A large annotated medical image dataset for the development and evaluation of segmentation algorithms

By Amber L. Simpson, M. Antonelli, S. Bakas, M. Bilello, Keyvan Farahani, B. van Ginneken, A. Kopp-Schneider, Bennett A. Landman, Geert J. S. Litjens, Bjoern H Menze and 14 more. arXiv.org, 2019

1,013 citations recorded by the source corpus.

Semantic segmentation of medical images aims to associate a pixel with a label in a medical image without human initialization. The success of semantic segmentation algorithms is contingent on the availability of high-quality imaging data with corresponding labels provided by experts. We sought to create a large collection of annotated medical image datasets of various clinically relevant anatomies available under open source license to facilitate the development of semantic segmentation algorithms. Such a resource would allow: 1) objective assessment of general-purpose segmentation methods through comprehensive benchmarking and 2) open and free access to medical image data for any researcher interested in the problem domain. Through a multi-institutional effort, we generated a large, curated dataset representative of several highly variable segmentation tasks that was used in a crowd-sourced challenge - the Medical Segmentation Decathlon held during the 2018 Medical Image Computing and Computer Aided Interventions Conference in Granada, Spain. Here, we describe these ten labeled image datasets so that these data may be effectively reused by the research community.

Subfields: Computer Vision and Pattern Recognition, Radiology, Nuclear Medicine and Imaging

View Paper: https://www.semanticscholar.org/paper/4654aa505e5bcdb089d0df202cd7ceabc9d2d41f

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Radiology, Nuclear Medicine and Imaging

Citations in the corpus

UNETR++: Delving Into Efficient and Accurate 3D Medical Image Segmentation (2022, methodology): “…ﬁve datasets: Synapse for Multi-organ CT Segmentation [19], BTCV for Multi-organ CT Segmentation [19], ACDC for Automated Cardiac Diagnosis [1], Brain Tumor Segmentation (BraTS) [24] and the Medical Segmentation Decathlon-Lung [30]. of right ventricle (RV), left ventricle (LV) and myocardium (MYO).”
INet: Convolutional Networks for Biomedical Image Segmentation (2021, methodology): “…of the proposed method in the segmentation of biomedical images, we carried out experiments on nine public imaging datasets: three brain (BraTS2013, LMS, BTD) one heart MRI [42]–[44], one liver one spleen CT [44], [45], one lung X-ray [9], one colon endoscopic [46], and one nerve ultrasonic [47].”
Medical Image Segmentation Using Deep Learning: A Survey (2020, background): “Spleen Medical Segmentation Decathlon (MSD) [164] http://medicaldecathlon.”
MedMNIST v2 - A large-scale lightweight benchmark for 2D and 3D biomedical image classification (2021, neutral): “22, 23 Chest X-Ray BC (2) 5,856 4,708 / 524 / 624 RetinaMNIST DeepDRiD Team24 Fundus Camera OR (5) 1,600 1,080 / 120 / 400 BreastMNIST Al-Dhabyani et al.”
A Review of Deep-Learning-Based Medical Image Segmentation Methods (2021, background): “[105] created a large, open source, hand-annotated medical image data set of various anatomical parts.”
Segment Anything in Medical Images (2023, background): “, T1, T2, ADC, FLAIR) [24][10][6][41][5][8][29][42], abdomen tumor, COVID-19 infection, gallbladder, head and neck tumor, liver, pancreas, pleural effusion, and stomach segmentation in CT images [34][19][7][33][32][42][4][28][27][12], breast tumor, liver, and vessel segmentation in ultrasound images”
TransFuse: Fusing Transformers and CNNs for Medical Image Segmentation (2021, methodology): “4)Prostate Segmentation, where volumetric Prostate Multi-modality MRIs from the Medical Segmentation Decathlon [22] are used.”
Swin UNETR: Swin Transformers for Semantic Segmentation of Brain Tumors in MRI Images (2022, background): “Recently, deep learning-based brain tumor segmentation techniques [31,21,32,20] have achieved state-of-the-art performance in various benchmarks [7,35,2].”
Generalizing Deep Learning for Medical Image Segmentation to Unseen Domains via Deep Stacked Transformation (2020, methodology): “…MRI and Heart MRI datasets from the MSD challenge are selected as the source domain data of our Task 1 and Task 2 , respectively, because: 1) MSD [35] is a recent large scale annotated medical image dataset which represents the state-of-the-art dataset with high quality; 2) moreover, more than…”
AbdomenCT-1K: Is Abdominal Organ Segmentation a Solved Problem? (2020, methodology): “NIH Pancreas(9) [21], [22], [23] Pancreas 80 1 The Cancer Imaging Archive 2015 VISCERAL Anatomy Benchmark(10) [86] 20 anatomical structures 80/40 1 ISBI and ECIR 2015 Liver Tumor Segmentation Benchmark (LiTS)(11) [16] Liver and tumor 131/70 7 ISBI and MICCAI 2017 Medical Segmentation Decathlon (MSD)”
nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation (2020, methodology): “We demonstrate the value of nnU-Net as an out-of-the-box segmentation tool by applying it to 11 international biomedical image segmentation challenges comprising 23 different datasets and 53 segmentation tasks 6,14,16–24 (https://cremi.org/).”
UNETR: Transformers for 3D Medical Image Segmentation (2021, methodology): “chitectures [22, 23, 21] have achieved state-of-the-art results in various medical semantic segmentation tasks [2, 38, 19].”
CHAOS Challenge - Combined (CT-MR) Healthy Abdominal Organ Segmentation (2020, background): “The volumetric overlap is found to be 97.21%, while RVD is 0.347%, ASSD is 0.611 (0.263 mm), RMSD is 1.04 (0.449 mm), and MSSD is 13.038 (5.632 mm).”
Learning with Limited Annotations: A Survey on Deep Semi-Supervised Learning for Medical Image Segmentation (2022, methodology): “[108] 2D CT, MRI ACDC dataset [30], SCGM [114], Spleen dataset [113] Perturbations”
Transformers in Medical Imaging: A Survey (2022, methodology): “UNETR achieves impressive performance on BTCV [203] and MSD [171] segmentation datasets as shown in Fig.”
The Medical Segmentation Decathlon (2021, background): “The Decathlon challenge made ten data sets available online [11], where each data set had between one and three region-of-interest (ROI) targets (17 targets in total).”
Federated learning for medical image analysis: A survey (2023, neutral): “(2021) [143] Prostate segmentation MSD-Prostate [144] PROMISE12 [135] ProstateX [139] NCI-ISBI 2013 [136] MRI U-Net”