Biology subjects

Parisey, N.

Publications and source records attributed to Parisey, N..

6 recordsLinked to original sources

bioRxiv2026

Text guidance is powerful but prompt-sensitive for weakly-supervised leaf symptom segmentation

Accurate segmentation of plant disease symptoms is essential for crop monitoring and phenotyping, yet it typically requires costly pixel-level annotations. Weakly supervised semantic segmentation (WSSS) alleviates this burden using image-level labels, but its performance depends on the quality of spatial priors such as class activation maps (CAMs). We investigate whether text-guided segmentation with the Segment Anything Model 3 (SAM3) can serve as an alternative weak supervision signal. Three pseudo-mask generation strategies are compared: (i) CAMs refined with SAM or SAM3, (ii) zero-shot text-guided SAM3, and (iii) a hybrid approach combining weak spatial cues with text prompts. The resulting pseudo-masks are used to train a DeepLabV3 model. Text guidance alone matches or outperforms conventional WSSS, achieving up to 0.46 IoU without spatial supervision and 0.61 IoU on a public dataset, although performance is sensitive to text prompt formulation. The hybrid strategy improves robustness, reaching 0.50 IoU on the primary dataset and 0.58 IoU on the additional dataset while reducing prompt sensitivity. Overall, text guidance is a promising alternative to conventional weak supervision, while hybrid approaches provide a more robust solution for plant disease segmentation.

plant biology↗

bioRxiv2022

Identifying fungal leaf spots on oilseed rape: could pictures help?

From sowing in late summer until harvest in following summer, oilseed rape can be infected by several fungi, which foliar symptoms (leaf spots) coexist on the crop. Training an expert at their identification is quick for the typical symptoms with characteristic appearance. However, in many cases the size, colour and morphology are similar and for the atypical symptoms, there is a risk of confusion or in-decidability. Also, scouting the fields for expert training is not possible at all seasons and all diseases might not be seen in all years and all places. The aim of our study was to produce large sets of pictures annotated by several experts, from which tables illustrating the diversity of symptom appearance were chosen. These tables will enable assistance to diagnostic and expert training.

ecology↗

bioRxiv2022

Imaging with spatio-temporal modelling to characterize the dynamics of plant-pathogen lesions

Within-host spread of pathogens is an important process for the study of plant-pathogen interactions. However, the development of plant-pathogen lesions remains practically difficult to characterize beyond the common traits such as lesion area. Here, we address this question by combining image-based phenotyping with mathematical modelling. We consider the spread of Peyronellaea pinodes on pea stipules that were monitored daily with visible imaging. We assume that pathogen propagation on host-tissues can be described by the Fisher-KPP model where lesion spread depends on both a logistic growth and an homogeneous diffusion. Model parameters are estimated using a variational data assimilation approach on sets of registered images. This modelling framework is used to compare the spread of an aggressive isolate on two pea cultivars with contrasted levels of partial resistance. We show that the expected slower spread on the most resistant cultivar is actually due to a significantly lower diffusion coefficient. This study shows that combining imaging with spatial mechanistic models can offer a mean to disentangle some processes involved in host-pathogen interactions and further development may allow a better identification of quantitative traits thereafter used in genetics and ecological studies.

ecology↗

bioRxiv2021

Optimal spatial monitoring of populations described by reaction-diffusion models

Using spatialised population measurements and related geographic habitat data, it is quite feasible nowadays to derive parsimonious spatially explicit population models and to carry on their parameter estimation. To achieve such goal, reaction-diffusion models are fairly common in conservation biology and agricultural plant health where they are used, for example, for landscape planning or epidemiological surveillance. Unfortunately, if the mathematical methods and computational power are readily available, biological measurements are not. Despite the high throughput of some habitat related remote sensors, the experimental cost of biological measurements are, in our view, one of the worst bottleneck against a widespread usage of reaction-diffusion models. Hence, in this paper, we will recall some classical methods for optimal experimental design that we deem useful to spatial ecologist. Using two case studies, one in landscape ecology and one in conservation biology, we will show how to construct a priori experimental design minimizing variance of parameter estimates, enabling optimal experimental setup with pre and post hoc filtering for accommodating additional constraints.

ecology↗

bioRxiv2020

Besides stem canker severity, oilseed rape host genotype matters for the production of Leptosphaeria maculans fruiting bodies

For fungal cyclic epidemics on annual crops, the pathogen carry-over is an important step in designing disease control strategies. However, it remains particularly difficult to estimate and predict. Plant resistance affects the pathogen development within the epidemics but we lack data on the inter-annual transmission of inoculum. We addressed this question by considering Leptosphaeria maculans on 15 oilseed rape genotypes in field during 4 growing seasons. Stem canker severity of host genotypes was visually scored at harvest while the number of fruiting bodies produced on incubated stubble was quantified using an automated image analysis framework. Our results confirm that higher severity at harvest leads to higher fruiting body production and is significantly affected by host genotype and Nitrogen supply. Most interestingly, we show that the production of fruiting bodies is significantly and substantially affected by host genotype, independently of severity at harvest. Tracking individual stems through incubation, we confirm for the first time that the oilseed rape genotype has a direct effect, not only through disease severity. While the genericity of this finding should be investigated on other fungi, this major effect of genotype on inoculum carry-over should be taken into account in models of varietal deployment strategies.

microbiology↗

bioRxiv2018

Automated image processing to support the analysis of between-year transmission of Leptosphaeria maculans in field conditions

Understanding the transmission of inoculum between periods where the host plants are present is central for predicting the development of plant diseases and optimising mitigation strategies. However, the production at the end of the growing period, the survival during the intercrop period, and the emergence or emission of inoculum after sowing or planting can be highly variable, difficult to assess and generally inferred indirectly from symptoms data. As a result, there is a lack of large data sets which is a major brake for the study of these epidemiological processes. Here we focus on Leptosphaeria maculans that causes the black leg of oilseed rape. After having infected leaves, at early stages of the plant, and migrating into the stem, it causes a basal stem canker before harvest. It then survives on stubble left in the field from which ascospores are emitted at the beginning of the next growing period. In this study we first developed an image processing framework to estimate the density of fruiting bodies produced on stubble. Then, we used this framework to analyse automatically a large number of stems collected in oilseed rape fields among a cultivated area. Having performed a quality assessment of the processing chain we used the output data to investigate how the potential level of inoculum may change with the source field, the considered year and the stem canker severity at harvest. Besides the insights gain into the blackleg of oilseed rape, this work shows how image-based phenotyping may support epidemiological studies by increasing substantially the precision of high throughput disease data.

epidemiology↗