho
DHDavid Hughes
David Hughes
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.

1 January 2019 in Current Biology
Verified
Journal Article
Closed access
DHDavid Hughes
David Hughes
FLFrederic Libersat
Frederic Libersat

2 October 2017 in Scientific Reports
Verified
Journal Article
Open access
PDF file
CBCharissa de Bekker
Charissa de Bekker
RORobin A Ohm
Robin A Ohm
HEHarry C. Evans
Harry C. Evans
ABAndreas Brachmann
Andreas Brachmann
DHDavid Hughes
David Hughes
Much can be gained from revealing the mechanisms fungal entomopathogens employ. Especially intriguing are fungal parasites that manipulate insect behavior because, presumably, they secrete a wealth of bioactive compounds. To gain more insight into their strategies, we compared the genomes of five ant-infecting Ophiocordyceps species from three species complexes. These species were collected across three continents, from five different ant species in which they induce different levels of manipulation. A considerable number of (small) secreted and pathogenicity-related proteins were only found in these ant-manipulating Ophiocordyceps species, and not in other ascomycetes. However, few of those proteins were conserved among them, suggesting that several different methods of behavior modification have evolved. This is further supported by a relatively fast evolution of previously reported candidate manipulation genes associated with biting behavior. Moreover, secondary metabolite clusters, activated during biting behavior, appeared conserved within a species complex, but not beyond. The independent co-evolution between these manipulating parasites and their respective hosts might thus have led to rather diverse strategies to alter behavior. Our data indicate that specialized, secreted enterotoxins may play a major role in one of these strategies.

24 August 2015 in Scientific Reports
Verified
Journal Article
Open access
PDF file
LQLauren E. Quevillon
Lauren E. Quevillon
EHEphraim M. Hanks
Ephraim M. Hanks
SBShweta Bansal
Shweta Bansal
DHDavid Hughes
David Hughes
High-density living is often associated with high disease risk due to density-dependent epidemic spread. Despite being paragons of high-density living, the social insects have largely decoupled the association with density-dependent epidemics. It is hypothesized that this is accomplished through prophylactic and inducible defenses termed ‘collective immunity’. Here we characterise segregation of carpenter ants that would be most likely to encounter infectious agents (i.e. foragers) using integrated social, spatial and temporal analyses. Importantly, we do this in the absence of disease to establish baseline colony organization. Behavioural and social network analyses show that active foragers engage in more trophallaxis interactions than their nest worker and queen counterparts and occupy greater area within the nest. When the temporal ordering of social interactions is taken into account, active foragers and inactive foragers are not observed to interact with the queen in ways that could lead to the meaningful transfer of disease. Furthermore, theoretical resource spread analyses show that such temporal segregation does not appear to impact the colony-wide flow of food. This study provides an understanding of a complex society’s organization in the absence of disease that will serve as a null model for future studies in which disease is explicitly introduced.

1 November 2019 in Current Biology
Verified
Journal Article
Closed access
JAJoão Araújo
João Araújo
DHDavid Hughes
David Hughes