G.B. Pant Institute of Himalayan Environment and Development

Medical image fusion (MIF) significantly improves diagnosis accuracy by combining complementary data from several imaging modalities. However, the conventional fusion methods often fail to maintain the edge details, uniformity of texture and most relevant characteristics. To overcome these challenges, a novel multi-modality medical image fusion using Triple generator network (MIMO-TGAN) is proposed for accurate brain abnormality detection. The proposed MIMO-TGAN uses the multi-modality images such as Magnetic Resonance Imaging (MRI), Computed Tomography (CT), and Positron Emission Tomography (PET) images for MIF process. Initially, the gathered MRI and CT images are denoised with scalable range-based adaptive trilateral (SCRAT) filter and PET images are processed by multi-scale retinex (MSR) technique. The proposed MIMO-TGAN model contains three generators with a discriminator and an abnormality detection phase. The Generator-1 consists of grouped convolutional with different attention blocks for retrieving the relevant features from MRI and CT images to generate a synthesized image. Simultaneously, Generator-2 retrieves the most important voxel features from PET images and MRI images to generate another synthesized image. Finally, Generator-3 is designed to combine the both synthesized images from two generators by using four fusion rules to accurately detect brain abnormalities. The experimental evaluations demonstrate that the proposed MIMO-TGAN achieves an accuracy of 99.43% and 98.24% for dataset-1and dataset-2 respectively. Moreover, the proposed MIMO-TGAN achieves the entropy values of 7.05 and 6.84 for dataset-1 and dataset-2, and overall SSIM of 0.94 that indicates superior fusion quality results compared to existing methods.

Accurate prediction of carbon exchange in the Central Indian Himalaya, a global biodiversity hotspot with extreme vertical gradients and monsoon variability, remains critical for regional carbon assessments. Despite substantial sequestration potential in Himalayan pine forests, mechanistic drivers of net ecosystem exchange (NEE) are poorly constrained, while alpine grassland carbon dynamics remain enigmatic due to observational scarcity at high elevations. Here, we integrate bidirectional long short-term memory (BiLSTM) networks with SHAP explainability to predict hourly NEE and quantitatively rank environmental controls across contrasting ecosystems. Continuous eddy covariance data from a needleleaf forest (Kosi-Katarmal, 1217 m; April-October 2020-2022) and alpine grassland (Darma Valley, 3240 m; July-October 2022-2023) were analyzed using air temperature, relative humidity, net radiation, soil conditions, vapor pressure deficit, NEE derivatives, and multi-scale lag features (1-24 h). The BiLSTM model achieved exceptional performance (forest: R² = 0.94-0.95, RMSE = 2.18-2.86 μmol m⁻² s⁻¹; grassland: R² = 0.95-0.96, RMSE = 1.09-1.73 μmol m⁻² s⁻¹). Critically, SHAP attribution unveiled fundamentally divergent control mechanisms: forest NEE was governed by rapid temporal dynamics (NEE derivative, SHAP: 0.70) and radiation-temperature coupling (SHAP: 0.02 each), signifying energy-driven photosynthetic control. Conversely, grassland NEE exhibited strong short-term memory (1-h lag, SHAP: 0.35) and atmospheric constraint dominance (temperature, SHAP: 0.06, humidity: 0.03), reflecting stomatal regulation and evaporative demand at high elevation. These findings demonstrate that forest carbon exchange operates as an energy-limited, dynamically responsive system, whereas grasslands function as atmospheric-demand limited systems with pronounced temporal persistence. Our results provide a mechanistic framework for ecosystem-specific carbon flux modeling and demonstrate the efficacy of explainable AI for process understanding in data-sparse mountain regions.

Meteorological drought is a major natural hazard in semi-arid regions, where high climate variability and strong dependence on precipitation intensify pressure on water resources and socio-economic systems. This study examined the spatiotemporal characteristics of meteorological drought in the Wadi Sly basin (northwestern Algeria) over the period 1967–2022, using long-term monthly precipitation records from seven meteorological stations. The Standardized Precipitation Index (SPI) was calculated at multiple time scales (1-, 3-, 6-, 9-, and 12-month) to characterize drought onset, severity, persistence, and temporal variability. In addition, drought severity probability and frequency analyses were conducted to evaluate the likelihood and recurrence of different drought classes. The results indicate pronounced inter-annual and decadal variability in drought conditions, with severe and prolonged drought episodes occurring during the mid-1980s, early-to-mid-1990s, and late 2010s. During these periods, SPI values frequently fell below −2.0, signifying extreme drought conditions. Spatial analysis reveals strong basin-wide synchronicity of drought events, suggesting the influence of large-scale atmospheric drivers, although localized variations in drought intensity remain evident. Overall, near-normal conditions dominate the record (accounting for approximately 60–70% of observations), while moderately dry conditions occur more frequently than moderately wet conditions at several stations. Drought characteristics exhibit strong scale dependence, with short-term droughts prevailing at shorter SPI time scales, while longer time scales emphasize drought persistence and accumulation. Overall, the findings indicate an increasing prominence of long-duration drought conditions in recent decades, as evidenced by recurrent low SPI values at longer aggregation scales. Such conditions may pose heightened risks to groundwater recharge processes and long-term water resource availability. Despite the limitations inherent in precipitation-based indices, this study provides a robust statistical framework for drought characterization and contributes valuable insights for improved drought monitoring, early warning systems, and climate-resilient water resource management in semi-arid basins.

Purpose: This study aims to explore the transformative role of artificial intelligence (AI) and machine learning (ML) in modern agriculture, emphasising their potential to revolutionise traditional farming practices, enhance resource efficiency, and address global food security challenges. Design/methodology/approach: Through an extensive review of recent technological advancements and practical applications, this chapter examines the integration of AI and ML with human-centric agricultural practices. The review focuses on various domains such as yield prediction, pest and disease detection, crop monitoring, precision farming, and livestock management. Findings: The study identifies AI and ML as critical enablers of smart agriculture, facilitating real-time data analysis through advanced ML algorithms, computer vision, and sensor-based systems. These technologies provide actionable insights for farmers, improving decision-making processes and overall farm productivity. Furthermore, the paper underscores the importance of these tools in promoting sustainable and resilient farming systems. Practical implications: By showcasing how AI and ML can enhance agricultural efficiency and sustainability, this study offers valuable guidance for policymakers, agritech developers, and farming communities. It highlights the need for strategic implementation to maximise benefits while ensuring ethical and equitable technology access. Originality/value: This chapter contributes a timely and comprehensive perspective on AI-driven innovation in agriculture. It positions AI and ML not just as technological tools but as transformative forces with the potential to reshape global agricultural practices for a more food-secure and environmentally responsible future.Keywords:

This article reveals the creation and execution of a digital twin (DT) for the robotic inspection system of mechanical parts, applying the machine vision method. The examination is carried out on the KUKA iiwa robot that has a camera for the purpose of identifying and sorting the parts that are positioned on the work area. A vision system based on a YOLO convolutional neural network (CNN) was utilized for the tasks of object detection and classification. The dataset was automatically created via the use of 3D CAD models along with domain randomization to mitigate the risk of overfitting. The simulation environment was set up in IsaacSim, and the vision system was interfaced through ROS Noetic with hand-eye calibration also included. The performance of the YOLO network evidenced an overall precision of 94.8% and recall of 87.7% in the test dataset indicating the parts' effective detection and classification. The research underlines the convergence of simulation, deep learning, robotics, and the potential for automated inspections in manufacturing sectors.