Farhan Amin

Background: Lung cancer remains a major cause of cancer-related mortality, and accurate analysis of pulmonary abnormalities on computed tomography (CT) images is important for supporting timely diagnosis. However, automated lung cancer classification remains challenging because pulmonary lesions exhibit substantial variability in size, morphology, intensity, and anatomical location, while image noise and irrelevant surrounding structures can adversely affect model performance. This study investigates an integrated deep learning framework combining image restoration, anatomical segmentation, and transformer-based classification for computer-aided lung cancer analysis. Methods: The proposed framework consists of three sequential stages. First, Fast Non-Local Means (FNLM) denoising is applied to reduce image noise while preserving relevant anatomical structures. Second, lung regions are isolated using intensity-based thresholding followed by morphological operations and contour-based refinement. The resulting lung masks are applied to the denoised CT images to suppress irrelevant surrounding structures. Finally, the segmented lung images are resized and normalized and supplied to a Vision Transformer (ViT), which learns contextual representations for classification. The framework is evaluated using a single open-access CT image dataset under two classification settings: binary classification of cancerous and non-cancerous CT images and multiclass discrimination among normal, benign, and malignant cases. Results: In binary lung cancer classification, the proposed framework correctly classified 982 of 1000 CT images, achieving an accuracy of 98.20%, with sensitivity, specificity, precision, and F1 scores of 98.20%. For multiclass classification of normal, benign, and malignant cases, the framework correctly classified 951 of 1005 CT images, corresponding to an overall accuracy of 94.63%. Class-wise precision, recall, specificity, F1 score, and ROC-AUC were additionally evaluated to provide a more comprehensive assessment of classification performance. These findings reflect performance under the evaluated experimental conditions. Conclusions: The proposed framework demonstrates promising performance for CT-based lung cancer classification by integrating FNLM denoising, conventional lung-region extraction, and transformer-based representation learning. However, because the evaluation relied on a single open-access dataset with image-level partitioning, the generalizability of the findings to independent patient populations and clinical settings remains uncertain. Further patient-disjoint evaluation, external validation using multi-institutional datasets, and prospective clinical assessment are required before its potential clinical utility can be established.

To ensure the sustainable development of agriculture, the vegetation and water resources must be monitored accurately in the context of a changing climate. Spatial resolution, however, is a limitation of conventional satellite-based remote sensing methods, and they may not be able to detect the fine-scale patterns of agricultural activity in areas with low water availability. The integrated framework of multisensor remote sensing with super-resolution image enhancement for detecting agriculture and groundwater (GW) conditions in the Al Qassim region of Saudi Arabia is presented. Landsat, Gravity Recovery and Climate Experiment (GRACE), PlanetScope, and Climatic Research Unit Time-Series (CRU-TS) climate data were combined to assess the vegetation indices (VIs) Normalized Difference Vegetation Index (NDVI), Soil-Adjusted Vegetation Index (SAVI), Modified Soil-Adjusted Vegetation Index (MSAVI2), and their associations with temperature, rainfall, and groundwater storage (GWS). Three image enhancement techniques were considered, such as Bicubic interpolation, Super-Resolution Convolutional Neural Network (SRCNN), and the Super-Resolution Generative Adversarial Network (SRGNN)-based super-resolution model. The SRGAN model had the smallest root mean square error (RMSE) of 8.27 image intensity units in the 0–255 scaled reflectance space, and the highest peak signal-to-noise ratio (PSNR) and structural similarity index (SSIM) scores of 36.48 and 0.935, respectively. The results indicate that there is clear seasonal and annual variability in the vegetation indices, and there is a relationship between the vegetation condition at the regional scale and the climate variables, such as temperature, rainfall, and GWS. Spatial analysis also suggests that responses to vegetation and GW are not the same throughout the study area, due to the variable nature of agriculture, water availability, and climatic stresses encountered locally. The results show that SRGAN can enhance the spatial interpretability of the multi-sensor remote sensing imagery (RSE) and aid in climate-resilient monitoring of agriculture and water resource management.

Automated retinal vessel segmentation can support quantitative analysis of fundus photographs and facilitate measurement of vascular characteristics associated with ocular and systemic disease. Although deep-learning methods achieve high segmentation performance on benchmark datasets, their reliability may decrease when applied to images obtained under different acquisition conditions. This study evaluated the cross-dataset generalizability of a U-Net segmentation architecture incorporating an ImageNet-pretrained ResNet34 encoder. Sixteen images from the DRIVE dataset were used for model training and four were retained for in-domain validation. Images were resized to 256 × 256 pixels and processed using a consistent normalization pipeline, while spatial and photometric augmentation was applied during training. Without target-domain retraining or domain adaptation, the trained model was subsequently evaluated on the independent CHASE_DB1 dataset. Segmentation performance was assessed using the Dice coefficient, intersection over union (IoU), and sensitivity, together with qualitative examination of predicted vessel maps. On the held-out DRIVE images, the model achieved a Dice coefficient of 0.7263, IoU of 0.5703, and sensitivity of 0.7201. On CHASE_DB1, performance decreased to 0.6269, 0.4573, and 0.5502, respectively. The disproportionately larger reduction in sensitivity indicated an increased frequency of missed vessel pixels under cross-dataset evaluation. Qualitative examination showed greater fragmentation and omission of fine and peripheral vascular structures in CHASE_DB1 images. These findings demonstrate that within-dataset performance may overestimate the robustness of retinal vessel segmentation models when transferred to a different imaging domain. External evaluation should therefore form an important component of retinal imaging artificial intelligence research. Future work should investigate higher-resolution processing, vessel-width-specific evaluation, multi-domain training, domain-adaptation strategies, and larger multicentre datasets to improve model generalizability.

Accurate estimation of crop evapotranspiration (ETc) is fundamental to irrigation planning and sustainable agricultural water management, particularly under increasing climate variability and water scarcity. Conventional flux measurement systems, including eddy covariance towers and lysimeters, provide high-quality observations but are costly, maintenance-intensive, and spatially constrained, limiting their scalability for precision water management. This study presents AquaVolt-AI, a physics-informed machine learning framework that integrates Sentinel-2 optical imagery, NASA ECOSTRESS thermal observations, and meteorological data with the FAO-56 dual crop-coefficient formulation to generate spatially explicit ETc estimates without requiring dedicated on-site sensing infrastructure for routine operation. The framework couples a dynamic residual neural network with physics-based constraints and an automated state-estimation mechanism designed to maintain inference during satellite data gaps and external data-service interruptions. AquaVolt-AI was evaluated over 36 days (28 June–3 August 2026) at the UC Davis Russell Ranch Sustainable Agriculture Facility using ground-based CIMIS observations for validation and ECOSTRESS thermal data as an auxiliary model input. ETc was represented across a 16 × 16 virtual sensing grid comprising 256 spatial sectors at 10 m resolution. The framework achieved a root mean square error of 0.3000 mm day⁻¹ and a mean absolute error of 0.2688 mm day⁻¹. During a consecutive 9-day satellite data gap, the physics-informed state estimator maintained continuous ETc predictions without detectable empirical drift in the evaluated period. These findings demonstrate the feasibility of integrating Earth observation, meteorological information, and physics-informed machine learning within a low-infrastructure computational framework for spatially resolved ETc monitoring. The approach provides a scalable foundation for precision irrigation assessment and data-driven agricultural water management, although broader multi-season and multi-site validation is required to establish transferability across cropping systems and agroclimatic environments. The main novelty of this study is in coupling a bounded residual neural correction to the FAO-56 dual crop coefficient model within a fully serverless architecture, eliminating on-site sensing hardware while preserving physical plausibility during data outages.

LULCC is a major driver of habitat fragmentation, biodiversity loss, and declining landscape connectivity, particularly in rapidly urbanizing regions. Although Islamabad has experienced substantial urban expansion and forest-cover change, long-term city-scale assessments linking land-cover dynamics with landscape connectivity remain limited. This study examined spatiotemporal LULCC in Islamabad from 1991 to 2021 and assessed whether recent forest recovery improved landscape structural connectivity. Landsat images acquired in 1991, 2001, 2011, and 2021 were classified into five land-cover categories: water, forest, built-up area, bare land, and agricultural land. Classification was performed using the Random Forest (RF) algorithm in Google Earth Engine (GEE). Landscape composition and spatial configuration were quantified using FRAGSTATS 4.3, while forest fragmentation was evaluated using the Landscape Fragmentation Tool v2.0 (LFT) with a 100 m edge threshold. The classifications achieved overall accuracies above 90%, with Kappa coefficients (K) greater than 0.85. Built-up area increased from 76.31 km², representing 7.55% of the study area, in 1991, to 258.62 km², or 25.60%, in 2021, demonstrating rapid urban expansion and associated habitat conversion. Forest cover increased to 340.86 km² in 2001, declined to 271.96 km² in 2011, and subsequently recovered to 409.22 km² in 2021. Despite this increase in forest extent, fragmentation metrics indicated persistent spatial subdivision and limited structural connectivity. High patch density (PD), reduced landscape aggregation, and changes in the largest patch index (LPI) indicated persistent spatial subdivision and limited habitat continuity. These findings highlight the value of integrating RF-based land-cover classification, multitemporal remote sensing, and landscape metrics for urban environmental monitoring. The findings suggest that future land-use planning should consider landscape connectivity, protection of existing forest patches, and spatially coordinated restoration alongside continued reforestation.

Background: Neonatal asphyxia is a life-threatening perinatal condition associated with neonatal mortality and long-term neurological impairment. Therefore, early identification of high-risk newborns is essential for timely clinical intervention; however, accurate prediction of neonatal asphyxia remains a key challenge due to heterogeneous maternal and perinatal risk factors, for example, overlapping clinical patterns, and class imbalance in medical datasets. Thus, to solve this issue, this research proposes an imbalance-sensitive machine-learning framework for the prediction of neonatal asphyxia using perinatal and maternal clinical variables. In contrast to conventional SMOTE, which interpolates minority samples without explicitly prioritizing their diagnostic difficulty, the proposed framework is designed to strengthen the representation of informative asphyxia cases located in uncertain or overlapping decision regions while reducing the influence of less representative synthetic generation. Methods: In summary, in this research, a novel synthetic oversampling framework, named HEM-SMOTE, is proposed to improve minority-class representation by generating more informative synthetic asphyxia samples through hybrid distance-guided neighbor selection. The proposed method integrates local Euclidean-distance-based similarity with Mahalanobis-distance-based covariance awareness to identify representative minority neighbors prior to the generation of synthetic samples. Results: To measure the performance, we compared our proposed framework with the traditional classifiers, for instance, logistic regression, support vector machine, random forest, balanced random forest, Extra Trees, gradient boosting, XGBoost, LightGBM, and CatBoost. The comparison is performed based on accuracy, balanced accuracy, precision, recall, specificity, F1-score, and AUROC. The experimental results show that the proposed framework achieved the strongest overall performance, and achieved 95.35% accuracy as compared with conventional SMOTE and class-weighted baselines. The proposed approach provided more balanced predictive performance across discrimination and classification metrics.

Background/Objectives: Brain tumor segmentation from magnetic resonance imaging (MRI) plays an important role in clinical assessment and treatment planning. However, accurate segmentation remains challenging because of the complex anatomical structure of the brain, variations in tumor size and shape, and the imbalance between tumor and non-tumor regions in MRI datasets. These challenges highlight the need for reliable automated segmentation methods. Methods: This study proposes a multilevel deep learning model for automated brain tumor segmentation using MRI images. The BraTS dataset was used for model development and evaluation. To address class imbalance, a modified Synthetic Minority Oversampling Technique (SMOTE) was incorporated during preprocessing. A Multilevel Architecture-Based Modified U-Net was then employed to learn multiscale spatial features and generate pixel-wise tumor segmentation. The proposed framework was evaluated using the Dice coefficient, Jaccard coefficient, Matthews Correlation Coefficient (MCC), and accuracy. Results: The experimental results demonstrate that the proposed model consistently outperformed the Berkeley Wavelet Transform (BWT)-based method and the conventional U-Net across different tumor grades. Higher Dice, Jaccard, and MCC values indicate improved agreement between the predicted segmentation and the expert-annotated ground truth masks, demonstrating more accurate and consistent tumor delineation. Conclusions: The proposed multilevel deep learning model provides an effective framework for automated brain tumor segmentation from MRI images. By combining imbalance-aware preprocessing with a lightweight Modified U-Net architecture, the proposed method improves segmentation performance while maintaining a relatively simple network design. Future work will focus on validating the proposed framework using external clinical datasets and comparing it with recent state-of-the-art segmentation models.

With increased Internet of Things (IoT) devices, complexity and protection are more challenging. Lightweight cryptographic algorithms are secure and suitable for limited-resource environments; however, their hash functions provide encrypted data but not integrity. Strong security features are available, but setup is difficult and expensive. Network security mechanisms increase power consumption and latency. As IoT networks grow, managing cryptographic keys and securely authenticating large numbers of devices become complex tasks. Efficient key management strategies are required to ensure the scalability required. Existing state-of-the-art solutions lack standardization, scalability, complex and costly. Thus, this research proposes a secure solution for IoT resource-constrained devices, combining strong data integrity and lightweight encryption, and is thus named a hybrid. This hybrid approach integrates SHA-512 and the present cipher in our proposed approach and thus ensuring higher security than state-of-the-art models. This intelligent combination not only enhances the algorithm’s resistance against cryptographic attacks but also improves its processing speed. The proposed approach is used to reduce the processing time for encryption in the IoT platform and to preserve the trade-off between security and efficiency. In terms of memory use, execution time, and precision, the proposed approach is compared with recent state-of-the-art research. The experimental results indicate that our approach is efficient using the avalanche, authentication success rate, collision events, and execution time. The efficiency is 53% to 65%, and the avalanche effect indicates sensitivity to input variations, suggesting moderate-to-considerable reactivity to small data changes. The experimental tests conducted across 10,000 and 80,000 runs reveal no collisions and found that the proposed approach is resilient in managing unique IDs. Moreover, our approach performs consistently, with an average execution time of 0.088246 s, ranging from 0.075954 to 0.094583 s. Finally, our approach provides a practical and scalable solution for securing IoT devices in resource-constrained environments, addressing practical problems for IoT devices.

Access to safe drinking water is a fundamental determinant of global health. The presence of contaminated water affects the citizens’ health. Per- and polyfluoroalkyl substances (PFAS) are often referred to as forever chemicals. They pose a persistent and growing threat to drinking water. In the literature, machine learning methods are used to identify the forever chemicals in water. However, traditional methods are not efficient and scalable. Thus, to solve this issue. This study develops a large-scale machine-learning framework for PFAS risk screening in US public water systems. The proposed framework incorporates data ingestion, preprocessing, and feature engineering. We have used SMOTE for correcting imbalanced data. We performed experimentation and also evaluated our ensemble-based framework integrating Gradient boosting, bagging, and meta-learning strategies. The proposed framework achieves a maximum ROC-AUC of 0.9574, with the best-performing stacking ensemble achieving a precision of 0.75, a recall of 0.68, and an F1-score of 0.71. The simulation results show that the proposed ensemble learning framework is useful for screening and identifying water systems.