Medical Image Segmentation

The Digital Retinal Images for Vessel Extraction (DRIVE) dataset is a dataset for retinal vessel segmentation. It consists of a total of JPEG 40 color fundus images; including 7 abnormal pathology cases. The images were obtained from a diabetic retinopathy screening program in the Netherlands. The images were acquired using Canon CR5 non-mydriatic 3CCD camera with FOV equals to 45 degrees. Each image resolution is 584*565 pixels with eight bits per color channel (3 channels). The set of 40 images was equally divided into 20 images for the training set and 20 images for the testing set. Inside both sets, for each image, there is circular field of view (FOV) mask of diameter that is approximately 540 pixels. Inside training set, for each image, one manual segmentation by an ophthalmological expert has been applied. Inside testing set, for each image, two manual segmentations have been applied by two different observers, where the first observer segmentation is accepted as the ground-truth for performance evaluation. Source: Ant Colony based Feature Selection Heuristics for Retinal Vessel Segmentation Image Source: https://drive.grand-challenge.org/

The LIDC-IDRI dataset contains lesion annotations from four experienced thoracic radiologists. LIDC-IDRI contains 1,018 low-dose lung CTs from 1010 lung patients. Source: A 3D Probabilistic Deep Learning System for Detection and Diagnosis of Lung Cancer Using Low-Dose CT Scans Image Source: https://thesai.org/Publications/ViewPaper?Volume=8&Issue=10&Code=IJACSA&SerialNo=15

HAM10000 is a dataset of 10000 training images for detecting pigmented skin lesions. The authors collected dermatoscopic images from different populations, acquired and stored by different modalities. Source: https://www.kaggle.com/kmader/skin-cancer-mnist-ham10000 Image Source: https://www.kaggle.com/kmader/skin-cancer-mnist-ham10000

Kvasir-SEG is an open-access dataset of gastrointestinal polyp images and corresponding segmentation masks, manually annotated by a medical doctor and then verified by an experienced gastroenterologist. Source: Kvasir-SEG: A Segmented Polyp Dataset Image Source: https://datasets.simula.no/kvasir-seg/

The STARE (Structured Analysis of the Retina) dataset is a dataset for retinal vessel segmentation. It contains 20 equal-sized (700×605) color fundus images. For each image, two groups of annotations are provided.. Source: DPN: Detail-Preserving Network with High Resolution Representation for Efficient Segmentation of Retinal Vessels Image Source: https://www.researchgate.net/figure/Results-of-the-different-methods-applied-to-the-STARE-dataset-a-original-image-b_fig4_279215756

The LUNA challenges provide datasets for automatic nodule detection algorithms using the largest publicly available reference database of chest CT scans, the LIDC-IDRI data set. In LUNA16, participants develop their algorithm and upload their predictions on 888 CT scans in one of the two tracks: 1) the complete nodule detection track where a complete CAD system should be developed, or 2) the false positive reduction track where a provided set of nodule candidates should be classified. Source: Validation, comparison, and combination of algorithms for automatic detection of pulmonary nodules in computed tomography images: the LUNA16 challenge

The KVASIR Dataset was released as part of the medical multimedia challenge presented by MediaEval. It is based on images obtained from the GI tract via an endoscopy procedure. The dataset is composed of images that are annotated and verified by medical doctors, and captures 8 different classes. The classes are based on three anatomical landmarks (z-line, pylorus, cecum), three pathological findings (esophagitis, polyps, ulcerative colitis) and two other classes (dyed and lifted polyps, dyed resection margins) related to the polyp removal process. Overall, the dataset contains 8,000 endoscopic images, with 1,000 image examples per class. Source: Two-Stream Deep Feature Modelling for Automated Video Endoscopy Data Analysis Image Source: https://datasets.simula.no/kvasir/

The dataset used in this challenge consists of 165 images derived from 16 H&E stained histological sections of stage T3 or T42 colorectal adenocarcinoma. Each section belongs to a different patient, and sections were processed in the laboratory on different occasions. Thus, the dataset exhibits high inter-subject variability in both stain distribution and tissue architecture. The digitization of these histological sections into whole-slide images (WSIs) was accomplished using a Zeiss MIRAX MIDI Slide Scanner with a pixel resolution of 0.465µm. Source: Sirinukunwattana et al. Image source: Sirinukunwattana et al.

Despite the considerable progress in automatic abdominal multi-organ segmentation from CT/MRI scans in recent years, a comprehensive evaluation of the models' capabilities is hampered by the lack of a large-scale benchmark from diverse clinical scenarios. Constraint by the high cost of collecting and labeling 3D medical data, most of the deep learning models to date are driven by datasets with a limited number of organs of interest or samples, which still limits the power of modern deep models and makes it difficult to provide a fully comprehensive and fair estimate of various methods. To mitigate the limitations, we present AMOS, a large-scale, diverse, clinical dataset for abdominal organ segmentation. AMOS provides 500 CT and 100 MRI scans collected from multi-center, multi-vendor, multi-modality, multi-phase, multi-disease patients, each with voxel-level annotations of 15 abdominal organs, providing challenging examples and test-bed for studying robust segmentation algorithms under diverse targets and scenarios. We further benchmark several state-of-the-art medical segmentation models to evaluate the status of the existing methods on this new challenging dataset. We have made our datasets, benchmark servers, and baselines publicly available, and hope to inspire future research. Information can be found at https://amos22. grand-challenge. org.

The Medical Segmentation Decathlon is a collection of medical image segmentation datasets. It contains a total of 2,633 three-dimensional images collected across multiple anatomies of interest, multiple modalities and multiple sources. Specifically, it contains data for the following body organs or parts: Brain, Heart, Liver, Hippocampus, Prostate, Lung, Pancreas, Hepatic Vessel, Spleen and Colon. Source: A large annotated medical image dataset for the development and evaluation of segmentation algorithms Image Source: Simpson et al

This project aims to provide all the materials to the community to resolve the problem of echocardiographic image segmentation and volume estimation from 2D ultrasound sequences (both two and four-chamber views). To this aim, the following solutions were set up. 1. Introduction of the largest publicly-available and fully-annotated dataset for 2D echocardiographic assessment (to our knowledge). The CAMUS dataset, containing 2D apical four-chamber and two-chamber view sequences acquired from 500 patients, is made available for download. 2. Deployment of a dedicated Girder online platform. This platform aims to assess in a reproducible manner the performance of methods for segmenting cardiac structures (left ventricle endocardium and epicardium and left atrium borders) and extracting clinical indices (left ventricle volumes and ejection fraction). The CAMUS online platform is now available and will be maintained and kept open as long as the data remains relevant for clinical research.

The PROMISE12 dataset was made available for the MICCAI 2012 prostate segmentation challenge. Magnetic Resonance (MR) images (T2-weighted) of 50 patients with various diseases were acquired at different locations with several MRI vendors and scanning protocols. Source: Constrained Deep Networks: Lagrangian Optimization via Log-Barrier Extensions Image Source: https://promise12.grand-challenge.org/

The BRATS2017 dataset. It contains 285 brain tumor MRI scans, with four MRI modalities as T1, T1ce, T2, and Flair for each scan. The dataset also provides full masks for brain tumors, with labels for ED, ET, NET/NCR. The segmentation evaluation is based on three tasks: WT, TC and ET segmentation. Source: Scribble-based Hierarchical Weakly Supervised Learning for Brain Tumor Segmentation Image Source: https://www.google.com/search?q=A+Modified+U-Net+Convolutional+Network+Featuring+a+Nearest-neighbor+Re-sampling-based+Elastic-Transformation+for+Brain+Tissue+Characterization+and+Segmentation&oq=A+Modified+U-Net+Convolutional+Network+Featuring+a+Nearest-neighbor+Re-sampling-based+Elastic-Transformation+for+Brain+Tissue+Characterization+and+Segmentation&aqs=chrome..69i57j69i64l3.296j0j4&sourceid=chrome&ie=UTF-8

The BraTS 2015 dataset is a dataset for brain tumor image segmentation. It consists of 220 high grade gliomas (HGG) and 54 low grade gliomas (LGG) MRIs. The four MRI modalities are T1, T1c, T2, and T2FLAIR. Segmented “ground truth” is provide about four intra-tumoral classes, viz. edema, enhancing tumor, non-enhancing tumor, and necrosis. Source: Brain MRI Tumor Segmentation with Adversarial Networks Image Source: https://sites.google.com/site/braintumorsegmentation/home/brats2015

The colorectal nuclear segmentation and phenotypes (CoNSeP) dataset consists of 41 H&E stained image tiles, each of size 1,000×1,000 pixels at 40× objective magnification. The images were extracted from 16 colorectal adenocarcinoma (CRA) WSIs, each belonging to an individual patient, and scanned with an Omnyx VL120 scanner within the department of pathology at University Hospitals Coventry and Warwickshire, UK. Source: HoVer-Net: Simultaneous Segmentation and Classification of Nuclei in Multi-Tissue Histology Images Image Source: Graham et al

PanNuke is a semi automatically generated nuclei instance segmentation and classification dataset with exhaustive nuclei labels across 19 different tissue types. The dataset consists of 481 visual fields, of which 312 are randomly sampled from more than 20K whole slide images at different magnifications, from multiple data sources. In total the dataset contains 205,343 labeled nuclei, each with an instance segmentation mask. Source: PanNuke Dataset Extension, Insights and Baselines Image Source: https://jgamper.github.io/PanNukeDataset/

CHASE_DB1 is a dataset for retinal vessel segmentation which contains 28 color retina images with the size of 999×960 pixels which are collected from both left and right eyes of 14 school children. Each image is annotated by two independent human experts. Source: MixModule: Mixed CNN Kernel Module for Medical Image Segmentation Image Source: https://www.mdpi.com/2073-8994/9/11/276

CVC-ClinicDB is an open-access dataset of 612 images with a resolution of 384×288 from 31 colonoscopy sequences.It is used for medical image segmentation, in particular polyp detection in colonoscopy videos. Source: ResUNet++: An Advanced Architecture for Medical Image Segmentation Image Source: https://polyp.grand-challenge.org/CVCClinicDB/

The HRF dataset is a dataset for retinal vessel segmentation which comprises 45 images and is organized as 15 subsets. Each subset contains one healthy fundus image, one image of patient with diabetic retinopathy and one glaucoma image. The image sizes are 3,304 x 2,336, with a training/testing image split of 22/23. Source: Connection Sensitive Attention U-NET for Accurate Retinal Vessel Segmentation Image Source: https://www.researchgate.net/figure/Examples-of-fundus-images-from-HRF-database-with-corresponding-hand-labelled-gold_fig1_260625531

The goal of the Automated Cardiac Diagnosis Challenge (ACDC) challenge is to: - compare the performance of automatic methods on the segmentation of the left ventricular endocardium and epicardium as the right ventricular endocardium for both end diastolic and end systolic phase instances; - compare the performance of automatic methods for the classification of the examinations in five classes (normal case, heart failure with infarction, dilated cardiomyopathy, hypertrophic cardiomyopathy, abnormal right ventricle). The overall ACDC dataset was created from real clinical exams acquired at the University Hospital of Dijon. Acquired data were fully anonymized and handled within the regulations set by the local ethical committee of the Hospital of Dijon (France). Our dataset covers several well-defined pathologies with enough cases to (1) properly train machine learning methods and (2) clearly assess the variations of the main physiological parameters obtained from cine-MRI (in particular diastolic volume and ejection fraction). The dataset is composed of 150 exams (all from different patients) divided into 5 evenly distributed subgroups (4 pathological plus 1 healthy subject groups) as described below. Furthermore, each patient comes with the following additional information : weight, height, as well as the diastolic and systolic phase instants. The database is made available to participants through two datasets from the dedicated online evaluation website after a personal registration: i) a training dataset of 100 patients along with the corresponding manual references based on the analysis of one clinical expert; ii) a testing dataset composed of 50 new patients, without manual annotations but with the patient information given above. The raw input images are provided through the Nifti format. Source: Automated Cardiac Diagnosis Challenge Image source: Automated Cardiac Diagnosis Challenge

Lesion segmentation data includes the original image, paired with the expert manual tracing of the lesion boundaries in the form of a binary mask. The Training Data file is a ZIP file, containing 900 dermoscopic lesion images in JPEG format. All images are named using the scheme ISIC_<image_id>.jpg, where <image_id> is a 7-digit unique identifier. EXIF tags in the images have been removed; any remaining EXIF tags should not be relied upon to provide accurate metadata. The Training Ground Truth file is a ZIP file, containing 900 binary mask images in PNG format. All masks are named using the scheme ISIC_<image_id>_Segmentation.png, where <image_id> matches the corresponding Training Data image for the mask. All mask images will have the exact same dimensions as their corresponding lesion image. Mask images are encoded as single-channel (grayscale) 8-bit PNGs (to provide lossless compression), where each pixel is either: 0: representing the background of the image or areas outside the lesion 255: representing the foreground of the image or areas inside the lesion

This dataset contains a large number of segmented nuclei images. The images were acquired under a variety of conditions and vary in the cell type, magnification, and imaging modality (brightfield vs. fluorescence). The dataset is designed to challenge an algorithm's ability to generalize across these variations. Each image is represented by an associated ImageId. Files belonging to an image are contained in a folder with this ImageId. Within this folder are two subfolders: images contains the image file. masks contains the segmented masks of each nucleus. This folder is only included in the training set. Each mask contains one nucleus. Masks are not allowed to overlap (no pixel belongs to two masks). The second stage dataset will contain images from unseen experimental conditions. To deter hand labeling, it will also contain images that are ignored in scoring. The metric used to score this competition requires that your submissions are in run-length encoded format. Please see the evaluation page for details. As with any human-annotated dataset, you may find various forms of errors in the data. You may manually correct errors you find in the training set. The dataset will not be updated/re-released unless it is determined that there are a large number of systematic errors. The masks of the stage 1 test set will be released with the release of the stage 2 test set.

LiTS17 is a liver tumor segmentation benchmark. The data and segmentations are provided by various clinical sites around the world. The training data set contains 130 CT scans and the test data set 70 CT scans. Image Source: https://arxiv.org/pdf/1707.07734.pdf

Spine or vertebral segmentation is a crucial step in all applications regarding automated quantification of spinal morphology and pathology. With the advent of deep learning, for such a task on computed tomography (CT) scans, a big and varied data is a primary sought-after resource. However, a large-scale, public dataset is currently unavailable. VerSe is a large scale, multi-detector, multi-site, CT spine dataset consisting of 374 scans from 355 patients. The challenge was held in two iterations in conjunction with MICCAI 2019 and 2020. The tasks evaluated for include: vertebral labelling and segmentation. Image source: https://arxiv.org/pdf/2001.09193v5.pdf

WORD is a dataset for organ semantic segmentation that contains 150 abdominal CT volumes (30,495 slices) and each volume has 16 organs with fine pixel-level annotations and scribble-based sparse annotation, which may be the largest dataset with whole abdominal organs annotation.

BRATS 2013 is a brain tumor segmentation dataset consists of synthetic and real images, where each of them is further divided into high-grade gliomas (HG) and low-grade gliomas (LG). There are 25 patients with both synthetic HG and LG images and 20 patients with real HG and 10 patients with real LG images. For each patient, FLAIR, T1, T2, and post-Gadolinium T1 magnetic resonance (MR) image sequences are available. Source: Learning Fixed Points in Generative Adversarial Networks: From Image-to-Image Translation to Disease Detection and Localization Image Source: https://arxiv.org/pdf/1708.00377.pdf

Under Institutional Review Board (IRB) supervision, 50 abdomen CT scans of were randomly selected from a combination of an ongoing colorectal cancer chemotherapy trial, and a retrospective ventral hernia study. The 50 scans were captured during portal venous contrast phase with variable volume sizes (512 x 512 x 85 - 512 x 512 x 198) and field of views (approx. 280 x 280 x 280 mm3 - 500 x 500 x 650 mm3). The in-plane resolution varies from 0.54 x 0.54 mm2 to 0.98 x 0.98 mm2, while the slice thickness ranges from 2.5 mm to 5.0 mm. The standard registration data was generated by NiftyReg. Source: MICCAI 2015 Multi-Atlas Abdomen Labeling Challenge Image source: MICCAI 2015 Multi-Atlas Abdomen Labeling Challenge

The ISIC 2018 dataset was published by the International Skin Imaging Collaboration (ISIC) as a large-scale dataset of dermoscopy images. This Task 1 dataset is the challenge on lesion segmentation. It includes 2594 images. Source: Bi-Directional ConvLSTM U-Net with Densley Connected Convolutions Image Source: https://challenge2018.isic-archive.com/task1/

Head and Neck Tumor Segmentation

Retinal OCTA SEgmentation dataset (ROSE) consists of 229 OCTA images with vessel annotations at either centerline-level or pixel level. Source: ROSE: A Retinal OCT-Angiography Vessel Segmentation Dataset and New Model Image Source: https://imed.nimte.ac.cn/dataofrose.html

Introduced by Da et al. in DigestPath: a Benchmark Dataset with Challenge Review for the Pathological Detection and Segmentation of Digestive-System ### Grand-Challenge Page ### 1. Signet ring cell dataset Signet ring cell carcinoma is a type of rare adenocarcinoma with poor prognosis. Early detection of such cells leads to huge improvement of patients' survival rate. However, there is no existing public dataset with annotations for studying the problem of signet ring cell detection. This dataset has positive samples and negative samples. Training positive samples contain 77 images from 20 WSIs, with cell bounding boxes written in xml. Training negative samples contain 378 images from 79 WSIs.These negative WSIs have no signet ring, but could contain other kinds of tumor cells. Each signet ring cell is labeled by experienced pathologists with a rectangle bounding box tightly surrounding the cell. Each image is of size 2000X2000. The training images are from 2 organs, including gastric mucosa and intestine. Because of the difficulty of manual annotation, there exist some signet ring cells who are missed by pathologists. In other words, this dataset is a noisy dataset with its positive images not fully annotated. All whole slide images were stained by hematoxylin and eosin and scanned at X40. ### 2. Colonoscopy tissue segment dataset Colonoscopy pathology examination can find cells of early-stage colon tumor from small tissue slices. Pathologists need to daily examine hundreds of tissue slices, which is a time consuming and exhausting work. Here we propose a challenge task on automatic colonoscopy tissue segmentation and screening, aiming at automatic lesion segmentation and classification of the whole tissue (benign vs. malignant). This dataset has positive samples and negative samples. Training positive samples contain 250 images of tissue from 93 WSIs, with pixel-level annotation in jpg format, where 0 means background and 255 for foreground (malignant lesion). You could simply get binary mask by a threshold 128. Training negative samples contain 410 images of tissue from 231 WSI. This negative images have no annotation because they don't have any malignant lesion. The average size of all images are of 5000x5000 pixels, some of them are extremely huge. We will also provide another 152 patients' 212 tissues as the testing set, in which 90 images from 65 patients contain lesion. All whole slide images were stained by hematoxylin and eosin and scanned at X20. Sign the DATABASE USE AGREEMENT first and download the dataset at the homepage! Da Q, Huang X, Li Z, et al. DigestPath: a Benchmark Dataset with Challenge Review for the Pathological Detection and Segmentation of Digestive-System[J]. Medical Image Analysis, 2022: 102485. （https://doi.org/10.1016/j.media.2022.102485）

MosMedData contains anonymised human lung computed tomography (CT) scans with COVID-19 related findings, as well as without such findings. A small subset of studies has been annotated with binary pixel masks depicting regions of interests (ground-glass opacifications and consolidations). CT scans were obtained between 1st of March, 2020 and 25th of April, 2020, and provided by municipal hospitals in Moscow, Russia. Source: MosMedData: Chest CT Scans With COVID-19 Related Findings Dataset

This dataset has 1,842 images with pixel-level DR-related lesion annotations, and 1,000 images with image-level labels graded by six board-certified ophthalmologists with intra-rater consistency. The proposed dataset will enable extensive studies on DR diagnosis. Source: A Benchmark for Studying Diabetic Retinopathy: Segmentation, Grading, and Transferability

The RSNA-ASNR-MICCAI BraTS 2021 challenge utilizes multi-institutional pre-operative baseline multi-parametric magnetic resonance imaging (mpMRI) scans, and focuses on the evaluation of state-of-the-art methods for (Task 1) the segmentation of intrinsically heterogeneous brain glioblastoma sub-regions in mpMRI scans. Furthemore, this BraTS 2021 challenge also focuses on the evaluation of (Task 2) classification methods to predict the MGMT promoter methylation status.

Consists of annotated frames containing GI procedure tools such as snares, balloons and biopsy forceps, etc. Beside of the images, the dataset includes ground truth masks and bounding boxes and has been verified by two expert GI endoscopists. Source: Kvasir-Instrument: Diagnostic and therapeutic tool segmentation dataset in gastrointestinal endoscopy

CoIR (Code Information Retrieval) benchmark, is designed to evaluate code retrieval capabilities. CoIR includes 10 curated code datasets, covering 8 retrieval tasks across 7 domains. In total, it encompasses two million documents. It also provides a common and easy Python framework, installable via pip, and shares the same data schema as benchmarks like MTEB and BEIR for easy cross-benchmark evaluations.