Search

Growing mangoes is an important part of life in southern India and a major economic driver for the area. Nevertheless, several leaf diseases often impede mango tree development and production, substantially affecting harvest output and quality. Detecting and identifying mango leaf diseases early can be challenging due to the diverse crop varieties, climatic circumstances, and numerous disease signs. While deep-learning methods have been developed to address this problem, they generally need help to detect illnesses across geographies and crop types. To tackle this difficulty, this research offers a transfer learning model that uses Explainable Artificial Intelligence (XAI) characteristics to identify and categorize leaf diseases. This research proposes MLTNet (Mango Leaf Disease Classification with Transfer Learning, Feature Localization, and Visual Explanations) in this study. Our study utilized a dataset from Southern India comprising 1,275 high-quality images of mango leaves affected by diseases like rust and powdery mildew, augmented to 11,480 images across 14 classes to enhance model training and robustness. This novel model utilizes Explainable Artificial Intelligence (XAI) techniques such that leaf disease detection and categorization may achieve higher levels of accuracy. The research work lies in the development of the MLTNet model, which integrates Explainable Artificial Intelligence (XAI) techniques with the ResNet50 architecture to enhance classification accuracy in mango leaf disease detection. This model uniquely employs advanced data pre-processing methods like Error Level Analysis and incorporates Grad-CAM for feature localization and visual explanations. We compared MLTNet's performance with state-of-the-art models like ResNet-50, VGG-16, and InceptionV3, focusing on accuracy, interpretability, and computational efficiency. MLTNet demonstrated superior performance, achieving a training accuracy of 94.3% and a test accuracy of 86.3%, which notably surpasses other models under similar conditions. This success is attributed to the model's ability to leverage complex features from the augmented dataset and the added interpretability provided by XAI techniques. [ABSTRACT FROM AUTHOR]

The vast majority of people in rural Andhra Pradesh depend on employment for their primary source of income. Therefore, success in terms of improving employment opportunities can lead to success in reducing poverty and raising living standards. Simply having access to employment is not enough. Labour force participation rates are generally high across the districts in the state of Andhra Pradesh and yet large numbers of employed people cannot lift their families out of poverty. Therefore, reducing poverty requires a joint emphasis on the quality and quantity of employment. The greater the employment focus, the more the effective economic growth in fighting poverty. However, economic growth alone cannot be counted on to generate significant improvement in the quality and quantity of employment opportunities. What is needed is an employment-centred approach to growth and development. The link between employment and living standards is straightforward. However, one has to be clear on what one means by employment in the context of rural Andhra Pradesh. Often labour market policy recommendations assume a wage employment relationship as the dominant framework for understanding the employment dynamics. Equilibrium earnings are determined by an interaction between the supply and demand for labour in a wage labour market. It is critical to understand the structure of employment in rural Andhra Pradesh if appropriate employment policies are to be formulated. [ABSTRACT FROM AUTHOR]

Background & Objectives: Japanese encephalitis (JE) is one of the dreaded mosquito-borne viral diseases mostly prevalent in south Asian countries including India. Early warning of the disease in terms of disease intensity is crucial for taking adequate and appropriate intervention measures. The present study was carried out in Dibrugarh district in the state of Assam located in the northeastern region of India to assess the accuracy of selected forecasting methods based on historical morbidity patterns of JE incidence during the past 22 years (1985-2006)., Methods: Four selected forecasting methods, viz. seasonal average (SA), seasonal adjustment with last three observations (SAT), modified method adjusting long-term and cyclic trend (MSAT), and autoregressive integrated moving average (ARIMA) have been employed to assess the accuracy of each of the forecasting methods. The forecasting methods were validated for five consecutive years from 2007-2012 and accuracy of each method has been assessed., Results: The forecasting method utilising seasonal adjustment with long-term and cyclic trend emerged as best forecasting method among the four selected forecasting methods and outperformed the even statistically more advanced ARIMA method. Peak of the disease incidence could effectively be predicted with all the methods, but there are significant variations in magnitude of forecast errors among the selected methods. As expected, variation in forecasts at primary health centre (PHC) level is wide as compared to that of district level forecasts., Interpretation & Conclusion: The study showed that adopted forecasting techniques could reasonably forecast the intensity of JE cases at PHC level without considering the external variables. The results indicate that the understanding of long-term and cyclic trend of the disease intensity will improve the accuracy of the forecasts, but there is a need for making the forecast models more robust to explain sudden variation in the disease intensity with detail analysis of parasite and host population dynamics.

Purpose: Alteration and overexpression of HER2 proto-oncogene have been implicated in the carcinogenesis and prognosis of ovarian cancer. We evaluated this hypothesis among women with ovarian carcinoma patients from Tiruchirapalli, Tamil Nadu, India., Methods: Genomic DNA was extracted from 72 case patients and 288 control subjects and was examined for I655V polymorphism by PCR-RFLP based assay. Immunohistochemistry analysis was carried out in order to study the overexpression of HER2 protein. The observed number of each genotype was compared with that expected for a population in the Hardy-Weinberg equilibrium. In analysing the relation between genotype and overexpression of HER2 protein, Cochran-Mantel-Haenszel statistics was used., Results: We found that 20.8% of the case patients and 16.3% of the control subjects were heterozygous for the Val allele and 10 case patients (13.8%) and 3 control subjects (1.1%) were homozygous for this allele (P < 0.001). Compared with women with Ile/Ile genotype, women with Val/Val genotype had an elevated risk of ovarian cancer. The genotype distributions were consistent with the Hardy-Weinberg equilibrium. The risk increased with the number of Val allele and women homozygous for the Val allele had 15-fold (OR = 15.3; 95%CI = 4.09-57.31) increased risk of cancer. The patients homozygous for the Valine allele showed strong HER2 protein expression., Conclusion: The results showed that the valine allele may be an indicator of genetic susceptibility to ovarian carcinoma in the study population.

Background, Aim, and Scope: The aim of the present study was to evaluate the distribution of petroleum hydrocarbons in ten commercial fish species and water samples in three estuaries along Tamilnadu coast, Bay of Bengal, India., Materials and Methods: Fish and water samples collected from Tamilnadu coast, India, were extracted and analyzed for petroleum hydrocarbons by ultraviolet fluorescence (UVF) spectroscopy., Results: The petroleum hydrocarbon concentration (PHC) in coastal waters and fish species varied between 2.28 and 14.02 μg/l and 0.52 and 2.05 μg/g, respectively. The highest PHC concentration was obtained in Uppanar estuarine waters (14.02 ± 0.83) and the lowest was observed in Vellar estuarine waters (2.28 ± 0.25)., Discussion: Among the ten fish species, Sardinella longiceps have high PHC concentration from all the locations. This study suggests that S. longiceps can be used as a good biological indicator for petroleum hydrocarbon pollution in water., Conclusions: The concentration of petroleum hydrocarbons in coastal waters along Tamilnadu coast is markedly higher than that in the background, but there is no evidence for its increase in fish of this region. From a public health point, petroleum hydrocarbon residue levels in all fish samples analyzed in this study are considerably lower than the hazardous levels., Recommendations and Perspectives: At present, as Tamilnadu coastal area is in a rapid development stage of new harbour, chemical industries, power plants, oil exploration and other large-scale industries, further assessment of petroleum hydrocarbons and the various hydrodynamic conditions acting in the region are to be studied in detail and continuous pollution monitoring studies should be conducted for improving the aquatic environment. The results will also be useful for pollution monitoring program along the coastal region and also to check the levels of petroleum hydrocarbons.

Eight mollusc species and sediment samples collected from three different stations along Tamilnadu coast, Bay of Bengal, India were analysed for the levels of petroleum hydrocarbons to elucidate the status of the petroleum residues in mollusc meant for human consumption. The concentrations of petroleum hydrocarbons in sediments along Tamilnadu coast varied from 5.04-25.5 microg/g dw (dry weight). High concentration of petroleum hydrocarbons in the sediment of Uppanar estuary (25.5 +/- 1.45 microg/g dw) was perhaps land and marine based anthropogenic sources of this region. The petroleum hydrocarbon residues in eight mollusc species collected from Uppanar, Vellar and Coleroon estuaries varied between 2.44-6.04 microg/g ww (wet weight). Although the concentration of petroleum hydrocarbons in sediment of the Uppanar region was markedly higher than the background, the petroleum hydrocarbon residues in mollusc collected from Uppanar estuary did not suggest bioaccumulation. The results signified that industrial growth has affected the aquatic environments and regular monitoring will help to adopt stringent pollution control measures for better management of the aquatic region.

Background & Objectives: The kala-azar fever (Visceral leishmaniasis) is continuing unabated in India for over a century, now being largely confined to the eastern part of India mainly in Bihar state and to some extent in its bordering states like West Bengal and Uttar Pradesh. Two study sites namely Patepur block in Vaishali district with high endemicity in northern part and Lohardagga block in Lohardagga district with absolute non-endemicity in southern part of Bihar were selected for the study with the following objectives : (i) to study the macro-ecosystem in relation to distribution of vector -Phlebotomus argentipes; (ii) to identify/map the risk prone areas or villages in a block for quick remedial measures; and (iii) to make use of satellite remote sensing and GIS to demonstrate the utility for rapid assessment of landuse/landcover and their relation with the incidence of kalaazar leading to the mapping of risk prone areas., Methods: Indian Remote Sensing (IRS)-1D LISS III satellite data for the periods of March and November 2000 were analysed in Silicon graphic image processing system using ERDAS software. False color composites (FCC) were generated and landuse/landcover was assessed using Maximum likelihood supervised classification techniques based on ground truth training sets. During the study the GIS functions are used to quantify the remotely sensed landscape proportions of 5 km2 buffer surrounding each known group of villages of high occurrence of sandflies in endemic and nonendemic study sites. Instead of traditional ground based survey methods to vector surveillance, the present study used a combination of remote sensing (RS) and geographical information system (GIS) approach to develop landscape predictors of sandfly abundance-an indicator of human vector contact and as a measure of risk prone areas., Results: Statistical analysis using the remotely sensed landscape variables showed that rural villages surrounded by higher proportion of transitional swamps with soft stemmed edible plants and banana, sugarcane plantations had higher sandfly abundance and would, therefore, be at higher risk prone areas for man-vector contact., Interpretation & Conclusion: The present study clearly brought out the usefulness of satellite remote sensing technology in generating the crucial information on spatial distribution of landuse/landcover classes with special emphasis on indicator landcover classes thereby helping in prioritising the area to identify risk prone areas of kala-azar through GIS application tools.