leong1230 / deep-learning-in-medical-application-

Collection of Deep Learning in mediacal applications.

Medical Image Dataset

Please respect all usage restrictions for any of the data and resources listed here.

[det.] 💨 Detection (Localization)
[seg.] 💨 Segmentation
[reg.] 💨 Registration
[cla.] 💨 Disease Classification
[age.] 💨 Biological Age Estimating
[rec.] 💨 Reconstruction
[pred.] 💨 Age/Disease Predition
[syn.] 💨 Image Synthesis/Generation (usually for data augmentation)
[trac.] 💨 Tracking

Retina/OCT

• Kaggle contest dataset for Diabetic retinopathy detection 👉[det.][syn.]
• DRIVE: Digital Retinal Images for Vessel Extraction 👉[seg.]
• MICCAI 2020 REFUGE challenge dataset: Glaucoma detection and optic disc/cup segmentation on a standard dataset of retinal fundus images. 👉[seg.][det.]
• STARE
• CHASE_DB1

Chest/Lung

• [X-ray]CHEST X-RAY 14: first large-scale chest X-ray dataset, released on 2017. 👉[seg.]
• [CT] NIH dataset 👉
• [X-ray] CheXpert(from 斯坦福 吴恩达团队): 224k images, PA and L views, 13 labels, automatic ruled-based labeler. 👉 Paper
• [X-ray] PADCHEST: 160k images, Multiple views, almost 200 labels. 27% hand labelled, other using RNN. 👉 Paper
• [X-ray] MIMIC-CXR: 377k images. PA and L views, 13 labels, automatic ruled-based labeler. 👉
• [CT/Pathology] the National Lung Screening Trial (NLST):include data on participant characteristics, screening exam results, diagnostic procedures, lung cancer, and mortality. 👉

Pathology/Microscopy/Gene

MICCAI Challenge on Circuit Reconstruction from Electron Microscopy Images 👉[reg.][seg.][det.][rec.]
The Cancer Genome Atlas (TCGA), a landmark cancer genomics program, molecularly characterized over 20,000 primary cancer and matched normal samples spanning 33 cancer types. 👉[pred.]

Endoscopy

Cholec80 dataset：contains 80 videos of cholecystectomy surgeries performed by 13 surgeons. The videos are captured at 25 fps. 👉 Paper [det.][trac.]

Brain

• (ADAM) Intracranial Aneurysm Detection and Segmentation Challenge（颅内动脉瘤检测和分割竞赛)👉

Kidney

• MICCAI 2019 Kits-challenge. 👉 or github

Cardiac

• [US] EchoNet-Dynamic 👉
• [SPECT] SPECTF Heart Data Set: 267 SPECT images that are descibed by 45 binary attributes.👉 (SPECT Heart Data Set👉
• [CCTA] Automated Segmentation of Coronary Arteries: 40 data. 👉
• [MR] Second Annual Data Science Bowl: hundreds of cardiac MRI images in DICOM format, for determining the left ventricle volume.👉
• [MR ]Sunnybrook Cardiac Data (SCD): the 2009 Cardiac MR Left Ventricle Segmentation Challenge data, consist of 45 cine-MRI images from a mixed of patients and pathologies. 👉
• [MR] SCMR CONSENSUS CONTOUR DATA: 15 Cardiac MRI with contour GT.👉
• [MR] Congenital Heart Disease (CHD) Atlas:he purpose of the study is to create a database of cardiac images and computational models in CHD patients that will enable clinicians and scientists to examine detailed ventricular shape and function in individual patients and compare them statistically against a database of other, similar CHD patient examinations.👉
• [MR] ACDC: Segmentation and dignose. composed by 150 patients with 3D cine-MR datasets.👉
• [MR] ukbb_cardiac:cardiac MR images, need registration. 👉
• [ECG]MIT-BIH Arrhythmia Database:arrhythmia analysis and related subjects，containing 48 half-hour excerpts of two-channel ambulatory ECG recordings. 👉
• [ECG]Arrhythmia Data Set(UIC):Distinguish between the presence and absence of cardiac arrhythmia and classify it in one of the 16 groups.👉
• [Multi-data]MESA: It aims to investigate the manifestation of subclinical to clinical cardiovascular disease before signs and symptoms develop. All imaging examinations used a four-element phased-array surface coil placed anterior and posterior, ECG gating, and brachial artery blood pressure monitoring.👉
• [Multi-data]Heart Disease Dataset: contains 4 database and 76 attributes.👉 or 👉
• [Multi-data]Z-Alizadeh Sani Data Set：contains the records of 303 patients, each of which have 54 features.The features are arranged in four groups: demographic, symptom and examination, ECG, and laboratory and echo features.👉
• [Collection] Cardiac Atlas Project👉

Software

• 3D Slicers: free open source software for image analysis and scientific visualization, providing image registration, processing of DTI (diffusion tractography), an interface to external devices for image guidance support, and GPU-enabled volume rendering, among other capabilities. 👉

Other resources

• [Github] Harvard Medical Data for Machine Learning 👉
• Academic torrents 👉
• [Github] medical-imaging-datasets 👉
• MedPix：aiming to teach medical konwledge to people major in medicine. Including cases for many disease, but not suitable for ML, since the database is small and often show only one case for each disease. 👉
• UK Biobank:released imaging data on 5,000 scanned participants from its imaging pilot study. 👉
• MedMNIST: a collection of 10 pre-processed medical open datasets. 👉

Other dataset include non-medical data

• Awesome Public Dataset 👉
• Data is Plural 👉
• Kaggle Challenge
• Google Dataset Search

Software

• Amira 3D：medical data segmentation
• Mimics 16.0：doctors can use to label the GT for segmentation
• ukbb_cardiac:a toolbox used for processing and analysing cardiovascular magnetic resonance (CMR) images, including preprocessing, FCN training and testing, evaluation, ect.
• 3D Slicer: registration and so on.