Explore the vast range of titles available.

One fascinating and challenging research topic in the field of computer vision and image processing is the classification of mushroom images into multiple categories. Mushrooms fall into many categories worldwide. While some varieties of mushrooms are edible, others are toxic or not edible, and the shapes and color combinations present an intriguing challenge for their classification from images. This challenge is becoming easier to solve because of the recent success of deep learning (DL)-based models, particularly convolutional neural networks (CNN), on images for feature extraction and classification. Furthermore, the integration of transfer learning approaches with different attention mechanisms significantly improves classification accuracy over time. Even so, there is still a huge potential for improved performance by experimenting with more sophisticated DL-based approaches and introducing more mushroom categories, each with more images. In this work, we achieved validation accuracies of 95.85% and 93.68% on two different mushroom image datasets, outperforming other comparable DL-based systems by integrating the convolutional block attention module (CBAM) with a transfer learning-based Xception architecture. A 5-fold cross-validation approach is employed to enhance the model’s generalizability and ensure greater robustness and consistency in the results. The novelty of our model lies in integrating the convolutional block attention module (CBAM) with a transfer learning-based Xception architecture, selectively unfreezing the last 4 blocks to enhance feature extraction. This unique combination improves the model’s attention to critical spatial and channel information, resulting in superior performance compared to conventional transfer learning approaches.

Social media platforms like Twitter and Facebook have gradually become vital for communication and information exchange. However, this often leads to the spread of unreliable or false information, such as harmful rumors. Currently, graph convolutional networks (GCNs), particularly TextGCN, have shown promise in text classification tasks, including rumor detection. Their success is due to their ability to identify structural patterns in rumors and effectively use neighborhood information. We present a novel rumor detection model using TextGCN, which utilizes a word-document graph to represent rumor texts. This model uses dual embedding from two pre-trained transformer models: generative pre-trained transformers (GPT) and bidirectional encoder representations from transformers (BERT). These embeddings serve as node representations within the graph, enhancing rumor detection. Combining these deep neural networks effectively extracts significant contextual features from rumors. This graph undergoes convolution, and through graph-based learning, the model detects a rumor. We evaluated our model using publicly available rumor datasets, such as PHEME, Twitter15, and Twitter16. It achieved 88.64% accuracy on the PHEME dataset, surpassing similar models, and performed well on Twitter15 and Twitter16 with accuracies of 81.98% and 83.41%, respectively.

Solving math word problems of varying complexities is one of the most challenging and exciting research questions in artificial intelligence (AI), particularly in natural language processing (NLP) and machine learning (ML). Foundational language models such as GPT must be evaluated for intelligence, and solving word problems is a key method for this assessment. These problems become especially difficult when presented in low-resource regional languages such as Bengali. Word problem solving integrates the cognitive domains of language processing, comprehension, and transformation into real-world solutions. During the past decade, advances in AI and machine learning have significantly progressed in addressing this complex issue. Although researchers worldwide have primarily utilized datasets in English and some in Chinese, there has been a lack of standard datasets for low-resource languages such as Bengali. In this pioneering study, we introduce the first Bengali Math Word Problem Benchmark Data Set (BMWP), comprising 8653 word problems. We detail the creation of this dataset and the benchmarking methods employed. Furthermore, we investigate operation prediction from Bengali word problems using state-of-the-art deep learning (DL) techniques. We implemented and compared various standard DL-based neural network architectures, achieving an accuracy of$$92 \\pm 2\\%$$92 ± 2 % . The data set and the code will be available at https://github.com/SanchitaMondal/BMWP .

A bstract Top polarization is an important probe of new physics that couples to the top sector, and which may be discovered at the 14 TeV LHC. Taking the example of the MSSM, we argue that top polarization measurements can put a constraint on the soft supersymmetry breaking parameter A t . In light of the recent discovery of a Higgs-like boson of mass ∼ 125 GeV, a large A t is a prediction of many supersymmetric models. To this end, we develop a detector level analysis methodology for extracting polarization information from hadronic tops using boosted jet substructure. We show that with 100 fb −1 of data, left and right 600 GeV stops can be distinguished to 4 σ , and 800 GeV stops can be distinguished to 3 σ .

Aim This study aims​ to report preclinical validation, and the first clinical treatment of total bone marrow irradiation (TMI) and total bone marrow and lymph nodal irradiation (TMLI) using Volumetric modulated arc therapy in Halcyon-E ring gantry linear accelerator. Preclinical validation includes simulation, planning, patient-specific QA, and dry run. Material and method Four patients, two female and two male, with body weights of 116 kg, 52 kg, 64 kg, and 62 kg; with two with chronic myeloid leukemia, one each with acute lymphoblastic leukemia and acute myeloid leukemia (AML) were simulated and planned for TMI/TMLI. Patients were immobilized with a full-body vacuum bag. Head first supine (HFS) and Feet first supine (FFS) CT scans were acquired from head to knee and knee to toe. Planning target volume (PTV) was created with a uniform margin of 6 mm over the total bone marrow/bone marrow + lymph nodes. HFS and FFS PTVs were optimized independently using 6MV unflatten energy for 12 Gy in 6 fractions. Plans were merged to create the resultant dose distribution using a junction bias dose matching technique. The total number of isocenters was ≤ 10 per CT set, and two to four full arcs were used for each isocenter. A junction dose gradient technique was used for dose feathering between arcs between adjacent isocenters. Result Only one female patient diagnosed as AML received the TMLI treatment, while the other three patients dropped out due to clinical complications and comorbidities that developed in the time between simulation and treatment. The result presented has been averaged over all four patients. For PTV, 95% dose was normalised to 95% volume, PTV_V107% receiving 3.3 ± 3.1%. Total lung mean and V12Gy were 1048.6 ± 107.1 cGy and 19.5 ± 12.1%. Maximum lens doses were 489.5 ± 35.5 cGy (left: L) and 497 ± 69.2 cGy (right: R). The mean cardiac and bilateral kidney doses were 921.75 ± 89.2 cGy, 917.9 ± 63.2 cGy (L), and 805.9 ± 9.7 cGy (R). Average Monitor Unit was 7738.25 ± 1056.6. The median number of isocenters was 17(HFS+FFS), average MU/Dose (cGy) ratio per isocenter was 2.28 ± 0.3. Conclusion Halcyon-E ring gantry linear accelerator capable of planning and delivering TMI/TMLI.​​

This study describes a new method for gripping and sensing a physical object (or material) with a prosthetic arm that uses haptic technologies, kinaesthetic communication, and machine learning. Haptic technology is a method of determining if an object is firm or soft as if it were gripped by a human and determining how much gripping force the object can withstand without crushing it. The bending moment and gripping force are measured using a flex sensor in human fingers and a pressure sensor applied by the tip of the human fingers. Three different types of objects (soft sponge, hard sponge, and plastic) are studied and tested in this work by pressing them with varying gripping pressures (soft, firm, and firmer). In addition, a model (Haptic Intelligence Recorder arm) is proposed that can anticipate the object type and gripping force based on the recorded intelligence data. The major goal is to educate our prosthetic hand to be able to grip various items with varying finger pressures, much like we can do naturally. Finally, a glove is created that is tailored to the intelligence arm’s ability to anticipate grabbing items.