Browse Items (16481 total)
Sort by:
-
Privacy-Preserving Federated Learning for Prognostic Modeling in Rare Diseases: A Scalable Case Study on Kawasaki Disease
Predictive modeling in rare diseases faces major challenges, including data scarcity, class imbalance, and strict privacy regulations that limit cross-border collaboration. These challenges are particularly critical in Kawasaki disease (KD)a rare vasculitis in childrenwhere 10% to 20% of patients are resistant to intravenous immunoglobulin (IVIG), the standard first-line treatment. This significantly increases the risk of coronary artery abnormalities (CAA), making early and accurate prediction of resistance to IVIG essential for improving patient outcomes. Our work proposes a federated learning (FL) approach to address the constraints imposed by security and privacy concerns. We investigate convolutional neural networks (CNN) as the shared model, collaboratively trained across clients. Coupled with strategies to address class imbalance resulting from the rarity of the condition, the federated approach yielded promising results when evaluated against conventional machine learning (ML) models. The proposed approach demonstrated strong performance, achieving 94% accuracy, 93% precision, 89% recall, and 91% F1 score. To ensure robustness and generalizability, an independent dataset was also used, where the proposed model excelled similarly. These results highlight the potential of FL to overcome data privacy barriers and provide a scalable, secure solution for predictive modeling in rare diseases, supporting its integration into medical prediction workflows. 2025 by the authors of this article. -
Metabolomics Pathway Prediction Using Enhanced-Graph Convolutional Networks with Graph Attention Networks
Metabolomics, the comprehensive study of small molecules in biological systems, has a central role to play in the diagnosis of diseases, biomarker detection, and the design of new drugs. Although there have been major breakthroughs in analytical toolsets such as mass spectrometry (MS) coupled with chromatography, it is hard to predict metabolomics pathways because biochemical interactions are inherently complex. To meet this end, the current research suggests a deep learning-based approach using graph neural networks (GNN), which have shown high efficiency for graph-structured biological data. We specifically propose an enhanced graph convolutional network integrated with graph attention networks (EGCNGAT) to enhance pathway prediction performance. The hybrid framework employs graph convolutional networks (GCN) to represent molecular structural data and graph attention networks (GAT) to provide context-sensitive feature importance, thus improving the models capacity for learning complex pathway patterns. Comparative experiments against current deep learning approaches show that the introduced EGCN-GAT model obtains an accuracy of 98.90 percent, which is a 0.26 percent increase compared to the baseline MLGL-MP model. In addition, it demonstrates a 0.94 percent gain in precision as well as a slight gain in recall. The findings validate the performance of the proposed method and highlight its utility for developing pathway-level predictions in metabolomics studies. 2025 by the authors of this article. Published under CC-BY. -
Stability and bifurcation analysis of a fractional-order preypredator model with ratio-dependent functional response
This paper explores the dynamics of a fractional preypredator system with a ratio-dependent functional response with memory and hereditary effects in predatorprey interactions. The model is developed by the Caputo fractional derivative, and the existence, uniqueness, positivity, and boundedness of solutions are proven to satisfy biological reality. Stability conditions for local and global stability of both predator-free and coexistence equilibria are proven through linearization and Lyapunov function techniques. The fractional order is used as a bifurcation parameter, and the appearance of Hopf bifurcations is analytically explained with demonstration of the influence of memory on oscillations. To examine discrete-time dynamics, the piecewise constant argument is used to derive a discrete counterpart of the fractional model. The discrete model indicates a wide range of rich complex oscillatory phenomena, including period-doubling and NeimarkSacker bifurcations, leading to periodic, quasiperiodic, and chaotic oscillations. Numerical computations, including bifurcation diagrams, phase portraits, and Lyapunov exponents, verify the analytical results and describe the routes of transition to chaos. A comparative analysis to compare integer- and fractional-order cases indicates that memory effects enhance dynamical richness and sensitivity to parameters. The study provides a unified framework relating continuous fractional dynamics and their discrete implementations and provides additional insight into how memory and discretization interact to modify stability and bifurcation in ecological models. 2026 the Author(s), -
A comparative study of bayesian and classical methods for the weighted Lindley distribution under unified hybrid censoring with survival data applications
In survival analysis and reliability engineering, censoring schemes play a crucial role in efficient data collection and analysis. This study investigated the unified hybrid censoring scheme (UHCS), a versatile framework that integrates multiple censoring strategies, to evaluate the suitability of the Weighted Lindley (WL) distribution for modeling lifetime data. Maximum likelihood estimates (MLEs) and their corresponding asymptotic confidence intervals are derived for the parameters of the WL distribution. In the Bayesian framework, parameter estimation was performed under a squared error loss function. A detailed Monte Carlo simulation study was conducted to compare the performance of classical and Bayesian estimators across various sample sizes and censoring schemes. The simulation results revealed that Bayesian estimators consistently yielded lower mean squared errors (MSEs) than their classical counterparts, and the associated credible intervals were generally narrower than the frequentist confidence intervals. To demonstrate the practical applicability of the proposed methods, the analysis was applied to real-world survival datasets. The results highlighted the effectiveness of the WL distribution under UHCS, offering valuable insights for researchers and practitioners in reliability and survival analysis. 2025 the Author(s), licensee AIMS Press. -
X-Ray Spectral Variability of 13 TeV High-energy-peaked Blazars with XMM-Newton
We present a comprehensive study of the X-ray spectral variability observed in 13 TeV photon-emitting high-energy-peaked BL Lacertae objects (HBLs). These data come from 54 XMM-Newton EPIC-pn pointed observations made during its operational period from 2001 June through 2023 July. We performed spectral studies in the energy range of 0.6-10 keV by fitting X-ray spectra of the pointed observations with power-law and log-parabolic (PL and LP) models. We found at a 99% confidence level that 31 of these X-ray spectra were best fitted with a range of LP models with local photon indices (at 1.0 keV), ? ? 1.75-2.66, and convex curvature parameter ? ? 0.02-0.25. PL models with photon index ? ? 1.78-2.68 best described the spectra of 14-pointed observations. Nine PN spectra resulted in negative curvature parameters in fitting an LP model, and eight among them were significant (? ? 2?err). We fitted broken power-law models to these eight X-ray spectra and found spectral hardening in the range of ?? ? 0.06?0.54 for these observations. EPIC-MOS spectra were also studied for those eight observations to search for similar trends, and we were able to find them in only two, one observation each of PKS 0548-322 and Mrk 501. This indicates the possibility of the coexistence of an inverse Compton component along with the dominant synchrotron component for these two observations. We also performed correlation studies between various LP spectral parameters and briefly discuss their possible implications. 2025. The Author(s). Published by the American Astronomical Society. -
X-Ray Spectral Variability of the TeV High-energy Peaked Blazar PG 1553+113 with XMM-Newton
We present an extensive X-ray spectral variability study of the TeV photon-emitting high-energy-peaked BL Lacertae object PG 1553+113, using the data from the EPIC-PN camera of XMM-Newton, which observed the source during its operational period from 2001 September to 2024 November. X-ray spectra in the energy range, 0.67.0 keV, were fitted with absorbed power-law (PL) and absorbed log-parabola (LP) models. We found with 99% confidence that 14 of them were fit well by LP models having parameters in the range ??2.132.80, and ??0.040.18, one spectrum flavors an LP model with ?<0, while simple PL models with ??2.532.69 were sufficient to describe the X-ray spectra of the remaining 15. Of these 30 observations, 2 showed strong signatures of an additional inverse Compton component, while 1 showed weaker indications. On fitting joint Optical Monitor and EPIC-PN data with LP models, we found synchrotron peaks in the energy range of ?s?4.5948.61 eV. This indicates that the spectral evolution is probably caused by variations in particle acceleration or cooling conditions within the jet. 2026. The Author(s). Published by the American Astronomical Society. Original content from this work may be used under the terms of the https://creativecommons.org/licenses/by/4.0/. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI. -
Spectrotemporal Evolution in XTE J1701-462 during Its 2022 Outburst as Revealed by NICER
We present a comprehensive spectrotemporal study of the 2022 outburst of the transient neutron star low-mass X-ray binary (NS-LMXBs) XTE J1701?462 using 57 NICER observational epochs (E1?E57). The 0.8?10 keV lightcurve exhibits a FRED-like profile with multiple rebrightenings and intensity dips, indicating a nonmonotonic evolution of the accretion flow. Broadband spectral modeling with an absorbed Comptonized disk-blackbody model reveals a coherent evolution of spectral parameters consistent with changes in the disk?corona geometry driven by a varying mass accretion rate. The ??Fbol diagram shows distinct clustering, enabling the identification of six accretion states: LHS-1, IMS-1, HSS, IMS-2, LHS-2, and QS. These states trace the expected cycle of disk truncation, inward propagation, and recession, with notable deviations such as sustained coronal heating in IMS-1 and the HSS, likely caused by changes in coronal geometry or the limited bandpass of NICER. State-resolved hardnessintensity diagrams reveal that XTE J1701?462 exhibits a hybrid phenomenology: island and banana branches characteristic of atoll-state early in the outburst, followed by well-defined horizontal and normal branches during IMS-1 and the HSS. As the source decays through IMS-2 and LHS-2, the HID returns to isolated clumps with increasing hardness before entering quiescence. We detected a quasiperiodic oscillation (QPO) at ?27 Hz with a quality factor Q ? 3.4 during epochs E30 and E31. A Crab-based cross-calibration between NICER and NuSTAR shows that XTE J1701?462 reached a peak accretion rate of ?1.21 (Formula presented) ?Edd, suggesting near- or super-Eddington luminosities consistent with its 2006 outburst. 2026. The Author(s). Published by the American Astronomical Society. Original content from this work may be used under the terms of the https://creativecommons.org/licenses/by/4.0/. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI. -
Tracing Fe K X-Ray Reverberation Lag in the Energy-resolved Spectra of Narrow-line Seyfert 1 Galaxy Ton S180
We report the Fe K relativistic reverberation feature for the first time in the narrow-line Seyfert 1 galaxy Ton S180. Using a long observation from XMM-Newton we find that the Fe K emission lag peaks at 117 49 s in the lag energy spectrum computed for frequencies (0.3-8.5) 10?4 Hz. The lag amplitude drops to 22.85 14.20 s as the frequency increases to (8.5-30) 10?4 Hz. The time-averaged spectrum of the source shows a relatively narrow Fe K line at ?6.4 keV, indicating a low black hole spin ( a = 0 . 4 3 ? 0.14 + 0.10 ) based on the reflection modeling. We perform general relativistic transfer function modeling of the lag energy spectra individually. This provides an independent timing-based measure of the spin at a = 0.3 0 ? 0.17 + 0.34 , a black hole mass M BH = 0.2 9 ? 0.16 + 0.01 1 0 8 M ? , comparable to the previous measurement, and a coronal height h = 2.5 9 ? 0.33 + 5.17 r g . Further, we observe that the Fe K lag and the black hole mass fit well in the linear lag-mass relation shown by other Seyfert 1 galaxies. 2026. The Author(s). Published by the American Astronomical Society. -
Spin Frequency of Neutron Star in the Bright Atoll Source 4U 1705-44 Using NICER Observation
We conducted a systematic study of two thermonuclear type I X-ray bursts (B1 and B2) and the spectral properties of the bright atoll type neutron star low-mass X-ray binary 4U 1705?44 using NICER observation. In our analysis, a burst oscillation at ?702 Hz was detected during the double-peaked profile of burst B1. This provides the first strong evidence for the spin frequency of the neutron star, which establishes 4U 1705?44 as a rapid rotator. The measured convexity parameters of the bursts indicate that both bursts likely ignited at off-equatorial latitudes. The ignition depth of burst B1 was nearly twice that of B2, indicating substantial fuel consumption between the two events. The recurrence time of B2 (?1 hr) categorizes it as a short waiting time burst. The complementary spectral analysis of persistent emission, modeled with tbabsnthcomp+ diskbb), showed a hard state with a photon index ? < 1.12. The hardnessintensity diagram was consistent with the source being in the island state. Notably, the accretion disk appeared to extend close to the neutron star surface. 2026. The Author(s). Published by the American Astronomical Society. -
Discovery of a 459 Hz Burst Oscillation in XTE J1810?189 with NICER
We present a detailed temporal study of a type I X-ray burst from the neutron star low-mass X-ray binary (NS-LMXB) XTE J1810?189, observed on 2023 April 27 using the Neutron Star Interior Composition Explorer. The burst exhibited a rapid rise time of 2.55 s, followed by an exponential decay lasting for 7.5 s, with a total duration of ?13 s. Type I X-ray bursts are driven by thermonuclear burning on the surface of a neutron star in an NS-LMXB. As these bursts originate from the stellar surface they can exhibit highly coherent signals known as burst oscillations, which serve as probes of the neutron stars spin frequency. We report the detection of a burst oscillation signal at ?459 Hz at the cooling tail of the burst. The oscillation showed a strong Leahy-normalized power of PL = 35.95 at 458.92 Hz, corresponding to a single-trial significance of 5.53? and a multiple-trial corrected significance of 3.14?. The folded pulse profile in the 0.212 keV band is well described by a constant plus sinusoid with a fractional rms amplitude of 14.63%. These results suggest that the burst oscillation frequency of XTE J1810-189 directly reflects on the neutron stars spin, measured here to be ?2.18 ms, placing it among the rapidly rotating NS-LMXBs. This burst oscillation signal at the cooling tail of the burst can be interpreted through surface mode model or the asymmetric cooling wake model. 2025. The Author(s). Published by the American Astronomical Society. -
Probing the Origin of X-Ray Flares in the Low-hard State of GRS 1915+105 Using AstroSat and NuSTAR
We performed a detailed time-resolved spectral study of GRS 1915+105 during its low-flux rebrightening phase using the broadband capabilities of AstroSat and NuSTAR in 2019 MayJune. The AstroSat light curves revealed erratic X-ray flares with count rates rising by a factor of ?5. Flares with simultaneous LAXPC and SXT coverage were segmented and fitted using two degenerate but physically motivated spectral models: a reflection-dominated model (hereafter model A) and an absorption-dominated model (hereafter model B). In model A, the inner disk radius (Rin) shows a broken power-law dependence on flux, indicating rapid inward motion of the disk at higher flux levels. In contrast, model B shows variable column density in the range of 10231024 cm?2, displaying a strong anticorrelation with flux. Both models exhibit significant variation in the ionization parameter between low- and high-flux segments. The total unabsorbed luminosity in the 0.730 keV energy range ranged from 6.64 1036 to 6.33 1038 erg s?1Across both models, several spectral parameters exhibited step-function-like behavior around flux thresholds of 510 10?9 erg cm?2 s?1, indicating multiple spectral regimes. The disk flux contribution, more evident in model B, increased with total flux, supporting an intrinsic origin for the variability. These findings point to a complex interplay between intrinsic disk emission, structured winds, and variable local absorption in driving the flare activity. 2025. The Author(s). -
Spectral Evolution of GX 17+2 Using AstroSat and NICER Observations
We study the spectral evolution of the Z-track source GX 17+2 using AstroSat and NICER observations taken between 2016 and 2020. The AstroSat observations cover the period when the source is in the normal branch (NB) and the flaring branch (FB), while for the NICER ones the variability can be associated with the FB branch. The source spectra at different regions of the branches are well described by accretion disk emission, blackbody surface emission, and a thermal Comptonization component. In the NB, the total bolometric unabsorbed flux remains constant and the variation is due to changes in the Comptonization, disk fluxes. In particular, the inferred luminosity (LT) and accretion rate ( M ? ) remain constant, while there is significant variation in the inner disk radii and fraction of disk photons entering the corona, indicating changes in the geometry of the system. On the other hand, in the FB, there is significant variation in luminosity from ?4.0 to ?7.0 1038 erg s?1. Despite this significant variation in luminosity and in the inner disk radii, the accretion efficiency, defined as ? = L T / M ? c 2 , remains nearly constant at ?0.20 throughout the evolution of the source, as expected for a neutron star system. 2025. The Author(s). Published by the American Astronomical Society. -
First Polarimetric View of GX 349+2 with the Imaging X-Ray Polarimetry Explorer
We conducted a spectropolarimetric study of the bright Z source GX 349+2 using the Imaging X-ray Polarimetry Explorer (IXPE) observation. Our findings reveal a significant polarization degree (PD) of 1.1% 0.3% in the 2.0-8.0 keV energy range. Spectropolarimetric analysis was performed by modeling the source spectra with an absorbed multicolor disk component and a blackbody. This allowed us to constrain the polarization contributions from the disk and boundary/spreading layer. The results indicate that the observed polarization signal primarily originates from the disk and the spreading layer at the neutron stars surface, rather than the boundary layer. Additionally, we detect an excess polarization component, which we attribute to either an outflow or reflection processes within the system, indicating the presence of a third component, albeit not observed in the IXPE spectra. Furthermore, energy-resolved polarization analysis in the 2.0-4.0 and 4.0-8.0 keV energy ranges hinted at a marginal increase of PD with energy and rotation of polarization angle (PA). This also pointed to an energy-dependent dominance of emission and indicated that the variation in PA with energy (?17? in the 2.0-4.0 keV energy range and ?48? in the 4.0-8.0 keV energy range) is likely associated with the different nonorthogonal PAs of the disk and spreading layer components, which peak at different energies. 2025. The Author(s). Published by the American Astronomical Society. -
A Novel Survey for Young Substellar Objects with the W-band Filter. VII. Water-bearing Objects in the Core of the ? Ophiuchi Cloud Complex
We present a study of very low mass stars and brown dwarfs in the rich star-forming core of the ? Ophiuchi cloud complex. The selection of the sample relies on detecting the inherent water absorption characteristic in young substellar objects. Of the 22 water-bearing candidates selected, 15 have a spectral type of M6 or later. Brown dwarf candidates too faint for membership determination by Gaia have their proper motions derived by deep-infrared images spanning 6 yr. Astrometric analysis confirms 21/22 sources as members, with one identified as a contaminant. Infrared colors and the spectral energy distribution of each water-bearing candidate are used to diagnose the mass, age, and possible existence of circumstellar dust. A total of 15 sources exhibit evidence of disks in their spectral energy distributions, as late as in M8-type objects. Spectroscopy for bright candidates has confirmed one as an M8 member and verified two sources (with disks) exhibiting signatures of magnetospheric accretion. 2025. The Author(s). Published by the American Astronomical Society. -
Tracing Early Enrichment Pathways: Chemical and Chemodynamical Analysis of Two CEMP-no Stars**Based on the data collected from HDS/SUBARU
We present a detailed high-resolution spectroscopic and chemodynamic analysis of two carbon stars, HE 1148?0037 and HE 1246?1344, using SUBARU High Dispersion Spectrograph spectra (R ? 50,000). Our analysis confirms that both stars are extremely metal-poor and belong to the Carbon-Enhanced Metal-Poor (CEMP)-no class ([C/Fe] > 0.70 and [Ba/Fe] < ?1.0). Both of the stars are found to be group II CEMP-no stars in the A(C) versus [Fe/H] diagram. While both stars exhibit slightly enhanced ?-elements (Ca, Sc), their Mg, Fe-peak, and neutron-capture element abundances show distinct trends, indicating different progenitor pathways. From [O/Fe] and [Sr/Ba] ratios, we identify both metal-poor asymptotic giant branch stars and fast-rotating massive stars as potential progenitors. Further elemental abundance analysis suggests a multienrichment origin for these stars. Kinematic analysis reveals that HE 1148?0037 belongs to the halo, with a spatial velocity of 464.99 14 km s?1. Orbital calculations show that HE 1148?0037 follows a retrograde orbit and is associated with the high-energy retrograde halo, likely a member of the Iitoi substructure. In contrast, HE 1246?1344 follows a prograde orbit and is linked to Gaia-Sausage/Enceladus. 2026. The Author(s). Published by the American Astronomical Society. Original content from this work may be used under the terms of the https://creativecommons.org/licenses/by/4.0/. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI. -
Neurofeedback Therapy Meets Transformers: Rewiring Sleep Disorders Through AI-Driven EEG Modulation
Sleep disorders such as insomnia, sleep Apnea, and hypersomnia significantly impair neurophysiological functioning, yet conventional treatments like Cognitive Behavioral Therapy for Insomnia (CBT-I) remain resource-intensive and difficult to personalize. This study introduces a novel AI-powered neurofeedback simulation framework designed to detect dysregulated EEG frequency band activity across sleep stages and simulate targeted interventions. A Transformer-based model serves as the core component, offering a unique capability to model cross-epoch temporal dynamics and frequency-specific spectral patterns. Unlike traditional architectures that treat EEG epochs in isolation, the Transformer captures how EEG band activity evolves across the night, critical for identifying persistent dysregulation patterns and planning stage-specific interventions. Through its multi-head attention mechanism, the model can simultaneously monitor delta, theta, alpha, beta, and gamma fluctuations while preserving sleep architecture transitions using positional encoding. Dysregulated epochs are classified with 92% accuracy, and intervention simulations-such as beta suppression in N2 or delta enhancement in REM-led to measurable improvements: average WASO decreased by 23%, and Sleep Efficiency improved by 13%. This framework not only demonstrates the efficacy of Transformer-based temporal-spectral modelling in EEG but also lays the foundation for closed-loop, wearable-compatible, personalized neurofeedback systems for remote sleep therapy. 2026 A l A KA d V idh hi V -
IoT-Integrated CNN Deep Learning for Automated Breast Cancer Detection and Diagnosis
Breast cancer continues to be a primary cause of death in women, requiring prompt and accurate diagnosis to enhance treatment results. Traditional diagnostic techniques depend on manual assessment, which leads to possible misclassification, significant inter-observer variability, and delays in decision-making. Current deep learning models, including CNNs, frequently experience feature loss, gradient declining and restricted adaptability to real-time data. To overcome these restrictions, we present a hybrid framework combining CNN and ResNet that merges deep learning-based feature extraction with real-time data collecting from IoT devices. The proposed approach utilises CNNs for preliminary feature extraction, ResNet for hierarchical learning with residual connections, and IoT for real-time patient monitoring and automatic notifications. The dataset undergoes preprocessing through normalisation, augmentation, and histogram equalisation to improve image quality and learning efficacy. The model is trained with cross-entropy loss and the Adam optimiser, guaranteeing stability and excellent performance. The evaluation results indicate a substantial enhancement compared to baseline models, with an accuracy of 97, an F1-score of 95.3, and a recall rate of 96.4%, exceeding traditional deep learning (90 accuracy) and CNN-based models (80% accuracy). The suggested model similarly minimises mistakes, with RMSE and MSE values declining to 1.2 and 1.6, respectively, signifying reduced misclassification rates. The inclusion of IoT facilitates instantaneous data transmission with little latency, hence improving clinical decision-making and minimising diagnostic delays. This advanced system facilitates automated and precise breast cancer detection, providing an innovative method for early diagnosis, optimised treatment planning, and improved patient outcomes, while ensuring data privacy and security through encryption and commitment to healthcare regulations. 2026 Yamini Kalva, R. Ganesh Babu, Sindhu V, S. Gokul Pran, Garaga Srilakshmi, Kavitha C T, Sathish Kumar Shanmugam and V. Bhoopathy. This open-access article is distributed under a Creative Commons Attribution (CC-BY) 4.0 license. -
Optimized Feature Selection for Kidney Ultrasound Image Classification Using Binary Coati Weighted Mean Vector Algorithm
The analysis of medical images presents many challenges, especially when making precise diagnoses. In pediatric Chronic Kidney Disease (CKD), early identification is critical because of its gradual progression to significant kidney failure. This study proposes a diagnostic framework for pediatric ultrasound image classification that incorporated machine learning and advanced feature selection methods. This approach is divided into four stages: Preprocessing, feature extraction, feature selection, and classification. Initially, pediatric kidney ultrasound images are enhanced using gaussian median filter. Radiomics features were then extracted, including Gray Level Co-Occurrence Matrix (GLCM), Gray Level Size Zone Matrix (GLSZM), Gray Level Run Length Matrix (GLRLM), Neighboring Gray Tone Difference Matrix (NGTDM), Gray Level Dependence Matrix (GLDM), and first-order statistics. To optimize this feature space, we introduce the Binary Coati Weighted Mean Vector (BinCoWmv) optimization algorithm, which uses a customized fitness function. Herein, the selected features were evaluated using different classifiers: Random Forest (RF), Decision Tree (DT), Support Vector Machine (SVM), Nae Bayes (NB), K-Nearest Neighbor (KNN), and XG-Boost. Comparative evaluations with existing optimizers, such as the Coati Optimization Algorithm (COA), weighted average vector (INFO), Firefly Algorithm (FFA), and Harris Hawk Optimization (HHO), showed that BinCoWmv achieved a higher classification accuracy. Our framework improves diagnostic reliability and assists radiologist and nephrologist in the early detection of chronic kidney disease in children. 2025 Fizhan Kausar and Ramamurthy B. -
Exploring Applications, Datasets, Algorithms, and Technologies in Satellite Image Processing
Amidst an era filled with complex local and global problems, satellite data presents itself as a revolutionary tool with unmatched potential to tackle practical problems in a variety of fields. This article investigates how satellite imagery, which is available through open data programs and repositories, is a valuable tool for applications including wildlife conservation, urban planning, precision agriculture, and disaster management. It highlights the unique perspective that satellite data offers. Various sources for data acquisition, the applications that are suitable for a chosen satellite data and commonly used algorithms and techniques are discussed. Through case studies, the paper demonstrates how quick and reliable data provided by satellites can be used to solve complex real-world problems. The benefits of satellite data are emphasized, including its affordability, ability to monitor in real-time, and ability to support sustainable behaviours and policy-making. The study explores cutting-edge technologies, highlighting cloud computing and GIS integration as well as machine learning algorithms to build robust solutions using satellite data. The immense potential of satellite data is accompanied by challenges, including data integration, computational complexity, and ethical considerations. These challenges underscore the need for standardization and continuous efforts to fully realize the potential of satellite data in sustainable development and informed decision-making. 2025 Bijeesh TV, Bejoy BJ, Michael Moses Thiruthuvanthan and Raju G. -
Towards a Human-Like AGI Architecture: GeneralIntelligence Framework (GIF)
Artificial Intelligence (AI) has achieved significant breakthroughsbut remains limited by its specialization and inability to generalize acrossdomains, unlike human cognition. Current models such as Large LanguageModels (LLMs) and Multimodal Large Language Models (MLLMs) excel atspecific tasks but struggle with real-time adaptability and cross-domaingeneralization. This paper introduces the General Intelligence Framework(GIF), an approach designed to bridge this gap by mimicking human-likecognitive processes. By integrating Deep Learning (DL), Spiking NeuralNetworks (SNNs), and neuromorphic hardware, the framework fosters Real-Time Learning (RTL) and adaptability. The proposed framework holdspotential for industries like robotics, healthcare, education, astronomy,defense, autonomous systems, etc., where flexible, adaptive AI is critical.We hypothesize that the framework will enable AI systems to handleunforeseen inputs and tasks without requiring extensive retraining,representing a step toward achieving Artificial General Intelligence (AGI). 2025 Jiran Kurian Puliyanmakkal and Rohini V. This open-access article is distributed under a Creative Commons Attribution (CC-BY) 4.0 license.