Deep recognition of rice disease images: how many training samples do we really need?

Background: With the rapid development of deep learning, the recognition of rice disease images using deep neural networks has become a hot research topic. However, most previous studies only focus on the modification of deep learning models, while lacking research to systematically and scientifically explore the impact of different data sizes on the image recognition task for rice diseases. In this study, a functional model was developed to predict the relationship between the size of dataset and the accuracy rate of model recognition.

Effects of Protection Time on Infection of Rice Panicle Blast.

Panicle blast, caused by , is a destructive disease of rice worldwide. Clarifying the susceptibility of rice panicles at different stages is of great significance for effective disease management. Field experiments were conducted in two paddy fields at Wuyuan County in 2016 and 2017 to determine the effects of head covering and its timing on the infection of rice panicle blast.

Comparison of the efficacy and safety of total laparoscopic hysterectomy without and with uterine manipulator combined with pelvic lymphadenectomy for early cervical cancer.

Objective: Some studies have reported that the prognosis of total laparoscopic hysterectomy (TLH) for early-stage cervical cancer (CC) is worse than that of open surgery. And this was associated with the use of uterine manipulator or not. Therefore, this study retrospectively analyzes the efficacy and safety of TLH without uterine manipulator combined with pelvic lymphadenectomy for early-stage CC.

The development of a cSMART-based integrated model for hepatocellular carcinoma diagnosis.

Background: Hepatocellular carcinoma (HCC) generally arises from a background of liver cirrhosis (LC). Patients with cirrhosis and suspected HCC are recommended to undergo serum biomarker tests and imaging diagnostic evaluation. However, the performance of routine diagnostic methods in detecting early HCC remains unpromising.

Effects of Image Dataset Configuration on the Accuracy of Rice Disease Recognition Based on Convolution Neural Network.

In recent years, the convolution neural network has been the most widely used deep learning algorithm in the field of plant disease diagnosis and has performed well in classification. However, in practice, there are still some specific issues that have not been paid adequate attention to. For instance, the same pathogen may cause similar or different symptoms when infecting plant leaves, while the same pathogen may cause similar or disparate symptoms on different parts of the plant.

Temporal Progress and Spatial Patterns of Northern Corn Leaf Blight in Corn Fields in China.

Northern corn leaf blight (NCLB), caused by , is a devastating disease of corn in China. To enhance our understanding of NCLB epidemiology, the temporal progress and spatial patterns of NCLB were investigated. A susceptible corn cultivar, Xianyu 335, was planted in a field in Beijing in 2016 and 2017.

Evaluation of Rice Responses to the Blast Fungus Magnaporthe oryzae at Different Growth Stages.

Rice blast, caused by the fungus Magnaporthe oryzae, is the most damaging disease for rice worldwide. However, the reactions of rice to M. oryzae at different growth stages are largely unknown.

Spatial Analysis of Rice Blast in China at Three Different Scales.

In this study, spatial analyses were conducted at three different scales to better understand the epidemiology of rice blast, a major rice disease caused by Magnaporthe oryzae. At the regional scale, across the major rice production regions in China, rice blast incidence was monitored on 101 dates at 193 stations from 10 June to 10 September during 2009 to 2014, and surveyed in 143 fields in September 2016; at the county scale, three surveys were done covering one to five counties in 2015 to 2016; and, at the field scale, blast was evaluated in six fields in 2015 to 2016. Spatial cluster and hot spot analyses were conducted in the geographic information system on the geographical pattern of the disease at regional scale, and geostatistical analysis was performed at all three scales.

Dynamics of verticillium species microsclerotia in field soils in response to fumigation, cropping patterns, and flooding.

Verticillium dahliae is a soilborne, economically significant fungal plant pathogen that persists in the soil for up to 14 years as melanized microsclerotia (ms). Similarly, V. longisporum is a very significant production constraint on members of the family Brassicaceae.

The Sclerotinia sclerotiorum mating type locus (MAT) contains a 3.6-kb region that is inverted in every meiotic generation.

Sclerotinia sclerotiorum is a fungal plant pathogen and the causal agent of lettuce drop, an economically important disease of California lettuce. The structure of the S. sclerotiorum mating type locus MAT has previously been reported and consists of two idiomorphs that are fused end-to-end as in other homothallics.

Phylogenetic Analyses of Phytopathogenic Isolates of Verticillium spp.

ABSTRACT To better understand the genetic relationships between Verticillium dahliae isolates from lettuce and other phytopathogenic Verticillium spp. isolates from various hosts and geographic locations, the complete intergenic spacer (IGS) region of the nuclear ribosomal RNA gene (rDNA) and the beta-tubulin gene were amplified and sequenced. The sequences of the complete IGS region and the beta-tubulin gene were used alone and in combination to infer genetic relationships among different isolates of Verticillium with the maximum-likelihood distance method.

Stem and Crown Rot of Chickpea in California Caused by Sclerotinia trifoliorum.

The identities of Sclerotinia isolates obtained from chickpea plants showing stem and crown rot were determined using morphological characteristics, variations in group I introns, and internal transcribed spacer (ITS) sequences. Isolates could be separated into two groups based on growth rates at 22°C, fast growing (about 40 mm per day) versus slow growing (about 20 mm per day). All fast-growing isolates induced stronger color change of a pH-indicating medium than did slow-growing isolates at 22°C.