Brain Cancer - MRI Dataset

This dataset contains a comprehensive collection of MRI images for brain cancer research, specifically aimed at supporting medical diagnostics.

Categories

• Medical Education
• Brain Cancer
• Machine Learning
• Image Classification
• Brain Areas
• Deep Learning

Contributor

Md Mizanur Rahman

Publication Date

August 5, 2024

DOI

10.17632/mk56jw9rns.1

Description

The Bangladesh Brain Cancer MRI Dataset is a comprehensive collection of MRI images categorized into three distinct classes:

• Brain_Glioma: 2004 images
• Brain_Menin: 2004 images
• Brain Tumor: 2048 images

The dataset includes a total of 6056 images, uniformly resized to 512x512 pixels. These images were collected from various hospitals across Bangladesh with the direct involvement of experienced medical professionals to ensure accuracy and relevance. This dataset is valuable due to the difficulty in obtaining such medical imaging data and offers a reliable resource for developing and testing diagnostic tools.

Potential Applications

Researchers and practitioners can utilize this dataset for various applications, including:

• Image Processing: Enhancing and analyzing MRI images.
• Deep Learning: Training neural networks for automated classification and detection of brain cancer.
• Machine Learning: Developing predictive models for early diagnosis and treatment planning.

Citation

Rahman, Md Mizanur (2024), “Brain Cancer - MRI dataset”, Mendeley Data, V1, doi: 10.17632/mk56jw9rns.1

Please upvote if you find this dataset useful!

Brain Tumor Detection Dataset

Overview

This dataset contains high-quality MRI images of brain tumors with detailed annotations. The dataset is meticulously curated, cleaned, and annotated to aid in the development and evaluation of machine learning models for brain tumor detection and classification.

Dataset Composition

The dataset includes a total of 5,249 MRI images divided into training and validation sets. Each image is annotated with bounding boxes in YOLO format, and labels corresponding to one of the four classes of brain tumors.

Classes

• Class 0: Glioma
• Class 1: Meningioma
• Class 2: No Tumor
• Class 3: Pituitary

Data Split

1. Training Set:

• Glioma: 1,153 images
• Meningioma: 1,449 images
• No Tumor: 711 images
• Pituitary: 1,424 images

2. Validation Set:

• Glioma: 136 images
• Meningioma: 140 images
• No Tumor: 100 images
• Pituitary: 136 images

Image Characteristics

The images in the dataset are from different angles of MRI scans including sagittal, axial, and coronal views. This variety ensures comprehensive coverage of brain anatomy, enhancing the robustness of models trained on this dataset.

Annotations

The bounding boxes were manually annotated using the LabelImg tool by a dedicated team. This rigorous process ensures high accuracy and reliability of the annotations.

Source and Inspiration

This dataset was inspired by two existing datasets: 1. https://www.kaggle.com/datasets/masoudnickparvar/brain-tumor-mri-dataset 2. https://www.kaggle.com/datasets/sartajbhuvaji/brain-tumor-classification-mri

A thorough cleaning process was performed to remove noisy, mislabeled, and poor-quality images, resulting in a high-quality and well-labeled dataset.

Usage

This dataset is suitable for training and validating deep learning models for the detection and classification of brain tumors. The variety in MRI scan angles and the precision of annotations provide an excellent foundation for developing robust computer vision applications in medical imaging.

Citation

If you use this dataset in your research or project, please consider citing it appropriately to acknowledge the effort put into its creation and annotation.

This dataset contains a collection of multimodal medical images, specifically CT (Computed Tomography) and MRI (Magnetic Resonance Imaging) scans, for brain tumor detection and analysis. It is designed to assist researchers and healthcare professionals in developing AI models for the automatic detection, classification, and segmentation of brain tumors. The dataset features images from both modalities, providing comprehensive insight into the structural and functional variations in the brain associated with various types of tumors.

The dataset includes high-resolution CT and MRI images captured from multiple patients, with each image labeled with the corresponding tumor type (e.g., glioma, meningioma, etc.) and its location within the brain. This combination of CT and MRI images aims to leverage the strengths of both imaging techniques: CT scans for clear bone structure visualization and MRI for soft tissue details, enabling a more accurate analysis of brain tumors.

I collected these data from different sources and modified data for maximum accuracy.

Brain Tumor CT scan Images source

1. CT Brain Segmentation Computer Vision Project -- https://universe.roboflow.com/joshua-zgc7b/ct-brain-segmentation
2. CT and MRI brain scans -- https://www.kaggle.com/datasets/darren2020/ct-to-mri-cgan
3. CT Head Scans(jpg files) -- https://www.kaggle.com/datasets/clarksaben/ct-head-scans
4. Head CT Images for Classification -- https://www.kaggle.com/datasets/nipaanjum/head-ct-images-for-classification
5. Anonymous brain -- from private data
6. Unpaired MR-CT Brain Dataset for Unsupervised Image Translation -- https://data.mendeley.com/datasets/z4wc364g79/1

Brain Tumor MRI images source

1. Brain Tumor (MRI Scans) -- https://www.kaggle.com/datasets/rm1000/brain-tumor-mri-scans
2. Brain Tumor MRIs -- https://www.kaggle.com/datasets/vinayjayanti/brain-tumor-mris
3. Siardataset -- https://www.kaggle.com/datasets/masoumehsiar/siardataset
4. Brain tumors 256x256 -- https://www.kaggle.com/datasets/thomasdubail/brain-tumors-256x256
5. Brain Tumor MRI Image Classification Dataset -- https://www.kaggle.com/datasets/iashiqul/brain-tumor-mri-image-classification-dataset
6. Brain Tumor MRI (yes or no) -- https://www.kaggle.com/datasets/mohamada2274/brain-tumor-mri-yes-or-no
7. BRAIN TUMOR CLASS CLASS Computer Vision Project -- https://universe.roboflow.com/college-sf5ih/brain-tumor-class-class
8. Brain Tumor Detection Computer Vision Project -- https://universe.roboflow.com/tuan-nur-afrina-zahira/brain-tumor-detection-bmmqz
9. Tumor Detection Computer Vision Project -- https://universe.roboflow.com/brain-tumor-detection-wsera/tumor-detection-ko5jp

This dataset is collected from Kaggle ( https://www.kaggle.com/datasets/masoudnickparvar/brain-tumor-mri-dataset ). This dataset is a combination of the following three datasets :figshareSARTAJ datasetBr35H

There a total of 7023 MRI scans of brains split into four classes. Transfer learning and CNNs could be used for classifying these images, which could save patient lives through early tumor diagnosis.

Uncovering Knowledge: A Clean Brain Tumor Dataset for Advanced Medical Research

Introduction:

• This dataset, available in RAR archive format, consists of four classes, including three tumor classes (Pituitary, Glioma and Meningioma) and one class representing normal brain MRI scans.
• The strength of this dataset in comparison with other releases across the Kaggle is the cleanness of data. In this regard, we subjected the initial dataset to a meticulous data cleaning pipeline. This pipeline involved several steps aimed at enhancing the dataset's integrity and usability.
• The initial data source for this dataset is the brain tumor classification MRI dataset, which can be accessed at this link.

Data Cleaning Process:

• Removal of Duplicate Samples: We employed an image vector comparison method to identify and remove duplicate samples, ensuring that each data point is unique.
• Correction of Mislabeled Images: Using our domain knowledge, we carefully inspected and corrected falsely labeled images, ensuring that they were appropriately categorized. This step greatly enhances the accuracy of the dataset.
• Image Resizing: All images in the dataset were resized to a memory-efficient yet academically accepted size of (224, 224), facilitating easier processing and analysis. Statistics: *Before the cleaning pipeline, the dataset contained the following number of samples for each class from the initial data source:
• Normal: 500
• Glioma: 926
• Meningioma: 937
• Pituitary: 901

After applying the data cleaning pipeline, the number of samples in each category decreased on average by approximately 3-9%. This reduction ensures the data integrity while maintaining a sufficient number of samples for comprehensive analysis.

Data Augmentation:

To enhance the diversity and robustness of the dataset, we employed various image augmentation techniques. These techniques were applied to the images without altering the labels. Here is a summary of the augmentation methods used: - Salt and Pepper Noise: Introducing random noise by setting pixels to white or black based on a specified intensity. - Histogram Equalization: Applying histogram equalization to enhance the contrast and details in the images. - Rotation: Rotating the images clockwise or counterclockwise by a specified angle. - Brightness Adjustment: Modifying the brightness of the images by adding or subtracting intensity values. - Horizontal and Vertical Flipping: Flipping the images horizontally or vertically to create mirror images.

Use Cases and Potential Investigations:

This dataset offers significant potential for various advanced medical research and analysis applications. Some interesting use cases and potential investigations using this dataset include: - Tumor Classification: Developing advanced machine learning models for accurate and automated brain tumor classification. - Treatment Planning: Analyzing the tumor characteristics to aid in treatment planning and decision-making processes. - Radiomics Analysis: Extracting quantitative features from the images for radiomics analysis to uncover valuable insights and patterns. - Comparative Studies: Conducting comparative studies among different tumor types to understand their unique characteristics and behaviors.

Acknowledgement

• We would like to express our sincere gratitude to the original dataset publisher, sartajbhuvaji, for their valuable contribution.
• This dataset is released under the CC0 license, making it open and accessible for everyone to use. While not mandatory, citing the dataset would be greatly appreciated.

Important Note

Those researchers who want to use this dataset for real world use cases, must consult with medical field experts (radiologists, ...) on the ground truth of the labels and their usability for their angle of research.

Kaggle Brain Tumor

## Overview

Kaggle Brain Tumor is a dataset for object detection tasks - it contains Brain annotations for 3,906 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

This dataset provides synthetic yet comprehensive information about brain tumor cases, including medical imaging data, patient demographics, clinical attributes, and treatment details. It is specifically designed to support research and development in the field of medical image analysis, brain tumor detection, and classification.

By leveraging this dataset, researchers and healthcare professionals can train machine learning models, conduct statistical analysis, and enhance tumor detection and classification techniques using MRI scans and clinical data.

📌 Key Features: 🔬 Patient Demographics & Medical Data: Patient ID 🆔: A unique identifier for each patient (Anonymized). Age 🎂: The age of the patient at the time of diagnosis. Gender 🚻: The gender of the patient (Male/Female). 🧪 Tumor Details & Medical Imaging: Tumor Type 🏷️: Categorization of tumors, including: Meningioma Glioma Pituitary Tumor Tumor Location 📍: The specific area within the brain where the tumor is located. MRI Images 🖼️: High-quality MRI scans from different modalities, including: T1-weighted MRI T2-weighted MRI FLAIR (Fluid-attenuated inversion recovery) Other specialized MRI scans 📄 Clinical Observations & Treatment Plans: Clinical Notes 🏥: Important observations, symptoms, and notes recorded by healthcare professionals. Treatment Plan 💊: The approach taken for treating the tumor, which may include: Surgery Radiotherapy Chemotherapy A combination of treatments 📊 Use Cases & Applications: This dataset can be used for: ✅ Developing AI models for tumor detection & classification ✅ Analyzing patterns in tumor characteristics & treatment responses ✅ Medical image processing & segmentation research ✅ Predicting patient outcomes & optimizing treatment strategies

⚠️ Important Note: This dataset is synthetic and does not contain real patient data. It is artificially generated for educational, research, and analytical purposes, ensuring privacy and confidentiality compliance.

This brain tumor dataset contains 3064 T1-weighted contrast-inhanced images with three kinds of brain tumor. Detailed information of the dataset can be found in the readme file.The README file is updated:Add image acquisition protocolAdd MATLAB code to convert .mat file to jpg images

Brain Tumor Detection | Vision Transformer 99% Click -> Kaggle

  task_categories:

• image-classification
• image-segmentation tags:
• 'brain '
• MRI
• brain-MRI-images
• Tumor

Medical Image DataSet: Brain Tumor Detection

Medical Image Dataset: Brain Tumor Detection

The Brain Tumor MRI dataset, curated by Roboflow Universe, is a comprehensive dataset designed for the detection and classification of brain tumors using advanced computer vision techniques. It comprises 3,903 MRI images categorized into four distinct classes:

• Glioma: A tumor originating from glial cells in the brain.
• Meningioma: Tumors arising from the meninges, the protective layers surrounding the brain and spinal cord.
• Pituitary Tumor: Tumors located in the pituitary gland, affecting hormonal balance.
• No Tumor: MRI scans that do not exhibit any tumor presence.

Each image in the dataset is annotated with bounding boxes to indicate tumor locations, facilitating object detection tasks precisely. The dataset is structured into training (70%), validation (20%), and test (10%) sets, ensuring a robust framework for model development and evaluation.

The primary goal of this dataset is to aid in the early detection and diagnosis of brain tumors, contributing to improved treatment planning and patient outcomes. By offering a diverse range of annotated MRI images, this dataset enables researchers and practitioners to develop and fine-tune computer vision models with high accuracy in identifying and localizing brain tumors.

This dataset supports multiple annotation formats, including YOLOv8, YOLOv9, and YOLOv11, making it versatile and compatible with various machine-learning frameworks. Its integration with these formats ensures real-time and efficient object detection, ideal for applications requiring timely and precise results.

By leveraging this dataset, researchers and healthcare professionals can make significant strides in developing cutting-edge AI solutions for medical imaging, ultimately supporting more effective and accurate diagnoses in clinical settings.

https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F14850461%2Fe03fba81bb62e32c0b73d6535a25cb8d%2F3.jpg?generation=1734173601629363&alt=media" alt="">

This is the data set used in the paper Brain Tumor Classification (MRI), and the complete data set can be found at: https://www.kaggle.com/datasets/sartajbhuvaji/brain-tumor-classification-mri

A private collection resonance MRI images by brain tumor type.

Images without any type of marking or patient identification, interpreted by radiologists and provided for study purposes.

The images are separated by astrocytoma, carcinoma, ependymoma, ganglioglioma, germinoma, glioblastoma, granuloma, medulloblastoma, meningioma, neurocytoma, oligodendroglioma, papilloma, schwannoma and tuberculoma.

Accurate diagnosis of the brain tumor type at an earlier stage is crucial for the treatment process and helps to save the lives of a large number of people worldwide. Because they are non-invasive and spare patients from having an unpleasant biopsy, magnetic resonance imaging (MRI) scans are frequently employed to identify tumors. The manual identification of tumors is difficult and requires considerable time due to the large number of three-dimensional images that an MRI scan of one patient’s brain produces from various angles. Moreover, the variations in location, size, and shape of the brain tumor also make it challenging to detect and classify different types of tumors. Thus, computer-aided diagnostics (CAD) systems have been proposed for the detection of brain tumors. In this paper, we proposed a novel unified end-to-end deep learning model named TumorDetNet for brain tumor detection and classification. Our TumorDetNet framework employs 48 convolution layers with leaky ReLU (LReLU) and ReLU activation functions to compute the most distinctive deep feature maps. Moreover, average pooling and a dropout layer are also used to learn distinctive patterns and reduce overfitting. Finally, one fully connected and a softmax layer are employed to detect and classify the brain tumor into multiple types. We assessed the performance of our method on six standard Kaggle brain tumor MRI datasets for brain tumor detection and classification into (malignant and benign), and (glioma, pituitary, and meningioma). Our model successfully identified brain tumors with remarkable accuracy of 99.83%, classified benign and malignant brain tumors with an ideal accuracy of 100%, and meningiomas, pituitary, and gliomas tumors with an accuracy of 99.27%. These outcomes demonstrate the potency of the suggested methodology for the reliable identification and categorization of brain tumors.

A comprehensive dataset consisting of brain scans from multiple imaging modalities, including MRI, has been chosen for the training of the YOLOv7 model. The dataset has been carefully selected to provide exposure to a wide variety of clinical situations, including cases of pituitary, meningioma, glioma, and no tumor. The dataset was obtained from Kagle and Hayatabad Medical Complex Peshawar. The specifications of the dataset, consisting of 1594 healthy brain samples, 1321 glioma, 1339 meningioma, and 1457 pituitary brain tumor samples. Every class in the dataset, showcasing multiple views, including axial, coronal, and sagittal. These samples offer a thorough visual depiction of various instances within the dataset, having been meticulously pre-sorted and checked by an expert radiologist. At first, there was no appropriate label format for YOLO in the dataset. To overcome this, a precise re-labeling strategy was implemented with the help of a medical specialist, guaranteeing the creation of consistent ground truth images. The labelImg tool, a graphical image annotation tool, made this process easier. Labeling was carefully done, which is an important step in establishing factual information about the intended subject in the image. The exact measured locations of the subjects within the image, shown as

This Brain Tumor Prediction Dataset contains 250,000 patient records with 22 important medical features. The data includes MRI scan results, tumor size, genetic risks, symptoms, lifestyle habits, and treatment details. It is designed for predictive modeling, data analysis, and AI applications in healthcare.

✅ Why Use This Dataset?

Large-scale realistic medical data (250K rows) Includes tumor location, growth rate, and survival rate Useful for machine learning, deep learning, and medical research Perfect for classification and survival analysis Supports global health insights (data from multiple countries)

This dataset consolidates brain tumor MRI images from multiple Kaggle data sources to create a larger, centralised dataset for research and model development purposes.

The dataset comprises of 16,269 images containing four main classes : - Glioma (3,325 Images) - Meningioma (3,266 Images) - Pituitary (2,974 Images) - Healthy (6,704 Images)

Key Notes:

Duplicate images are likely due to dataset overlaps when sourcing. We strongly recommend users perform deduplication before training.

The dataset does not apply any cleaning, resizing, or augmentation — it's intended to be raw and inclusive for flexibility.

Recommendation:

This dataset is ideal for users who want to experiment with preprocessing, augmentation, and custom cleaning pipelines on a real-world, mixed-quality dataset. Please consult medical professionals if using this data for clinical or diagnostic applications.

File Structure

The dataset is organised as follows: - Each folder represents the 4 classes - The filenames of each image contain the original dataset source (Name based on user who published the dataset to Kaggle)

Data Sources:

This dataset combines the following five Kaggle datasets:

1. Brain Tumors Dataset (Excluded their augmented images) by Seyed Mohammad Hossein Hashemi
2. PMRAM Bangladeshi Brain Cancer MRI Dataset by Orville
3. Brain Tumor MRI Images (17 Classes) by Fernando Feltrin (Only T1 glioma/meningioma/healthy images used).
4. SIAR Dataset by Masoumeh Siar (Only healthy scans used as this was a binary dataset, and did not differentiate the tumor types).
5. Brain Tumor MRI Scans by Rajarshi Mandal

These datasets were selected for their popularity, quality, and complementary class coverage. We recommend checking the original sources for more information about data collection methods and original licensing.

License

This combined dataset is released under CC BY-SA 4.0 to comply with ShareAlike requirements of source datasets:

Brain tumors are a serious health challenge worldwide, and early detection plays a key role in effective treatment. MRI (Magnetic Resonance Imaging) is one of the most widely used techniques for diagnosing brain tumors because it provides detailed images of soft tissues in the brain.

This dataset contains MRI images categorized into two classes: 1. (yes) for images with brain tumors 2. (no) for images without brain tumors

The dataset is designed for binary classification tasks and can be used to train and evaluate machine learning or deep learning models, particularly Convolutional Neural Networks (CNNs).

Images have been collected from publicly available medical imaging repositories and open research publications that allow educational and research use. The dataset has been organized into separate folders to make it easier for beginners and researchers to use directly in model training.

The main inspiration behind this dataset is to provide a beginner-friendly, ready-to-use resource for students, researchers, and practitioners who want to apply computer vision techniques in the medical imaging domain. Projects can range from simple CNN classification to more advanced applications like transfer learning and model interpretability such as Grad-CAM visualization.

A brain tumor detection dataset consists of medical images from MRI or CT scans, containing information about brain tumor presence, location, and characteristics. This dataset is essential for training computer vision algorithms to automate brain tumor identification, aiding in early diagnosis and treatment planning.

Dataset Structure

The brain tumor dataset is divided into two subsets:

• Training set: Consisting of 893 images, each accompanied by corresponding annotations.
• Testing set: Comprising 223 images, with annotations paired for each one.

Applications

The application of brain tumor detection using computer vision enables early diagnosis, treatment planning, and monitoring of tumor progression. By analyzing medical imaging data like MRI or CT scans, computer vision systems assist in accurately identifying brain tumors, aiding in timely medical intervention and personalized treatment strategies.

Usage

Train

To train a YOLO11n model on the brain tumor dataset for 100 epochs with an image size of 640, utilize the provided code snippets. For a detailed list of available arguments, consult the model's Training page.

from ultralytics import YOLO

# Load a model
model = YOLO("yolo11n.pt") # load a pretrained model (recommended for training)

# Train the model
results = model.train(data="brain-tumor.yaml", epochs=100, imgsz=640)

Predict

from ultralytics import YOLO

# Load a model
model = YOLO("path/to/best.pt") # load a brain-tumor fine-tuned model

# Inference using the model
results = model.predict("https://ultralytics.com/assets/brain-tumor-sample.jpg")

Learn more ➡️ https://docs.ultralytics.com/datasets/detect/brain-tumor/

Brain Tumor Dataset

This dataset contains simulated data for brain tumor diagnosis, treatment, and patient details. It consists of 20 columns and 20,000 rows, providing information such as patient demographics, tumor characteristics, symptoms, treatment details, and follow-up requirements. The dataset is designed for machine learning projects focused on predicting the type and severity of brain tumors, as well as understanding various treatment methods and patient outcomes.

Columns:

1. Patient_ID: Unique identifier for each patient.
2. Age: Age of the patient (in years).
3. Gender: Gender of the patient (Male/Female).
4. Tumor_Type: Type of tumor (Benign/Malignant).
5. Tumor_Size: Size of the tumor in centimeters.
6. Location: The part of the brain where the tumor is located (e.g., Frontal, Temporal).
7. Histology: The histological type of the tumor (e.g., Astrocytoma, Glioblastoma).
8. Stage: The stage of the tumor (I, II, III, IV).
9. Symptom_1: The first symptom observed (e.g., Headache, Seizures).
10. Symptom_2: The second symptom observed.
11. Symptom_3: The third symptom observed.
12. Radiation_Treatment: Whether radiation treatment was administered (Yes/No).
13. Surgery_Performed: Whether surgery was performed (Yes/No).
14. Chemotherapy: Whether chemotherapy was administered (Yes/No).
15. Survival_Rate: The estimated survival rate of the patient (percentage).
16. Tumor_Growth_Rate: The growth rate of the tumor (cm per month).
17. Family_History: Whether the patient has a family history of brain tumors (Yes/No).
18. MRI_Result: The result of the MRI scan (Positive/Negative).
19. Follow_Up_Required: Whether follow-up is required (Yes/No).
20. Treatment_Response: The response to the treatment (Improved/Worsened/Stable).

Intended Use:

This dataset can be used for various machine learning tasks, such as: - Tumor classification: Predicting whether a tumor is benign or malignant. - Survival analysis: Estimating the survival rate based on different features like tumor type and treatment. - Outcome prediction: Predicting the treatment response or follow-up requirement.

Note: The data in this dataset is synthetic and does not represent real patient information. It is intended for educational and research purposes in the field of healthcare analytics and machine learning.