scholarly journals Comparison of Convolutional Neural Network Architectures for Classification of Tomato Plant Diseases

2020 ◽  
Vol 10 (4) ◽  
pp. 1245 ◽  
Author(s):  
Valeria Maeda-Gutiérrez ◽  
Carlos E. Galván-Tejada ◽  
Laura A. Zanella-Calzada ◽  
José M. Celaya-Padilla ◽  
Jorge I. Galván-Tejada ◽  
...  
Keyword(s):  
Operating Characteristic ◽  
Area Under The Curve ◽  
Plant Diseases ◽  
Fine Tuning ◽  
Network Architectures ◽  
Tomato Plants ◽  
Tomato Diseases ◽  
Sensitivity Specificity

Tomato plants are highly affected by diverse diseases. A timely and accurate diagnosis plays an important role to prevent the quality of crops. Recently, deep learning (DL), specifically convolutional neural networks (CNNs), have achieved extraordinary results in many applications, including the classification of plant diseases. This work focused on fine-tuning based on the comparison of the state-of-the-art architectures: AlexNet, GoogleNet, Inception V3, Residual Network (ResNet) 18, and ResNet 50. An evaluation of the comparison was finally performed. The dataset used for the experiments is contained by nine different classes of tomato diseases and a healthy class from PlantVillage. The models were evaluated through a multiclass statistical analysis based on accuracy, precision, sensitivity, specificity, F-Score, area under the curve (AUC), and receiving operating characteristic (ROC) curve. The results present significant values obtained by the GoogleNet technique, with 99.72% of AUC and 99.12% of sensitivity. It is possible to conclude that this significantly success rate makes the GoogleNet model a useful tool for farmers in helping to identify and protect tomatoes from the diseases mentioned.

scholarly journals Evaluation of scratch and pre-trained convolutional neural networks for the classification of Tomato plant diseases

2021 ◽  
Vol 10 (2) ◽  
pp. 467
Author(s):  
Mohammad Amimul Ihsan Aquil ◽  
Wan Hussain Wan Ishak
Keyword(s):  
Neural Networks ◽  
Convolutional Neural Networks ◽  
Tomato Plant ◽  
State Of The Art ◽  
Plant Diseases ◽  
Fine Tuning ◽  
Convolutional Network ◽  
Tomato Diseases

<span id="docs-internal-guid-01580d49-7fff-6f2a-70d1-7893ec0a6e14"><span>Plant diseases are a major cause of destruction and death of most plants and especially trees. However, with the help of early detection, this issue can be solved and treated appropriately. A timely and accurate diagnosis is critical in maintaining the quality of crops. Recent innovations in the field of deep learning (DL), especially in convolutional neural networks (CNNs) have achieved great breakthroughs across different applications such as the classification of plant diseases. This study aims to evaluate scratch and pre-trained CNNs in the classification of tomato plant diseases by comparing some of the state-of-the-art architectures including densely connected convolutional network (Densenet) 120, residual network (ResNet) 101, ResNet 50, ReseNet 30, ResNet 18, squeezenet and Vgg.net. The comparison was then evaluated using a multiclass statistical analysis based on the F-Score, specificity, sensitivity, precision, and accuracy. The dataset used for the experiments was drawn from 9 classes of tomato diseases and a healthy class from PlantVillage. The findings show that the pretrained Densenet-120 performed excellently with 99.68% precision, 99.84% F-1 score, and 99.81% accuracy, which is higher compared to its non-trained based model showing the effectiveness of using a combination of a CNN model with fine-tuning adjustment in classifying crop diseases.</span></span>

scholarly journals Evaluation of alpha-l-fucosidase for the diagnosis of hepatocellular carcinoma based on meta-analysis

2020 ◽  
Vol 0 (0) ◽  
Author(s):  
Lei Xi ◽  
Chunqing Yang
Keyword(s):  
Hepatocellular Carcinoma ◽  
Operating Characteristic ◽  
Meta Analysis ◽  
Area Under The Curve ◽  
Diagnostic Odds Ratio ◽  
Diagnostic Value ◽  
Receiver Operating Characteristic Curves ◽  
Sensitivity Specificity ◽  
Review Manager

AbstractObjectivesThe main aim of the present study was to assess the diagnostic value of alpha-l-fucosidase (AFU) for hepatocellular carcinoma (HCC).MethodsStudies that explored the diagnostic value of AFU in HCC were searched in EMBASE, SCI, and PUBMED. The sensitivity, specificity, and DOR about the accuracy of serum AFU in the diagnosis of HCC were pooled. The methodological quality of each article was evaluated with QUADAS-2 (quality assessment for studies of diagnostic accuracy 2). Receiver operating characteristic curves (ROC) analysis was performed. Statistical analysis was conducted by using Review Manager 5 and Open Meta-analyst.ResultsEighteen studies were selected in this study. The pooled estimates for AFU vs. α-fetoprotein (AFP) in the diagnosis of HCC in 18 studies were as follows: sensitivity of 0.7352 (0.6827, 0.7818) vs. 0.7501 (0.6725, 0.8144), and specificity of 0.7681 (0.6946, 0.8283) vs. 0.8208 (0.7586, 0.8697), diagnostic odds ratio (DOR) of 7.974(5.302, 11.993) vs. 13.401 (8.359, 21.483), area under the curve (AUC) of 0.7968 vs. 0.8451, respectively.ConclusionsAFU is comparable to AFP for the diagnosis of HCC.

scholarly journals Early Detection and Classification of Tomato Leaf Disease Using High-Performance Deep Neural Network

Sensors ◽  
2021 ◽  
Vol 21 (23) ◽  
pp. 7987
Author(s):  
Naresh K. Trivedi ◽  
Vinay Gautam ◽  
Abhineet Anand ◽  
Hani Moaiteq Aljahdali ◽  
Santos Gracia Villar ◽  
...  
Keyword(s):  
Neural Network ◽  
Crop Yield ◽  
High Performance ◽  
Early Stage ◽  
Tomato Plants ◽  
Leaf Disease ◽  
Proposed Model ◽  
Tomato Diseases

Tomato is one of the most essential and consumable crops in the world. Tomatoes differ in quantity depending on how they are fertilized. Leaf disease is the primary factor impacting the amount and quality of crop yield. As a result, it is critical to diagnose and classify these disorders appropriately. Different kinds of diseases influence the production of tomatoes. Earlier identification of these diseases would reduce the disease’s effect on tomato plants and enhance good crop yield. Different innovative ways of identifying and classifying certain diseases have been used extensively. The motive of work is to support farmers in identifying early-stage diseases accurately and informing them about these diseases. The Convolutional Neural Network (CNN) is used to effectively define and classify tomato diseases. Google Colab is used to conduct the complete experiment with a dataset containing 3000 images of tomato leaves affected by nine different diseases and a healthy leaf. The complete process is described: Firstly, the input images are preprocessed, and the targeted area of images are segmented from the original images. Secondly, the images are further processed with varying hyper-parameters of the CNN model. Finally, CNN extracts other characteristics from pictures like colors, texture, and edges, etc. The findings demonstrate that the proposed model predictions are 98.49% accurate.

scholarly journals Usefulness of sialic acid for diagnosis of sepsis in critically ill patients: a retrospective study

2020 ◽  
Vol 20 (1) ◽  
Author(s):  
Bo Yao ◽  
Wen-juan Liu ◽  
Di Liu ◽  
Jin-yan Xing ◽  
Li-juan Zhang
Keyword(s):  
Sialic Acid ◽  
Critically Ill ◽  
Critically Ill Patients ◽  
Incomplete Data ◽  
Acid Level ◽  
Operating Characteristic ◽  
Area Under The Curve ◽  
Diagnostic Efficacy ◽  
Receiver Operating Characteristic Curves ◽  
Sensitivity Specificity

Abstract Background Early diagnosis of sepsis is very important. It is necessary to find effective and adequate biomarkers in order to diagnose sepsis. In this study, we compared the value of sialic acid and procalcitonin for diagnosing sepsis. Methods Newly admitted intensive care unit patients were enrolled from January 2019 to June 2019. We retrospectively collected patient data, including presence of sepsis or not, procalcitonin level and sialic acid level. Receiver operating characteristic curves for the ability of sialic acid, procalcitonin and combination of sialic acid and procalcitonin to diagnose sepsis were carried out. Results A total of 644 patients were admitted to our department from January 2019 to June 2019. The incomplete data were found in 147 patients. Finally, 497 patients data were analyzed. The sensitivity, specificity and area under the curve for the diagnosis of sepsis with sialic acid, procalcitonin and combination of sialic acid and procalcitonin were 64.2, 78.3%, 0.763; 67.9, 84.0%, 0.816 and 75.2, 84.6%, 0.854. Moreover, sialic acid had good values for diagnosing septic patients with viral infection, with 87.5% sensitivity, 82.2% specificity, and 0.882 the area under the curve. Conclusions Compared to procalcitonin, sialic acid had a lower diagnostic efficacy for diagnosing sepsis in critically ill patients. However, the combination of sialic acid and procalcitonin had a higher diagnostic efficacy for sepsis. Moreover, sialic acid had good value for diagnosing virus-induced sepsis.

Tomato pest classification using deep convolutional neural network with transfer learning, fine tuning and scratch learning

2021 ◽  
pp. 1-10
Author(s):  
Gayatri Pattnaik ◽  
Vimal K. Shrivastava ◽  
K. Parvathi
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Transfer Learning ◽  
Data Augmentation ◽  
State Of The Art ◽  
Deep Convolutional Neural Network ◽  
Fine Tuning ◽  
Tomato Plants ◽  
Random Weights

Pests are major threat to economic growth of a country. Application of pesticide is the easiest way to control the pest infection. However, excessive utilization of pesticide is hazardous to environment. The recent advances in deep learning have paved the way for early detection and improved classification of pest in tomato plants which will benefit the farmers. This paper presents a comprehensive analysis of 11 state-of-the-art deep convolutional neural network (CNN) models with three configurations: transfers learning, fine-tuning and scratch learning. The training in transfer learning and fine tuning initiates from pre-trained weights whereas random weights are used in case of scratch learning. In addition, the concept of data augmentation has been explored to improve the performance. Our dataset consists of 859 tomato pest images from 10 categories. The results demonstrate that the highest classification accuracy of 94.87% has been achieved in the transfer learning approach by DenseNet201 model with data augmentation.

scholarly journals Sampling for IPM Decision Making: Where Should We Invest Time and Resources?

1999 ◽  
Vol 89 (11) ◽  
pp. 1104-1111 ◽  
Author(s):  
Jan P. Nyrop ◽  
Michael R. Binns ◽  
Wopke van der Werf
Keyword(s):  
Decision Making ◽  
Sample Size ◽  
Operating Characteristic ◽  
Crop Protection ◽  
Critical Density ◽  
Sampling Distribution ◽  
Fine Tuning ◽  
Sampling Distributions ◽  
Sample Plan

Guides for making crop protection decisions based on assessments of pest abundance or incidence are cornerstones of many integrated pest management systems. Much research has been devoted to developing sample plans for use in these guides. The development of sampling plans has usually focused on collecting information on the sampling distribution of the pest, describing this sampling distribution with a mathematical model, formulating a sample plan, and sometimes, but not always, evaluating the performance of the proposed sample plan. For crop protection decision making, classification of density or incidence is usually more appropriate than estimation. When classification is done, the average outcome of classification (the operating characteristic) is frequently robust to large changes in the sampling distribution, including estimates of the variance of pest counts, and to sample size. In contrast, the critical density, or critical incidence, about which classifications are made, has a large influence on the operating characteristic. We suggest that rather than investing resources in elaborate descriptions of sampling distributions, or in fine-tuning sample size to achieve desired levels of precision, greater emphasis should be placed on characterizing pest densities that signal the need for management action and on designing decision guides that will be adopted by practitioners.

Video-Polysomnographic Assessment for the Diagnosis of Disorders of Arousal in Children

Neurology ◽  
2020 ◽  
Vol 96 (1) ◽  
pp. e121-e130
Author(s):  
Régis Lopez ◽  
Christine Laganière ◽  
Sofiène Chenini ◽  
Anna Laura Rassu ◽  
Elisa Evangelista ◽  
...  
Keyword(s):  
Operating Characteristic ◽  
Slow Wave Sleep ◽  
Area Under The Curve ◽  
Class Iii ◽  
Receiver Operating Characteristic Curves ◽  
Arousal Index ◽  
Fragmentation Index ◽  
Increased Sensitivity ◽  
Disorders Of Arousal

ObjectivesTo highlight the slow-wave sleep (SWS) fragmentation and validate the video-polysomnographic (vPSG) criteria and cutoffs for the diagnosis of disorders of arousal (DOA) in children, as already reported in adults.MethodsOne hundred children (66 boys, 11.0 ± 3.3 years) with frequent episodes of DOA and 50 nonparasomniac children (32 boys, 10.9 ± 3.9 years) underwent vPSG recording to quantify SWS characteristics (number of N3 sleep interruptions, fragmentation index, slow/mixed and fast arousal ratios, and indexes per hour) and associated behaviors. We compared SWS characteristics in the 2 groups and defined the optimal cutoff values for the diagnosis of DOA using receiver operating characteristic curves.ResultsPatients with DOA had higher amounts of N3 and REM sleep, number of N3 interruptions, SWS fragmentation, and slow/mixed arousal indexes than controls. The highest area under the curve (AUC) values were obtained for SWS fragmentation and slow/mixed arousal indexes with satisfactory classification performances (AUC 0.80, 95% confidence interval [CI] 0.73–0.87; AUC 0.82, 95% CI 0.75–0.89). SWS fragmentation index cutoff value of 4.1/h reached a sensitivity of 65.0% and a specificity of 84.0%. Slow/mixed arousal index cutoff of 3.8/h reached a sensitivity of 69.0% and a specificity of 82.0%. At least one parasomniac episode was recorded in 63.0% of patients and none of the controls. Combining behavioral component by vPSG increased sensitivity of both biomarkers to 83% and 89%, respectively.ConclusionsWe confirmed that SWS fragmentation and slow/mixed arousal indexes are 2 relevant biomarkers for the diagnosis of DOA in children, with different cutoffs obtained than those validated in adults.Classification of EvidenceThis study provides Class III evidence that SWS fragmentation and slow/mixed arousal indexes on vPSG accurately identify children with DOA.

scholarly journals Diagnostic accuracy of Doppler ultrasound, CT and MRI in Budd Chiari syndrome: systematic review and meta-analysis

2020 ◽  
Vol 93 (1109) ◽  
pp. 20190847 ◽  
Author(s):  
Pankaj Gupta ◽  
Varun Bansal ◽  
Praveen Kumar-M ◽  
Saroj K Sinha ◽  
Jayanta Samanta ◽  
...  
Keyword(s):  
Diagnostic Accuracy ◽  
Doppler Ultrasound ◽  
Meta Analysis ◽  
Area Under The Curve ◽  
Diagnostic Odds Ratio ◽  
Budd Chiari Syndrome ◽  
Chiari Syndrome ◽  
Sensitivity Specificity ◽  
Ct And Mri

Objective: To evaluate the sensitivity, specificity, and diagnostic odds ratio (DOR) of Doppler ultrasound, CT, and MRI in the diagnosis of Budd Chiari syndrome (BCS). Methods: We performed a literature search in PubMed, Embase, and Scopus to identify articles reporting the diagnostic accuracy of Doppler ultrasound, CT, and MRI (either alone or in combination) for BCS using catheter venography or surgery as the reference standard. The quality of the included articles was assessed by using the Quality Assessment of Diagnostic Accuracy Studies-2 tool. Results: 11 studies were found eligible for inclusion. Pooled sensitivities and specificities of Doppler ultrasound were 89% [95% confidence interval (CI), 81–94%, I2 = 24.7%] and 68% (95% CI, 3–99%, I2 = 95.2%), respectively. Regarding CT, the pooled sensitivities and specificities were 89% (95% CI, 77–95%, I2 = 78.6%) and 72% (95% CI, 21–96%, I2 = 91.4%), respectively. The pooled sensitivities and specificities of MRI were 93% (95% CI, 89–96%, I2 = 10.6%) and 55% (95% CI, 5–96%, I2 = 87.6%), respectively. The pooled DOR for Doppler ultrasound, CT, and MRI were 10.19 (95% CI: 1.5, 69.2), 14.57 (95% CI: 1.13, 187.37), and 20.42 (95% CI: 1.78, 234.65), respectively. The higher DOR of MRI than that of Doppler ultrasound and CT shows the better discriminatory power. The area under the curve for MRI was 90.8% compared with 88.4% for CT and 86.6% for Doppler ultrasound. Conclusion: Doppler ultrasound, CT and MRI had high overall diagnostic accuracy for diagnosis of BCS, but substantial heterogeneity was found. Prospective studies are needed to investigate diagnostic performance of these imaging modalities. Advances in knowledge: MRI and CT have the highest meta-analytic sensitivity and specificity, respectively for the diagnosis of BCS. Also, MRI has the highest area under curve for the diagnosis of BCS.

External Validation of Different Scoring Systems for Suspected Choledocholithiasis

2019 ◽  
Vol 36 (6) ◽  
pp. 530-538
Author(s):  
Nicolò Tamini ◽  
Davide Paolo Bernasconi ◽  
Luca Gianotti
Keyword(s):  
Operating Characteristic ◽  
Bile Duct Stone ◽  
Characteristic Curve ◽  
External Validation ◽  
Area Under The Curve ◽  
Scoring Systems ◽  
Youden Index ◽  
Predictive Values ◽  
Sensitivity Specificity ◽  
Symptoms And Signs

Aim of the Study: The diagnosis of choledocholithiasis is challenging. Previously published scoring systems designed to calculate the risk of choledocholithiasis were evaluated to appraise the diagnostic performance. Patients and Methods: Data of patients who were admitted between 2013 and 2015 with the following characteristics were retrieved: bile stone-related symptoms and signs, and indication to laparoscopic cholecystectomy. To validate and appraise the performance of the 6 scoring systems, the acknowledged domains of each metrics were applied to the present cohort. Sensitivity, specificity, positive, negative predictive, Youden index, and receiver operating characteristic curve with the area under the curve (AUC) values of the scores were calculated. Results: Two-hundred patients were analyzed. The highest sensitivity and specificity were obtained from the Menezes’ (96.6%) and Telem’s (99.3%) metrics respectively. The Telem’s and Menezes’ scores had the best positive (75.0%) and negative (96.4%) predictive values respectively. The best accuracy, as computed by the Youden index and AUC, was found for the Soltan’s scoring system (0.628 and 0.88, respectively). Conclusion: The available scoring systems are precise only in identifying patients with a negligible risk of common bile duct stone, but overall insufficiently accurate to suggest the routine use in clinical practice.

scholarly journals Glycaemic index methodology

2005 ◽  
Vol 18 (1) ◽  
pp. 145-171 ◽  
Author(s):  
F. Brouns ◽  
I. Bjorck ◽  
K. N. Frayn ◽  
A. L. Gibbs ◽  
V. Lang ◽  
...  
Keyword(s):  
Energy Content ◽  
Area Under The Curve ◽  
Test Dose ◽  
Glycaemic Index ◽  
Food Groups ◽  
Methodological Considerations ◽  
Sampling Procedures ◽  
Different Sources

AbstractThe glycaemic index (GI) concept was originally introduced to classify different sources of carbohydrate (CHO)-rich foods, usually having an energy content of >80 % from CHO, to their effect on post-meal glycaemia. It was assumed to apply to foods that primarily deliver available CHO, causing hyperglycaemia. Low-GI foods were classified as being digested and absorbed slowly and high-GI foods as being rapidly digested and absorbed, resulting in different glycaemic responses. Low-GI foods were found to induce benefits on certain risk factors for CVD and diabetes. Accordingly it has been proposed that GI classification of foods and drinks could be useful to help consumers make ‘healthy food choices’ within specific food groups. Classification of foods according to their impact on blood glucose responses requires a standardised way of measuring such responses. The present review discusses the most relevant methodological considerations and highlights specific recommendations regarding number of subjects, sex, subject status, inclusion and exclusion criteria, pre-test conditions, CHO test dose, blood sampling procedures, sampling times, test randomisation and calculation of glycaemic response area under the curve. All together, these technical recommendations will help to implement or reinforce measurement of GI in laboratories and help to ensure quality of results. Since there is current international interest in alternative ways of expressing glycaemic responses to foods, some of these methods are discussed.

Sign in / Sign up

Export Citation Format

Share Document