Biomedical and Health Sciences
In structure-based drug design (SBDD), a major challenge is generating high-affinity 3D ligand molecules that can effectively bind to specific protein targets, which requires accurately capturing complex protein-ligand interactions. Although existing diffusion models have demonstrated potential in molecular generation tasks, they often struggle with accurately capturing the complex interactions between proteins and ligands. To address this problem, we propose MSIDiff, a multi-stage interaction-aware diffusion model for protein-specific molecular generation.
- Categories:
196 Views
Osteoarthritis (OA) is a prevalent degenerative joint disease,particularly affecting the knees. Early and accurate detection of OA and its severity, often graded using the Kellgren-Lawrence (KL) scale, is crucial for timely intervention and management. This study explores the application of deep learning techniques to automatically detect OA and assign KL grades from knee X-ray images. We propose a novel deep learning architecture that effectively extracts relevant features from X-ray images and classifies them into different KL grades.
- Categories:
424 Views
This dataset comprises a comprehensive collection of PubMed abstracts and associated metadata focusing on the topic of multiple sclerosis (MS) in relation to social determinants and environmental factors, spanning publications from January 1, 2018, to November 15, 2024.
- Categories:
69 Views
C-SRR Radiation Patterns. The pixels from the radiation pattern generated by positioning the C-SRR over the phantom model with and without cancerous tissue were extracted using various window shapes and sizes to form the dataset. For this, pixel sampling operations such as average, minimum, maximum, and median are performed. Pixel reconfiguration to triangle, square, symmetric and asymmetric is also performed. In average pixel sampling, the average value of the pixel color is divided by the total number of pixels.
- Categories:
144 Views
Abstract— Pluripotent cell types retain several characteristics that make them optimal cell source material for applications in drug development, disease modeling, and therapeutic applications. Human induced pluripotent stem cells (hiPSCs) are currently the most accessible cell source material to cultivate and derive cell-based therapeutic solutions at scale. However, a disconnect exists between quality characteristics of phenotype in the pluripotent state, and downstream metrics for efficacy.
- Categories:
91 Views
We develope a novel TCM hallucination detection dataset, Hallu-TCM, sine no prior work has attempted this task in TM. We selected 1,260 TCM exam questions including 16 TCM subjects, input them into GPT-4, and collected their feedback. In the first level, we utilize Qwen-Max interface to annotate feedback multiple times with the binary label. If Qwen-Max consistently provided the same label across annotations, we adopted that label. For contentious cases, we recruited higher-degree research students who can understand and solve complex questions, including three Ph.D.
- Categories:
184 Views
This video demonstrates the real-time data acquisition and noise reduction capabilities of a CMOS capacitive sensor array (CSA) implemented on an FPGA. The system captures the evaporation process of a deionized water droplet placed on the sensor array, using multiple sampling (MS) and pixel-wise accumulation (PWA) techniques to enhance signal quality and reduce random noise. The system efficiently processes and transmits the data, showcasing the gradual reduction in the droplet's size as it evaporates.
- Categories:
112 Views
SynGen6 is a synthetic genomic dataset that encompasses six distinct populations. We utilized Principal Component Analysis (PCA) and ϵ-local differential privacy (LDP) to generate synthetic samples. We then simulated phenotype vectors associated with significant SNPs, mirroring real-world gene-disease associations. We also generated synthetic SNPs to watermark the dataset enabling verification of outsourced computations. Lastly, synthetic relatives were created to support research on kinship inference and family-based genomic analyses.
- Categories:
81 Views
To download this dataset without purchasing an IEEE Dataport subscription, please visit: https://zenodo.org/records/13896353
Please cite the following paper when using this dataset:
- Categories:
932 Views
This dataset webpage contains datasets and models
Existing_BLDC1.zip
presented in my 3rd last Speaker Presentation in Conference - 2025*.
- Categories:
280 Views