Children diagnosed with Attention-Deficit/Hyperactivity Disorder (ADHD) face many difficulties in maintaining their concentration (in terms of attention levels) and controlling their behaviors. Previous studies have mainly focused on identifying brain regions involved in cognitive processes or classifying ADHD and control subjects. However, the classification of attention levels of ADHD subjects has not yet been explored. Here, a robust Swin Transformer (Swin-T) model is proposed to classify the attention levels of ADHD subjects. The experimental cognitive task ‘Surround suppression’ includes two events: Stim ON and Stim OFF related to the high and low attention levels of a subject. In the proposed framework, ADHD-specific channels are initially identified from input Electroencephalography (EEG). Next, the significant, non-noisy connectivity features are extracted from those channels through the Singular Value Decomposition (SVD) method. Finally, the non-noisy features are passed to the robust Swin-T model for attention-level classification. The proposed model achieves 97.28% classification accuracy with 12 subjects. The robustness of the proposed model leads to potential benefits in EEG-based research and clinical settings, enhancing the reliability of ADHD assessments.

This dataset contains simultaneous EEG-fMRI recordings for inner speech experiments. Data were collected using a 3T MRI scanner and 64-channel BrainProducts EEG system. The EEG data have undergone preprocessing, including pulse artifact removal, using the BrainVision Analyzer software. No further data transformations have been applied to ensure the dataset remains BIDS-compliant as "raw".

This article presents an intervention study on the effects of the combined methods of 1) the Socratic method, 2) chain-of-thought (CoT) reasoning, 3) simplified gamification, and 4) formative feedback on university students’ math learning driven by large language models (LLMs). We call our approach Mathematics Explanations through Games by AI LLMs (MEGA). Some students struggle with math, and as a result, avoid math-related disciplines or subjects despite the importance of math across many fields, including signal processing. Oftentimes, students’ math difficulties stem from suboptimal pedagogy. We compared the MEGA method to the traditional step-by-step (CoT) method to ascertain which is better by using a within-group design after randomly assigning questions for the participants, who are university students. Samples (n=60) were randomly drawn from each of the two test sets of the Grade School Math 8 K (GSM8K) and Mathematics Aptitude Test of Heuristics (MATH) datasets, based on an error margin of 11%, a confidence level of 90%, and a manageable number of samples for the student evaluators. These samples were used to evaluate two capable LLMs at length [Generative Pretrained Transformer 4o (GPT4o) and Claude 3.5 Sonnet] out of the initial six that were tested for capability. The results showed that students agree in more instances that the MEGA method is experienced as better for learning for both datasets. It is even much better than the CoT (47.5% compared to 26.67%) in the more difficult MATH dataset, indicating that MEGA is better at explaining difficult math problems. We also calculated the accuracies of the two LLMs and showed that model accuracies differ for the methods. MEGA appears to expose the hallucination challenge that still exists with these LLMs better than CoT. We provide public access to the MEGA app, the preset instructions that we created, and the annotations by the students for transparency.

With the growing demand for raw materials, there is also an increased need in faster and more efficient processes of mineral exploration, especially locating possible materials to mine. Machine learning (ML) for supporting the labor-intensive and interpretive process of drill core logging becomes increasingly relevant to address these needs, since drill cores are the most direct and physically preserved record of subsurface geology available during mineral exploration. This paper reviews the current state of the art in ML-based drill core analysis, including its capabilities, limitations, and specific challenges related to generalization and practical deployment within geological workflows. The review focuses specifically on photographic images of drill core, which have been routinely used in mineral exploration for decades. This paper presents several major contributions: It offers a structured overview of current methods, organized around three key geological tasks, which are lithology prediction, geotechnical analysis, and mineralogical prediction. Additionally, it identifies potential research gaps and proposes directions for future work, concluding with an emphasis on advancing context-aware machine learning in drill core analysis through a human-in-the-loop approach.

Inner speech, or covert speech, refers to the internal generation of language without overt articulation. Decoding inner speech has significant implications for brain-computer interfaces (BCIs), particularly for assistive communication in individuals with speech and motor impairments. To facilitate research in this area, we introduce a publicly available dataset comprising simultaneously recorded electroencephalography (EEG) and functional magnetic resonance imaging (fMRI) data during inner speech production.Data were collected from three healthy, right-handed participants performing an inner speech task. The task involved silent repetition of visually presented words belonging to either a social or numerical category. The experiment consisted of 40 trials per word, with eight unique words and starts with a fixation period of two seconds. Stimuli were displayed for two seconds at the beginning of each session, followed by a 12-second rest period to allow hemodynamic responses to return to baseline. Participants were instructed to remain still and avoid movements to minimize artifacts.EEG was recorded using a 64-channel MR-compatible cap (BrainCap MR, EasyCap GmbH) at a 5 kHz sampling rate. Electrocardiogram (ECG) signals were simultaneously acquired using an additional electrode placed on the trapezius muscle to facilitate cardioballistic artifact correction. Gradient and cardioballistic artifacts were corrected using BrainVision Analyzer software.Functional MRI data were acquired using a 3T scanner with a 48-channel headcoil, and an echo-planar imaging (EPI) sequence optimized for whole-brain coverage. The repetition time (TR) was 2 s. High-resolution anatomical T1-weighted images were also acquired for structural reference. The dataset is publicly available in the OpenNeuro repository.The aim of this dataset is to provide a resource for studying inner speech processing, multimodal neuroimaging, EEG-fMRI fusion techniques, and BCI-driven speech prosthesis development.

Many base and precious metals are sourced from volcanic massive sulphide (VMS) deposits and understanding the geological characteristics of such deposits is crucial for new discoveries of this deposit type. Although key geological characteristics of modern VMS systems are relatively well understood, a remaining challenge is resolving the same geological characteristics in ancient, complex, altered and metamorphosed VMS deposits. One such deposit is the Palaeoproterozoic Rävliden North deposit, an 8.7 Mt (combined resources and reserves of 3.42 % Zn, 0.90 % Cu, 0.54 % Pb, 81 g/t Ag, and 0.24 g/t Au) replacement-style volcanic massive sulphide deposit in the felsic-bimodal western Skellefte district, northern Sweden. The VMS deposits in the Skellefte district are hosted in rocks subjected to greenschist to amphibolite facies metamorphism and occur at the lithostratigraphic contact between the metavolcanic 1.89 – 1.88 Ga Skellefte group (SG) and stratigraphically overlying metasiliciclastic 1.89 – 1.87 Ga Vargfors group (VG). Intense hydrothermal alteration commonly eradicates original rock textures, and polyphase deformation and metamorphism make geological interpretation and stratigraphic reconstruction difficult. Hence, to complement lithofacies analysis, immobile element chemostratigraphy is used in this study.

Rävliden North is predominantly hosted by felsic volcanic rocks of the herein defined Rävliden formation in the upper part of the SG that were deposited in half grabens related to rifting of a continental arc. Based on immobile elements and their ratios the felsic rocks fall into three groups, Rhy I, II and III. The chemostratigraphy and lithostratigraphy roughly coincide, where Rhy II (Zr/Al2O3 = 12.86, Al2O3/TiO2 = 36.07, Zr/TiO2 = 0.05) defines the rhyolites beneath the Rävliden formation that predominantly comprises Rhy I (Zr/Al2O3 = 17.23, Al2O3/TiO2 = 32.33, Zr/TiO2 = 0.06) and Rhy III (Zr/Al2O3 = 17.95, Al2O3/TiO2 = 36.53, Zr/TiO2 = 0.07), where Rhy I is the chief host to mineralisation. Mineralisation is partially hosted by graphitic phyllite that overlies the Rävliden formation and represents the base of the VG that indicates paused volcanism important for the build-up of massive sulphides beneath the seafloor. Facies analysis of rhyolites suggest that these were unconsolidated pumice rich rocks permeable for the upwelling hydrothermal fluids. Additionally, graphitic phyllite functioned as a permeability barrier inducing lateral fluid flow resulting in more effective sulphide precipitation.

This study demonstrates the effectiveness of combining stratigraphic, facies and chemostratigraphic analysis for targeting VMS deposits in complex, altered and metamorphosed rocks.

Electroencephalography (EEG) is a fundamental tool in the non-invasive evaluation of brain activity, providing insights into the intricate dynamics at play within neurode-generative disorders. Conventional methodologies often lack in effectively capturing the temporal and intricate intra- and inter-channel dynamics, leading to diminished predictive accuracy. To address this problem, we present an innovative framework that effectively captures temporal along with intra- and inter-channel dynamics for EEG analysis aimed at predicting neu-rodegenerative disorders, explicitly targeting Alzheimer's and dementia. The proposed method involves constructing aggregated recurrence matrices from EEG channels followed by kernel formation and convolution operation, effectively encapsulating intra- and inter-channel spatiotemporal patterns, thereby achieving a more comprehensive representation of neural dynamics. The proposed approach was validated using public datasets, revealing competitive performance. Implementation details with codes will be accessible on GitHub.

This study explores the integration of data fusion using machine learning methods for ore tracking frommine to mill, with the goal of developing predictive geometallurgical models. Conducted at BolidenMineral AB's Garpenberg Zn-Pb-Ag-(Cu-Au) mine in Sweden, the research utilizes extensive geological,operational, and legacy data to create a foundation for a digital twin geometallurgical model for theprocessing plant. By combining 3D geological data with mining and plant operational data, the projectaims to enhance the understanding of ore variability and its impact on processing performance. Thisapproach not only seeks to improve efficiency and reduce variability in production but also providesvaluable insights for more accurate prediction and simulation models in geometallurgy. The outcomesof this research could contribute significantly to the future of data-driven mine planning for optimizedperformance.