Javed Rashid

Proper and accurate classification of skin lesions is essential for melanoma diagnosis. Distinguishing lesion regions from surrounding healthy skin is difficult due to blurred tissue patterns, irregular shapes, and faint boundaries. Conventional segmentation techniques often lose high-frequency structural details during downsampling, which compromises boundary precision. This article proposes WIDENet, a lightweight segmentation framework that preserves fine lesion boundaries while maintaining computational efficiency. WIDENet combines a multi-resolution wavelet, a transformer-based contextual encoder, and an attentive feature refinement mechanism. In contrast to classical encoder-decoder systems, spatial downsampling is performed in the frequency domain via wavelet decomposition, thereby reducing feature dimensionality while preserving multiscale structural information. The encoder based on transformers enables long-range semantic perception. In contrast, the attention-learned refinement framework focuses on information related to boundaries and suppresses background noise. We tested WIDENet using three benchmark dermoscopic datasets, including $\mathrm {PH}^{2}$ , ISIC 2017, and ISIC 2018. Comparison with state-of-the-art models using U-Net, U-Net++, DeepLabV3(+), PAN, PSPNet, and LinkNet demonstrates that WIDENet consistently outperforms them. In particular, it achieves Dice scores of 0.9384 on $\mathrm {PH}^{2}$ and 0.8752 on ISIC 2017, and high precision, recall, and IoU across all datasets. These results demonstrate that WIDENet is not only correct but also highly computationally efficient, making it a promising tool for real-world melanoma screening and clinical decision support systems.

Finding rapid and precise methods of identifying diseases is necessary in order to safeguard agricultural yield in today’s global issues of food security. Citrus is a crucial crop in the Pakistani economy, which has decreasing productivity and quality due to foliar diseases. Conventional detection assessment requires manual work, and opinion. The proposed paper develops a Citrus Multiple Leaf Disease Detection system based on YOLOv11 architecture in real-time. To create and develop the CLDD dataset, 812 pictures and 3727 samples of citrus canker, greasy spot, and leaf miner illnesses in the Okara orchards in Punjab, Pakistan have been used. The data set was more appropriate to be generalized by data augmentation, when it had been prepared and annotated. The YOLOv11 model showed remarkable results with a precision of 79.9% and recall of 77.4% and mean Average Precision 81.2% mAP50. This system supports real-time applications by processing photos in 10.5 ms. Ablation research showed that increasing model robustness requires a CSPDarknet53 backbone, a PANet neck, and sophisticated augmentation approaches. Most disease classes performed well; however, citrus greasy spot showed slight variation due to environmental factors and similar illness symptoms. Class-enriched fine-tuning and the construction of artificial datasets should be improved for future use. The findings show that deep learning frameworks can automatically detect citrus illnesses in real time, enabling accurate agricultural solutions for Pakistan and other locations.

A reliable forecast of electricity generation is essential for long‐term energy planning and sustainable operation of the power system with the coexistence of renewable and nonrenewable energy sources. This paper presents a temporal convolutional network (TCN)‐based approach for the long‐term forecasting of total renewable electricity production (TREP) and total nonrenewable electricity production (TNREP) for European and American countries. The approach uses causal and dilated convolution to effectively extract long‐range dependencies. A multi‐country dataset from 2000 to 2025 is considered for analysis using country‐wise modeling with min–max normalization. The performance of the models is measured in terms of mean squared error (MSE), root mean squared error (RMSE), and mean absolute error (MAE), and is compared with bidirectional long short‐term memory (Bi‐LSTM), bidirectional gated recurrent unit (Bi‐GRU), and ensemble deep learning models. The performance of the proposed TCN model is always the best in terms of prediction error on all the datasets considered, with the lowest error values of MSE = 0.0035, RMSE = 0.0599, and MAE = 0.0480. The European datasets have lower values of forecasting errors because of the stable energy systems, while the American datasets produce higher total electricity with variability, resulting in relatively higher values of prediction errors. In summary, the proposed TCN framework has shown robustness and scalability in the long‐term forecasting of electricity generation across multiple countries.

Past studies have proposed machine learning-based solutions for classifying software requirements, helping analysts, software architects, and developers efficiently and effectively identify the types of requirements. A common step performed by these solutions is to represent requirements as meaningful vectors. These vectors are used to identify and classify software requirements. The quality of the vector representations of software requirements determines the effectiveness, efficiency, and performance of the proposed solution. This study introduced a domain-specific pretrained model, Req2Vec, that extracts a distributed representation of non-functional software requirements (NFRs). Req2Vec is an unsupervised, domain-specific pre-trained model trained on thirty million samples of functional and non-functional requirements (FR and NFR). Req2Vec is aware of various requirements, including both functional and non-functional requirements, and seamlessly integrates them to learn representations of requirements. Using a limited supervised dataset, Req2Vec was investigated for various requirement classification tasks, including binary classification, reduced sets of NFR classifications, and all NFR classifications. In binary classification, the Req2Vec model achieved precision, recall, and F1 scores of 0.91, 0.87, and 0.89, respectively. For the reduced set of NFR classification, the average scores are 0.89, 0.85, and 0.87. For all classes, the average scores are 0.83, 0.77, and 0.80. A comparative analysis revealed that Req2Vec yields promising results in advancing technologies, demonstrating its effectiveness in accurately classifying NFRs.

Skin cancer, particularly melanoma, remains a major public health challenge because of its rising incidence and mortality rates. Traditional methods of diagnosis, like dermoscopy and biopsies, are invasive, time‐consuming, and highly dependent on clinical experience. Furthermore, previous research has predominantly focused on 2D dermoscopic images, which do not capture important volumetric information required for the proper evaluation of the injury. This work introduces a new deep learning architecture based on the ResNet‐18 model, augmented by transfer learning, for binary classification of malignant and benign skin lesions. The model is trained on the ISIC 2024 3D Total Body Photography dataset and uses pre‐trained ImageNet weights to enable effective feature extraction. To counter the dataset's natural class imbalance and minimize overfitting, the model uses sophisticated data augmentation and oversampling methods. The suggested model boasts a staggering classification accuracy of 99.82%, surpassing many other 2D‐based alternatives. The utilization of 3D‐TBP offers a strong diagnostic benefit by allowing volumetric lesion analysis, retaining spatial and depth features usually lost in the conventional 2D images. The findings validate the clinical feasibility of the method, presenting a scalable, noninvasive, and very accurate early detection and diagnosis of melanoma using 3D skin imaging.

Brain tumors have a big effect on a person’s health by letting abnormal cells grow unchecked in the brain. This means that early and correct diagnosis is very important for effective treatment. Many of the current diagnostic methods are time-consuming, rely primarily on hand interpretation, and frequently yield unsatisfactory results. This work finds brain tumors in MRI data using DenseNet121 architecture with transfer learning. Model training made use of the Kaggle dataset. By preprocessing the stage, resizing the MRI pictures to minimize noise would help the model perform better. From one MRI scan, the proposed approach divides brain tissues into four groups: benign tumors, gliomas, meningiomas, and pituitary gland malignancies. The designed DenseNet121 architecture precisely classifies brain cancers. We assessed the models’ performance in terms of accuracy, precision, recall, and F1-score. The suggested approach proved successful in the multi-class categorization of brain tumors since it attained an average accuracy improvement of 96.90%. Unlike previous diagnostic techniques, such as eye inspection and other machine learning models, the proposed DenseNet121-based approach is more accurate, takes less time to analyze, and requires less human input. Although the automated method ensures consistent and predictable results, human error sometimes causes more unpredictability in conventional methods. Based on MRI-based detection and transfer learning, this paper proposes an automated method for the classification of brain cancers. The method improves the precision and speed of brain tumor diagnosis, which benefits both MRI-based classification research and clinical use. The development of deep-learning models may even further improve tumor identification and prognosis prediction.