Mendeley Data Showcase

## Dataset Information Dataset Title: Lower Limb and Feet Wound Image Dataset for Medical Analysis Version: 2 DOI: [10.17632/hsj38fwnvr.2](https://www.google.com/search?q=https://doi.org/10.17632/hsj38fwnvr.2) Publication Date: 11 December 2025 License: [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/) Contributor: Md Masudul Islam Associated Institutions: Bangladesh University of Business and Technology (BUBT) International Institute of Information Technology Hyderabad ## Description This dataset contains a collection of medical images focused on lower limb and feet conditions. It includes raw wound images, corresponding masked images (for segmentation tasks), and a control group of normal (healthy) feet images from both male and female participants. The dataset is designed to support medical image analysis, specifically for tasks such as wound detection, classification, and segmentation. ### Key Statistics Total Images: 8,129 Image Format: JPG Resolution: 331 x 331 pixels ## Dataset Structure The dataset is categorized into three main components: root_directory/ │ ├── Normal/ │ ├── [filename].jpg # Healthy feet images (Male & Female) │ └── ... # Total: 2,575 images │ ├── Wound_Main/ │ ├── [filename].jpg # Raw wound images │ └── ... # Total: 2,686 images │ └── Wound_Masked/ ├── [filename].jpg # Binary segmentation masks └── ... # Total: 2,686 images 1. Normal (Healthy) Images (2,575 images) Contains images of healthy feet (both left and right) from male and female participants. Male: 1,981 raw images (from 991 samples) Female: 776 raw images (from 388 samples) Source: Collected at Bangladesh University of Business and Technology (BUBT). 2. Wound Main (Raw) Images (2,686 images) Raw images of various lower limb and foot wounds. Source: Authors' collections (refer to the IEEE article below). 3. Wound Masked Images (2,686 images) Binary masks corresponding to the "Wound Main" images, suitable for training segmentation models. Source: Authors' collections. ## Methodology & Sources The data was aggregated from two primary sources: 1. Clinical Collection: 2,686 wound images and their corresponding masks were derived from the authors' previous research collections. 2. Institutional Collection: Healthy control samples were collected at the Bangladesh University of Business and Technology (BUBT), covering a demographic of 991 males and 388 females. ## Recommended Usage Medical Image Processing: Pre-processing and enhancement techniques for wound imagery. Image Classification: Distinguishing between healthy feet and those with wounds. Wound Segmentation: Using the raw and masked image pairs to train deep learning models (e.g., U-Net) to automatically identify wound boundaries.

Matrix of data taken from articles filtered according to PRISMA on agrivoltaic systems

Data was collected from 350 respondents using online survey. The survey did not collect an personal information or any kind of identifiable data point.

here is one raw data file (Shared leadership-1150211.sav) and nine processed files (including Output-1150211.spv and the relevant SEM files). All constructs in the manuscript correspond to the data files.

Entire data set

Sentences in Uzbek contain annotations of named entities. Each word is annotated with the BIOES tag.

Python code and result of A Cross-Linguistic Semantic Study with Large Language Models: Translating Core Concepts and Historical Names in The Analects

Research data on suction bucket foundation, porosity ratio generated through subroutines, and suction response of suction bucket foundation at different aspect ratios

This dataset consists of questionnaire survey data related to urban roaming and community urban tourism, used for structural equation modeling (SEM) analysis to examine the relationships among place attachment, physical environment, in-person experience sharing, electronic word-of-mouth, and travel intention. It contains three files: Raw questionnaire data in SPSS format. Data standardized for model analysis. AMOS model file (.amw) used for path analysis and hypothesis testing. Data were collected through formal questionnaires. Variables include place attachment (place identity, place dependence), physical environment, in-person experience sharing, electronic word-of-mouth, and travel intention. Data standardization was performed to facilitate normality tests and model fitting. The data can be directly imported into SPSS and AMOS for result replication, statistical analysis, or related tourism behavior studies.