Knee Magnetic Resonance Images


The outbreak of COVID-19 in Wuhan, China in December 2019 has rapidly spread across other countries in the world and has been declared as a global pandemic by WHO on 11th March, 2020. COVID-19 continues to have adverse effects on the health and economy of the global population and has brought immense pressure on the health care systems of the developing as well as developed countries.


Please refer the "Readme_CXR_Database_v1.0" for detailed instructions on how to use the dataset. 


The following pages show axial T2-weighted MRI obtained at 24 hours and at 3-15 months after MRgFUS. The images shown here were registered to the same reference frame that was used in the thermal simulations; every third image is shown. To segment the bone marrow lesions, the registered images were toggled back and forth between the two time points to detect obvious changes. The lesion segmentations were completed before the acoustic and thermal simulations were performed. They were originally done on the native T2-weighted images acquired at 3-15 months after FUS.


This set contains 1450 fundus images with 899 glaucoma data and 551 normal data.

All text about patient information and the date that the associated images were collected are replaced by 0, which is black.


BCI-Double-ErrP-Dataset is an EEG dataset recorded while participants used a P300-based BCI speller. This speller uses a P300 post-detection based on Error-related potentials (ErrPs) to detect and correct errors (i.e. when the detected symbol does not match the user’s intention). After the P300 detection, an automatic correction is made when an ErrP is detected (this is called a “Primary ErrP”). The correction proposed by the system is also evaluated, eventually eliciting a “Secondary ErrP” if the correction is wrong.


A detailed description of the data is given in “BCI-Double-ErrP-Dataset-instructions.pdf” and a Matlab code example is provided to extract P300 and ErrPs (primary and secondary).


There are 4 folders, one with the datasets of the P300 calibration (session 1), one with the datasets of the ErrP calibration (session 1), one with the datasets of the testing session (session 2), and a folder with the Matlab code to run the example.


BraTS has always been focusing on the evaluation of state-of-the-art methods for the segmentation of brain tumors in multimodal magnetic resonance imaging (MRI) scans. BraTS 2019 utilizes multi-institutional pre-operative MRI scans and focuses on the segmentation of intrinsically heterogeneous (in appearance, shape, and histology) brain tumors, namely gliomas. Furthemore, to pinpoint the clinical relevance of this segmentation task, BraTS’19 also focuses on the prediction of patient overall survival, via integrative analyses of radiomic features and machine learning algorithms.

Last Updated On: 
Fri, 02/28/2020 - 06:31

This demo is intended to implement Ultrasound Localization Microscopy by a modified sub-pixel convolutional neural network (mSPCN-ULM). The detailed information can be referred in Liu et al. "Deep Learning for Ultrasound Localization Microscopy (DOI: 10.1109/TMI.2020.2986781)".


Transcranial Doppler (TCD) echo data was recorded from healthy adults and neurocritical care adult patients. The insonated cerebral vessels were the middle cerebral artery (MCA) and the internal carotid artery (ICA). The ultrasound system used in this study was the Philips CX50.


There are two code examples included. One is a visualizer for plotting the spectrograms and the other is the actual code for tracing the maximal flow velocity.

  • Use the 'spectrogram_viewer.m' MATLAB script in the 'Spectrogram visualizer' folder to visualize the Doppler spectrograms. In this script, set the variables 'filepath' and 'filename' to point to the TCD data.
  • The full algorithm that computes the spectrogram from the Doppler echo and estimates the maximal flow velocity is found in the folder named 'TCD tracing code'. The main function is 'computeCBFVMain.m'. Note, the variables 'filepath' and 'filename' need to point to the TCD data.

Data acquisition: The transcranial Doppler data were collected from healthy volunteers at Massachusetts Institute of Technology (MIT) and from patients in neurocritical care at Boston Medical Center (BMC). Data collection occurred between 2016 and 2020, was approved by the MIT and BMC Institutional Review Boards, and informed consent was obtained from the subjects directly at MIT or from the patients or their legally authorized representatives at BMC. The data consists of 16 recordings from healthy subjects and 29 recordings neurocritical care patients. 

Published papers

F. Wadehn and T. Heldt, "Adaptive Maximal Blood Flow Velocity Estimation From Transcranial Doppler Echos," in IEEE Journal of Translational Engineering in Health and Medicine, vol. 8, pp. 1-11, 2020, Art no. 1800511, doi: 10.1109/JTEHM.2020.3011562.

R. Jaishankar, A. Fanelli, A. Filippidis, T. Vu, J. Holsapple and T. Heldt, "A Spectral Approach to Model-Based Noninvasive Intracranial Pressure Estimation," in IEEE Journal of Biomedical and Health Informatics, vol. 24, no. 8, pp. 2398-2406, Aug. 2020, doi: 10.1109/JBHI.2019.2961403.




ADAM is organized as a half day Challenge, a Satellite Event of the ISBI 2020 conference in Iowa City, Iowa, USA.


It includes 312 ROIs. An ROI is a rectangular BMP image region. A rectangular image region  is located within a PDAC tumor region or within a HP region of a slice CT image. ROIs of 1-153 are PDAC, ROIs of 154:312 are HP.