The Medical Information Mart for Intensive Care III (MIMIC-III) dataset is a large, de-identified and publicly-available collection of medical records. Each record in the dataset includes ICD-9 codes, which identify diagnoses and procedures performed. Each code is partitioned into sub-codes, which often include specific circumstantial details. The dataset consists of 112,000 clinical reports records (average length 709.3 tokens) and 1,159 top-level ICD-9 codes. Each report is assigned to 7.6 codes, on average. Data includes vital signs, medications, laboratory measurements, observations and notes charted by care providers, fluid balance, procedure codes, diagnostic codes, imaging reports, hospital length of stay, survival data, and more. The database supports applications including academic and industrial research, quality improvement initiatives, and higher education coursework. Source: MIT Laboratory for Computational Biology

We introduce here a new database called UBFC-rPPG (stands for Univ. Bourgogne Franche-Comté Remote PhotoPlethysmoGraphy) comprising two datasets that are focused specifically on rPPG analysis. The UBFC-RPPG database was created using a custom C++ application for video acquisition with a simple low cost webcam (Logitech C920 HD Pro) at 30fps with a resolution of 640x480 in uncompressed 8-bit RGB format. A CMS50E transmissive pulse oximeter was used to obtain the ground truth PPG data comprising the PPG waveform as well as the PPG heart rates. During the recording, the subject sits in front of the camera (about 1m away from the camera) with his/her face visible. All experiments are conducted indoors with a varying amount of sunlight and indoor illumination. The link to download the complete video dataset is available on request. A basic Matlab implementation can also be provided to read ground truth data acquired with a pulse oximeter.

The UCR Time Series Archive - introduced in 2002, has become an important resource in the time series data mining community, with at least one thousand published papers making use of at least one data set from the archive. The original incarnation of the archive had sixteen data sets but since that time, it has gone through periodic expansions. The last expansion took place in the summer of 2015 when the archive grew from 45 to 85 data sets. This paper introduces and will focus on the new data expansion from 85 to 128 data sets. Beyond expanding this valuable resource, this paper offers pragmatic advice to anyone who may wish to evaluate a new algorithm on the archive. Finally, this paper makes a novel and yet actionable claim: of the hundreds of papers that show an improvement over the standard baseline (1-nearest neighbor classification), a large fraction may be misattributing the reasons for their improvement. Moreover, they may have been able to achieve the same improvement with a much simpler modification, requiring just a single line of code.

The MIT-BIH Arrhythmia Database contains 48 half-hour excerpts of two-channel ambulatory ECG recordings, obtained from 47 subjects studied by the BIH Arrhythmia Laboratory between 1975 and 1979. Twenty-three recordings were chosen at random from a set of 4000 24-hour ambulatory ECG recordings collected from a mixed population of inpatients (about 60%) and outpatients (about 40%) at Boston's Beth Israel Hospital; the remaining 25 recordings were selected from the same set to include less common but clinically significant arrhythmias that would not be well-represented in a small random sample. The recordings were digitized at 360 samples per second per channel with 11-bit resolution over a 10 mV range. Two or more cardiologists independently annotated each record; disagreements were resolved to obtain the computer-readable reference annotations for each beat (approximately 110,000 annotations in all) included with the database. This directory contains the entire MIT-BIH Arrhythmia Database. About half (25 of 48 complete records, and reference annotation files for all 48 records) of this database has been freely available here since PhysioNet's inception in September 1999. The 23 remaining signal files, which had been available only on the MIT-BIH Arrhythmia Database CD-ROM, were posted here in February 2005. Much more information about this database may be found in the MIT-BIH Arrhythmia Database Directory.

The MMSE-HR benchmark consists of a dataset of 102 videos from 40 subjects recorded at 1040x1392 raw resolution at 25fps. During the recordings, various stimuli such as videos, sounds, and smells are introduced to induce different emotional states in the subjects. The ground truth waveform for MMSE-HR is the blood pressure signal sampled at 1000Hz. The dataset contains a diverse distribution of skin colors in the Fitzpatrick scale (II=8, III=11, IV=17, V+VI=4).

The ECGs in this collection were obtained using a non-commercial, PTB prototype recorder with the following specifications: 16 input channels, (14 for ECGs, 1 for respiration, 1 for line voltage) Input voltage: ±16 mV, compensated offset voltage up to ± 300 mV Input resistance: 100 Ω (DC) Resolution: 16 bit with 0.5 μV/LSB (2000 A/D units per mV) Bandwidth: 0 - 1 kHz (synchronous sampling of all channels) Noise voltage: max. 10 μV (pp), respectively 3 μV (RMS) with input short circuit Online recording of skin resistance Noise level recording during signal collection The database contains 549 records from 290 subjects (aged 17 to 87, mean 57.2; 209 men, mean age 55.5, and 81 women, mean age 61.6; ages were not recorded for 1 female and 14 male subjects). Each subject is represented by one to five records. There are no subjects numbered 124, 132, 134, or 161. Each record includes 15 simultaneously measured signals: the conventional 12 leads (i, ii, iii, avr, avl, avf, v1, v2, v3, v4, v5, v6) together with the 3 Frank lead ECGs (vx, vy, vz). Each signal is digitized at 1000 samples per second, with 16 bit resolution over a range of ± 16.384 mV. On special request to the contributors of the database, recordings may be available at sampling rates up to 10 KHz. Within the header (.hea) file of most of these ECG records is a detailed clinical summary, including age, gender, diagnosis, and where applicable, data on medical history, medication and interventions, coronary artery pathology, ventriculography, echocardiography, and hemodynamics. The clinical summary is not available for 22 subjects. Source: PTB

VIPL-HR database is a database for remote heart rate (HR) estimation from face videos under less-constrained situations. It contains 2,378 visible light videos (VIS) and 752 near-infrared (NIR) videos of 107 subjects. Nine different conditions, including various head movements and illumination conditions are taken into consideration. All the videos are recorded using Logitech C310, RealSense F200 and the front camera of HUAWEI P9 smartphone, and the ground-truth HR is recorded using a CONTEC CMS60C BVP sensor (a FDA approved device). Source: VIPL-HR: A Multi-modal Database for Pulse Estimation from Less-constrained Face Video

The Multi-domain Mobile Video Physiology Dataset (MMPD), comprising 11 hours(1152K frames) of recordings from mobile phones of 33 subjects. The dataset was designed to capture videos with greater representation across skin tone, body motion, and lighting conditions. MMPD is comprehensive with eight descriptive labels and can be used in conjunction with the rPPG-toolbox and PhysBench. MMPD is widely used for rPPG tasks and remote heart rate estimation. To access the dataset, you are supposed to download this data release agreement and request downloading by email.

Over the past few years a number of research groups have made rapid advances in remote PPG methods for estimating heart rate from digital video and obtained impressive results. How these various methods compare in naturalistic conditions, where spontaneous behavior, facial expressions, and illumination changes are present, is relatively unknown. To enable comparisons among alternative methods, the Vision for Vitals dataset was introduced. It is a novel dataset containing high-resolution videos time-locked with varied physiological signals from a diverse population. It contains more than 150+ subjects with over 1300+ videos along with ground truth heart rate and respiration rate annotations. It also includes blood pressure waveform signals as part of its physiological data.

This dataset is composed of two collections of heartbeat signals derived from two famous PhysioNet datasets in heartbeat classification, the MIT-BIH Arrhythmia Dataset and the PTB Diagnostic ECG Database. The number of samples in both collections is large enough for training a deep neural network. This dataset has been used in exploring heartbeat classification using deep neural network architectures, and observing some of the capabilities of transfer learning on it. The signals correspond to electrocardiogram (ECG) shapes of heartbeats for the normal case and the cases affected by different arrhythmias and myocardial infarction. These signals are preprocessed and segmented, with each segment corresponding to a heartbeat.

The China Physiological Signal Challenge 2018 aims to encourage the development of algorithms to identify the rhythm/morphology abnormalities from 12-lead ECGs. The data used in CPSC 2018 include one normal ECG type and eight abnormal types. Source: An Open Access Database for Evaluating the Algorithms of Electrocardiogram Rhythm and Morphology Abnormality Detection

The Medical Dataset for Abbreviation Disambiguation for Natural Language Understanding (MeDAL) is a large medical text dataset curated for abbreviation disambiguation, designed for natural language understanding pre-training in the medical domain. It was published at the ClinicalNLP workshop at EMNLP. Source: https://github.com/McGill-NLP/medal Image Source: https://github.com/McGill-NLP/medal

Electrocardiography (ECG) is a key diagnostic tool to assess the cardiac condition of a patient. Automatic ECG interpretation algorithms as diagnosis support systems promise large reliefs for the medical personnel - only on the basis of the number of ECGs that are routinely taken. However, the development of such algorithms requires large training datasets and clear benchmark procedures. In our opinion, both aspects are not covered satisfactorily by existing freely accessible ECG datasets. The PTB-XL ECG dataset is a large dataset of 21799 clinical 12-lead ECGs from 18869 patients of 10 second length. The raw waveform data was annotated by up to two cardiologists, who assigned potentially multiple ECG statements to each record. The in total 71 different ECG statements conform to the SCP-ECG standard and cover diagnostic, form, and rhythm statements. To ensure comparability of machine learning algorithms trained on the dataset, we provide recommended splits into training and test sets. In combination with the extensive annotation, this turns the dataset into a rich resource for the training and the evaluation of automatic ECG interpretation algorithms. The dataset is complemented by extensive metadata on demographics, infarction characteristics, likelihoods for diagnostic ECG statements as well as annotated signal properties. Source: PTB-XL

A dataset of 12-lead ECGs with annotations. The dataset contains 345 779 exams from 233 770 patients. It was obtained through stratified sampling from the CODE dataset ( 15% of the patients). The data was collected by the Telehealth Network of Minas Gerais in the period between 2010 and 2016. Source: CODE-15%

Data Description The training data contains twelve-lead ECGs. The validation and test data contains twelve-lead, six-lead, four-lead, three-lead, and two-lead ECGs: 1. Twelve leads: I, II, III, aVR, aVL, aVF, V1, V2, V3, V4, V5, V6 2. Six leads: I, II, III, aVR, aVL, aVF 3. Four leads: I, II, III, V2 4. Three leads: I, II, V2 5. Two leads: I, II Each ECG recording has one or more labels that describe cardiac abnormalities (and/or a normal sinus rhythm). We mapped the labels for each recording to SNOMED-CT codes. The lists of scored labels and unscored labels are given with the evaluation code; see the scoring section for details. # Data Sources The Challenge data include recordings from last year’s Challenge and many new recordings for this year’s Challenge: 1. CPSC Database and CPSC-Extra Database 2. INCART Database 3. PTB and PTB-XL Database 4. The Georgia 12-lead ECG Challenge (G12EC) Database 5. Augmented Undisclosed Database 6. Chapman-Shaoxing and Ningbo Database 7. The University of Michigan (UMich) Database The Challenge data include annotated twelve-lead ECG recordings from six sources in four countries across three continents. These databases include over 100,000 twelve-lead ECG recordings with over 88,000 ECGs shared publicly as training data, 6,630 ECGs retained privately as validation data, and 36,266 ECGs retained privately as test data. - The first source is the China Physiological Signal Challenge in 2018 (CPSC 2018), which was held during the 7th International Conference on Biomedical Engineering and Biotechnology in Nanjing, China. This source contains two databases: the data from CPSC 2018 (the CPSC Database) and unused data from CPSC 2018 (the CPSC-Extra Database). Together, these databases contain 13,256 ECGs (10,330 ECGs shared as training data, 1,463 retained as validation data, and 1,463 retained as test data). We shared the training set and an unused dataset from CPSC 2018 as training data, and we split the test set from CPSC 2018 into validation and test sets. Each recording is between 6 and 144 seconds long with a sampling frequency of 500 Hz. - The second source is the St Petersburg INCART 12-lead Arrhythmia Database. This source contains 74 annotated ECGs (all shared as training data) extracted from 32 Holter monitor recordings. Each recording is 30 minutes long with a sampling frequency of 257 Hz. - The third source is the Physikalisch-Technische Bundesanstalt (PTB) and includes two public datasets: the PTB and the PTB-XL databases. The source contains 22,353 ECGs (all shared as training data). Each recording is between 10 and 120 seconds long with a sampling frequency of either 500 or 1,000 Hz. - The fourth source is a Georgia database which represents a unique demographic of the Southeastern United States. This source contains 20,672 ECGs (10,344 ECGs shared as training data, 5,167 retained as validation data, and 5,161 retained as test data). Each recording is between 5 and 10 seconds long with a sampling frequency of 500 Hz. - The fifth source is an undisclosed American database that is geographically distinct from the Georgia database. This source contains 10,000 ECGs (all retained as test data). - The sixth source is the Chapman University, Shaoxing People’s Hospital (Chapman-Shaoxing) and Ningbo First Hospital (Ningbo) database. This source contains 45,152 ECGS (all shared as training data). Each recording is 10 seconds long with a sampling frequency of 500 Hz. - The seventh source is UMich Database from the University of Michigan. This source contains 19,642 ECGs (all retained as test data). Each recording is 10 seconds long with a sampling frequency of either 250 Hz or 500 Hz. Like other real-world datasets, different databases may have different proportions of cardiac abnormalities, but all of the labels in the validation or test data are represented in the training data. Moreover, while this is a curated dataset, some of the data and labels are likely to have errors, and an important part of the Challenge is to work out these issues. In particular, some of the databases have human-overread machine labels with single or multiple human readers, so the quality of the labels varies between databases. You can find more information about the label mappings of the Challenge training data in this table. The six-lead, four-lead, three-lead, and two-lead validation data are reduced-lead versions of the twelve-lead validation data: the same recordings with the same header data but only with signal data for the relevant leads. We are not planning to release the test data at any point, including after the end of the Challenge. Requests for the test data will not receive a response. We do not release test data to prevent overfitting on the test data and claims or publications of inflated performances. We will entertain requests to run code on the test data after the Challenge on a limited basis based on publication necessity and capacity. (The Challenge is largely staged by volunteers.) # Data Format All data was formatted in WFDB format. Each ECG recording uses a binary MATLAB v4 file (see page 27) for the ECG signal data and a plain text file in WFDB header format for the recording and patient attributes, including the diagnosis, i.e., the labels for the recording. The binary files can be read using the load function in MATLAB and the scipy.io.loadmat function in Python; see our MATLAB and Python example code for working examples. The first line of the header provides information about the total number of leads and the total number of samples or time points per lead, the following lines describe how each lead was encoded, and the last lines provide information on the demographics and diagnosis of the patient. For example, a header file A0001.hea may have the following contents: A0001 12 500 7500 05-Feb-2020 11:39:16 A0001.mat 16+24 1000/mV 16 0 28 -1716 0 I A0001.mat 16+24 1000/mV 16 0 7 2029 0 II A0001.mat 16+24 1000/mV 16 0 -21 3745 0 III A0001.mat 16+24 1000/mV 16 0 -17 3680 0 aVR A0001.mat 16+24 1000/mV 16 0 24 -2664 0 aVL A0001.mat 16+24 1000/mV 16 0 -7 -1499 0 aVF A0001.mat 16+24 1000/mV 16 0 -290 390 0 V1 A0001.mat 16+24 1000/mV 16 0 -204 157 0 V2 A0001.mat 16+24 1000/mV 16 0 -96 -2555 0 V3 A0001.mat 16+24 1000/mV 16 0 -112 49 0 V4 A0001.mat 16+24 1000/mV 16 0 -596 -321 0 V5 A0001.mat 16+24 1000/mV 16 0 -16 -3112 0 V6 #Age: 74 #Sex: Male #Dx: 426783006 #Rx: Unknown #Hx: Unknown #Sx: Unknown From the first line of the file, we see that the recording number is A0001, and the recording file is A0001.mat. The recording has 12 leads, each recorded at a 500 Hz sampling frequency, and contains 7500 samples. From the next 12 lines of the file (one for each lead), we see that each signal was written at 16 bits with an offset of 24 bits, the floating point number (analog-to-digital converter (ADC) units per physical unit) is 1000/mV, the resolution of the analog-to-digital converter (ADC) used to digitize the signal is 16 bits, and the baseline value corresponding to 0 physical units is 0. The first value of the signal (-1716, etc.), the checksum (0, etc.), and the lead name (I, etc.) are the last three entries of each of these lines. From the final 6 lines, we see that the patient is a 74-year-old male with a diagnosis (Dx) of 426783006, which is the SNOMED-CT code for sinus rhythm. The medical prescription (Rx), history (Hx), and symptom or surgery (Sx) are unknown. Please visit WFDB header format for more information on the header file and variables.

Abstract Lobachevsky University Electrocardiography Database (LUDB) is an ECG signal database with marked boundaries and peaks of P, T waves and QRS complexes. The database consists of 200 10-second 12-lead ECG signal records representing different morphologies of the ECG signal. The ECGs were collected from healthy volunteers and patients of the Nizhny Novgorod City Hospital No 5 in 2017–2018. The patients had various cardiovascular diseases while some of them had pacemakers. The boundaries of P, T waves and QRS complexes were manually annotated by cardiologists for all 200 records. Also, each record is annotated with the corresponding diagnosis. The database can be used for educational purposes as well as for training and testing algorithms for ECG delineation, i.e. for automatic detection of boundaries and peaks of P, T waves and QRS complexes. # Background Validating ECG delineation algorithms requires standardized databases with complexes and waves, manually annotated by specialists. Several collections are currently available: MIT-BIH Arrhythmia Database [1], European ST-T Database [2], and QT Database [3], however their annotation is not exhaustive. For example, MIT-BIH Arrhythmia Database and European ST-T Database has a markup only for QRS complexes. The QT Database contains annotations for P, QRS and T waves, but several complexes are unmarked. By assembling a new ECG database at Lobachevsky University (LUDB), we sought to eliminate these shortcomings. # Methods ECG 10 seconds records were obtained by the Schiller Cardiovit AT-101 cardiograph, with conventional 12 leads (i, ii, iii, avr, avl, avf, v1, v2, v3, v4, v5, v6). Signals are digitized at 500 samples per second. The boundaries and peaks of P, T waves and QRS complexes were determined by certified cardiologists by an eye inspection of each ECG signal and independently for each of 12 leads. The records were made by specialized medical staff (functional diagnostics nurses). All volunteers provided informed written consent before collecting the data. The research was approved by Lobachevsky University IRB (#23; 19 October 2017). # Data Description The database consists of 200 10-second 12-lead ECG signal records collected from 2017 to 2018: in total, 16797 P waves, 21966 QRS complexes, 19666 T waves (in total, 58429 annotated waves). The age of all volunteers ranged from a minimum of 11 years old to a maximum of >89 years old with an average 52 years old while the distribution by gender was 85 women and 115 men. The number of records with specified heart rate types in the dataset: | Rhythms |Number of ECGs| |------|--------| |Sinus rhythm |143| |Sinus tachycardia| 4| |Sinus bradycardia| 25| |Sinus arrhythmia| 8| |Irregular sinus rhythm 2| |Abnormal rhythm| 19| The number of records with specified types of the position of the electrical axis of the heart: |Electric axis of the heart| Number of ECGs| |------|--------| |Normal |75| |Left axis deviation |66| |Vertical |26| |Horizontal| 20| |Right axis deviation| 3| |Undetermined| 10| The number of records with specified types of conduction abnormalities: |Conduction abnormalities| Number of ECGs| |------|--------| |Sinoatrial blockade, undetermined |1| |I degree AV block |10| |III degree AV-block |5| |Incomplete right bundle branch block| 29| |Incomplete left bundle branch block |6| |Left anterior hemiblock |16| |Complete right bundle branch block |4| |Complete left bundle branch block |4| |Non-specific intravintricular conduction delay| 4| The numbers of records with specified types of extrasystolies: |Extrasystolies |Number of ECGs| |------|--------| |Atrial extrasystole, undetermined |2| |Atrial extrasystole, low atrial |1| |Atrial extrasystole, left atrial |2| |Atrial extrasystole, SA-nodal extrasystole |3| |Atrial extrasystole, type: single PAC |4| |Atrial extrasystole, type: bigemini |1| |Atrial extrasystole, type: quadrigemini |1| |Atrial extrasystole, type: allorhythmic pattern |1| |Ventricular extrasystole, morphology: polymorphic |2| |Ventricular extrasystole, localisation: RVOT, anterior wall |3| |Ventricular extrasystole, localisation: RVOT, antero-septal part |1| |Ventricular extrasystole, localisation: IVS, middle part |1| |Ventricular extrasystole, localisation: LVOT, LVS |2| |Ventricular extrasystole, localisation: LV, undefined |1| |Ventricular extrasystole, type: single PVC |6| |Ventricular extrasystole, type: intercalary PVC |2| |Ventricular extrasystole, type: couplet |2| The number of records with specified types of hypertrophies: |Hypertrophies |Number of ECGs| |------|--------| |Right atrial hypertrophy |1| |Left atrial hypertrophy |102| |Right atrial overload |17| |Left atrial overload |11| |Left ventricular hypertrophy |108| |Right ventricular hypertrophy |3| |Left ventricular overload |11| The number of records with cardiac pacing: |Cardiac pacing |Number of ECGs| |------|--------| |UNIpolar atrial pacing |1| |UNIpolar ventricular pacing |6| |BIpolar ventricular pacing |2| |Biventricular pacing |1| |P-synchrony |2| The number of records with ischemia: |Ischemia |Number of ECGs| |------|--------| |STEMI: anterior wall |8| |STEMI: lateral wall |7| |STEMI: septal |8| |STEMI: inferior wall |1| |STEMI: apical |5| |Ischemia: anterior wall |5| |Ischemia: lateral wall |8| |Ischemia: septal |4| |Ischemia: inferior wall |10| |Ischemia: posterior wall |2| |Ischemia: apical |6| |Scar formation: lateral wall |3| |Scar formation: septal |9| |Scar formation: inferior wall |3| |Scar formation: posterior wall |6| |Scar formation: apical |5| |Undefined ischemia/scar/supp.NSTEMI: anterior wall |12| |Undefined ischemia/scar/supp.NSTEMI: lateral wall |16| |Undefined ischemia/scar/supp.NSTEMI: septal |5| |Undefined ischemia/scar/supp.NSTEMI: inferior wall |3| |Undefined ischemia/scar/supp.NSTEMI: posterior wall| 4| |Undefined ischemia/scar/supp.NSTEMI: apical |11| The number of records with non-specific repolarization abnormalities: |Non-specific repolarization abnormalities |Number of ECGs| |------|--------| |Anterior wall |18| |Lateral wall |13| |Septal |15| |Inferior wall |19| |Posterior wall |9| |Apical |11| The number of records with other cases: |Other states |Number of ECGs| |------|--------| |Early repolarization syndrome |9|

The eICU Collaborative Research Database is a large multi-center critical care database made available by Philips Healthcare in partnership with the MIT Laboratory for Computational Physiology. The eICU Collaborative Research Database holds data associated with over 200,000 patient stays, providing a large sample size for research studies.

Data ## The data for this Challenge are from multiple sources: CPSC Database and CPSC-Extra Database INCART Database PTB and PTB-XL Database The Georgia 12-lead ECG Challenge (G12EC) Database Undisclosed Database The first source is the public (CPSC Database) and unused data (CPSC-Extra Database) from the China Physiological Signal Challenge in 2018 (CPSC2018), held during the 7th International Conference on Biomedical Engineering and Biotechnology in Nanjing, China. The unused data from the CPSC2018 is NOT the test data from the CPSC2018. The test data of the CPSC2018 is included in the final private database that has been sequestered. This training set consists of two sets of 6,877 (male: 3,699; female: 3,178) and 3,453 (male: 1,843; female: 1,610) of 12-ECG recordings lasting from 6 seconds to 60 seconds. Each recording was sampled at 500 Hz. The second source set is the public dataset from St Petersburg INCART 12-lead Arrhythmia Database. This database consists of 74 annotated recordings extracted from 32 Holter records. Each record is 30 minutes long and contains 12 standard leads, each sampled at 257 Hz. The third source from the Physikalisch Technische Bundesanstalt (PTB) comprises two public databases: the PTB Diagnostic ECG Database and the PTB-XL, a large publicly available electrocardiography dataset. The first PTB database contains 516 records (male: 377, female: 139). Each recording was sampled at 1000 Hz. The PTB-XL contains 21,837 clinical 12-lead ECGs (male: 11,379 and female: 10,458) of 10 second length with a sampling frequency of 500 Hz. The fourth source is a Georgia database which represents a unique demographic of the Southeastern United States. This training set contains 10,344 12-lead ECGs (male: 5,551, female: 4,793) of 10 second length with a sampling frequency of 500 Hz. The fifth source is an undisclosed American database that is geographically distinct from the Georgia database. This source contains 10,000 ECGs (all retained as test data). All data is provided in WFDB format. Each ECG recording has a binary MATLAB v4 file (see page 27) for the ECG signal data and a text file in WFDB header format describing the recording and patient attributes, including the diagnosis (the labels for the recording). The binary files can be read using the load function in MATLAB and the scipy.io.loadmat function in Python; please see our baseline models for examples of loading the data. The first line of the header provides information about the total number of leads and the total number of samples or points per lead. The following lines describe how each lead was saved, and the last lines provide information on demographics and diagnosis. Below is an example header file A0001.hea: A0001 12 500 7500 05-Feb-2020 11:39:16 A0001.mat 16+24 1000/mV 16 0 28 -1716 0 I A0001.mat 16+24 1000/mV 16 0 7 2029 0 II A0001.mat 16+24 1000/mV 16 0 -21 3745 0 III A0001.mat 16+24 1000/mV 16 0 -17 3680 0 aVR A0001.mat 16+24 1000/mV 16 0 24 -2664 0 aVL A0001.mat 16+24 1000/mV 16 0 -7 -1499 0 aVF A0001.mat 16+24 1000/mV 16 0 -290 390 0 V1 A0001.mat 16+24 1000/mV 16 0 -204 157 0 V2 A0001.mat 16+24 1000/mV 16 0 -96 -2555 0 V3 A0001.mat 16+24 1000/mV 16 0 -112 49 0 V4 A0001.mat 16+24 1000/mV 16 0 -596 -321 0 V5 A0001.mat 16+24 1000/mV 16 0 -16 -3112 0 V6 #Age: 74 #Sex: Male #Dx: 426783006 #Rx: Unknown #Hx: Unknown #Sx: Unknown From the first line, we see that the recording number is A0001, and the recording file is A0001.mat. The recording has 12 leads, each recorded at 500 Hz sample frequency, and contains 7500 samples. From the next 12 lines, we see that each signal was written at 16 bits with an offset of 24 bits, the amplitude resolution is 1000 with units in mV, the resolution of the analog-to-digital converter (ADC) used to digitize the signal is 16 bits, and the baseline value corresponding to 0 physical units is 0. The first value of the signal, the checksum, and the lead name are included for each signal. From the final 6 lines, we see that the patient is a 74-year-old male with a diagnosis (Dx) of 426783006. The medical prescription (Rx), history (Hx), and symptom or surgery (Sx) are unknown. Each ECG recording has one or more labels from different type of abnormalities in SNOMED-CT codes. The full list of diagnoses for the challenge has been posted here as a 3 column CSV file: Long-form description, corresponding SNOMED-CT code, abbreviation. Although these descriptions apply to all training data there may be fewer classes in the test data, and in different proportions. However, every class in the test data will be represented in the training data.

We designed a baseline wander (BLW) removal benchmark to evaluate various methods using a consistent test set and uniform conditions. Details of the data preprocessing pipeline are heavily based on papers [1]. All 105 signals from the QT Database were resampled from 250 Hz to 360 Hz to align with the NSTDB sampling frequency. Heartbeats were extracted using the annotations provided by specialists. During this process, we identified a small number of incorrect annotations for beat start/end points, leading to cases where two consecutive beats were erroneously merged into one. To address this issue, we discarded beats exceeding 512 samples (1422.22 ms) in length. We designated heartbeats from 14 signals, accounting for 13% of the total signals, as the test set. These signals were selected to include two signals from each of the seven datasets comprising the QT Database, ensuring a diverse representation of pathologies in the test set. This setup provides a more robust evaluation of the generalization capability of the methods under consideration. Noise from the NSTDB, specifically BLW caused by breathing and electrode movement (referred to as the "em" signal in the database), was used to contaminate the ECG signals. The "em" record was split to match the length of the beat samples, with Channels 1 and 2 concatenated. Additionally, 13% of each channel’s signal length was preserved to contaminate the signals assigned to the test set. This separation of the test signals ensures the reliability of results when employing learning algorithms, preventing any unfair advantage over classical methods. For noise injection, we adopted the same approach as the NSTDB, randomly injecting noise with amplitudes ranging from 0.2 to 2 times the ECG signal’s maximum peak value. The mixed dataset can be downloaded from here: https://drive.google.com/file/d/19qOwywAoxreEv4xONTk-smQdo-ZdoPBc/view. [1] Romero, F. P., Piñol, D. C., & Vázquez-Seisdedos, C. R. (2021). DeepFilter: An ECG baseline wander removal filter using deep learning techniques. Biomedical Signal Processing and Control, 70, 102992.

BUAA-MIHR dataset is a remote photoplethysmography (rPPG) dataset. BUAA-MIHR dataset for evaluation of remote photoplethysmography pipeline under multi-illumination situations. We recruited 15 healthy subjects (12 male, 3 female, 18 to 30 years old) in this experiment and a total number of 165 video sequences were recorded under various illuminations. The experiments were conducted in a darkroom in order to isolate from ambient light.

HiRID is a freely accessible critical care dataset containing data relating to almost 34 thousand patient admissions to the Department of Intensive Care Medicine of the Bern University Hospital, Switzerland (ICU), an interdisciplinary 60-bed unit admitting >6,500 patients per year. The ICU offers the full range of modern interdisciplinary intensive care medicine for adult patients. The dataset was developed in cooperation between the Swiss Federal Institute of Technology (ETH) Zürich, Switzerland and the ICU. The dataset contains de-identified demographic information and a total of 681 routinely collected physiological variables, diagnostic test results and treatment parameters from almost 34 thousand admissions during the period from January 2008 to June 2016. Data is stored with a uniquely high time resolution of one entry every two minutes.

The MIMIC-IV-ECG module contains approximately 800,000 diagnostic electrocardiograms across nearly 160,000 unique patients. These diagnostic ECGs use 12 leads and are 10 seconds in length. They are sampled at 500 Hz. This subset contains all of the ECGs for patients who appear in the MIMIC-IV Clinical Database. When a cardiologist report is available for a given ECG, we provide the needed information to link the waveform to the report. The patients in MIMIC-IV-ECG have been matched against the MIMIC-IV Clinical Database, making it possible to link to information across the MIMIC-IV modules.

The second Ninapro database includes 40 intact subjects and it is thoroughly described in the paper: "Manfredo Atzori, Arjan Gijsberts, Claudio Castellini, Barbara Caputo, Anne-Gabrielle Mittaz Hager, Simone Elsig, Giorgio Giatsidis, Franco Bassetto & Henning Müller. Electromyography data for non-invasive naturally-controlled robotic hand prostheses. Scientific Data, 2014" (http://www.nature.com/articles/sdata201453). Please, cite this paper for any work related to the Ninapro database. Please, use also the paper by Gijsberts et al., 2014 (http://publications.hevs.ch/index.php/publications/show/1629) for more information about the database.

This dataset is created from MIMIC-III (Medical Information Mart for Intensive Care III) and contains simulated patient admission notes. The clinical notes contain information about a patient at admission time to the ICU and are labelled for four outcome prediction tasks: Diagnoses at discharge, procedures performed, in-hospital mortality and length-of-stay. To obtain the data one first has to gain access to the MIMIC-III dataset and then run the scripts introduced in the linked repository.

The MIMIC PERform Testing dataset contains the following physiological signals recorded from 200 critically-ill patients during routine clinical care: - electrocardiogram (ECG) - photoplethysmogram (PPG) - impedance pneumography (imp), also known as respiratory (resp) - (in some cases) arterial blood pressure (abp) Each signal is sampled at 125 Hz. The dataset also contains some fixed parameters for each subject (such as whether the subject was an adult or neonate). The dataset is available in CSV, Matlab, and WaveForm Database formats here. Further details of the datasets are provided in the documentation accompanying the ppg-beats project documentation. The dataset was extracted from the MIMIC III Waveform Database.

the MTHS dataset contains 30Hz PPG signals obtained from 62 patients, including 35 men and 27 women. The ground truth data includes heart rate and oxygen saturation levels sampled at 1Hz. The HR and SPo2 measurement is obtained using a pulse oximeter (M70). An iPhone 5s was used to obtain the ppg recordings at 30 fps.

a dataset of multi-modal signals from wearable devices at four sites on the body. Each device continuously recorded synchronized signals from a 3-channel reflective photoplethysmogram (red, green, infrared PPG), 3-axis inertial sensor (accelerometer), temperature, and barometric altitude sensor. For reference, the sternum device continuously recorded a Lead-I electrocardiogram (ECG) from body-mounted gel electrodes to provide ground-truth heart rate (HR) estimates.

Introduction The China Physiological Signal Challenge 2019 (CPSC 2019) aims to encourage the development of algorithms for challenging QRS detection and heart rate (HR) estimation from short-term single-lead ECG recordings usually with low signal quality and/or abnormal rhythm waveforms. ECG signal provides an important role in non-invasively monitoring and clinical diagnosis for cardiovascular disease (CVD). Detection of QRS complex is an essential step for ECG signal processing, and can benefit the following HR calculation and abnormal situation analysis. Although detection methods of QRS complex have been severely tracked throughout the last several decades, accurate QRS location and HR estimation are still challenging in noisy signal episode or abnormal rhythm waveforms, especially when the ECG recordings are from the wearable dynamic ECG acquisition. It is true that many of the developed QRS detection algorithms can achieve high accuracy (over 99% in sensitivity and positive predictivity) when tested over the standard ECG databases such as MIT-BIH Arrhythmia Database or AHA Database [1]. However, these algorithms may not be able to perform well when used in the daily life environment that will cause severe noises and significantly reduce the signal quality. A recent study confirmed that none of the common QRS algorithms can obtain 80% detection accuracy when tested in a common dynamic noisy ECG database [2]. Thus, in this challenge, we provide a new ECG database containing noisy ECG episodes and/or signals with different arrhythmia patterns, encouraging participants to develop more efficient and robust algorithms QRS detection and HR estimation. In addition, it is worth to note that, although HR can be calculated from the detection results of QRS complexes, HR can be still estimated without QRS detection step [3,4]. # Challenge Data Training data consists of 2,000 single-lead ECG recordings collected from patients with cardiovascular disease (CVD), each of the recording last for 10 s. Test set contains similar ECG recordings of same lengths, which is unavailable to public and will remain private for the purpose of scoring for the duration of the Challenge and for some period afterwards. ECG recordings were obtained from multiple sources using a variety of instrumentation, although in all cases they are presented as 500 Hz sample rate here. All recordings were provided in MATLAB format (each including two .mat file: one is ECG data and another one is the corresponding QRS annotation file). Pan &Tompkins (P&T) algorithm [5,6] is also provided as benchmark or comparable algorithm. Although QRS detection and HR estimation are widely studied by lots of researchers for many years, accurate detection is still really challenging in this Challenge due to the QRS amplitude variation, QRS morphological variation, and occurrence of intense variability in the intervals between beats, different arrhythmias, as well as noises. # Reference 1. G.B. Moody, R.G. Mark, The impact of the MIT-BIH arrhythmia database, IEEE Engineering in Medicine & Biology Magazine the Quarterly Magazine of the Engineering in Medicine & Biology Society, 20 (2001) 45-50. 2. Liu, F.F.; Wei, S.S.; Li, Y.B.; Jiang, X.E.; Zhang, Z.M.; Liu, C.Y., Performance analysis of ten common qrs detectors on different ecg application cases. Journal of Healthcare Engineering 2018, 2018, ID 9050812. 3. J.J. Gieraltowski, K. Ciuchcinski, I. Grzegorczyk, K. Kosna, Heart rate variability discovery: Algorithm for detection of heart rate from noisy, multimodal recordings, Computing in Cardiology, 2014, pp. 253-256. 4. J. Gieraltowski, K. Ciuchcinski, I. Grzegorczyk, K. Kosna, M. Solinski, P.Podziemski, RS slope detection algorithm for extraction of heart rate from noisy, multimodal recordings, Physiological Measurement, 36 (2015) 1743-1761. 5. P.S. Hamilton, W.J. Tompkins, Quantitative investigation of QRS detection rules using the MIT/BIH arrhythmia database, Biomedical Engineering, IEEE Transactions on, (1986) 1157-1165. 6. J. Pan, W.J. Tompkins, A real-time QRS detection algorithm, Biomedical Engineering, IEEE Transactions on, (1985) 230-236. 7. ANSI-AAMI (1998). Testing and reporting performance results of cardiac rhythm and st segment measurement algorithms, ANSI-AAMI:EC57.

HYPE Dataset - Version 1.0.0 REFERENCE PAPER ------------------- Morassi Sasso, A., Datta, S., Jeitler, M., Steckhan, N., Kessler, C. S., Michalsen, A., Arnrich, B., & Böttinger, E. (2020). HYPE: Predicting Blood Pressure from Photoplethysmograms in a Hypertensive Population. In M. Michalowski & R. Moskovitch (Eds.), Artificial Intelligence in Medicine. AIME 2020. Lecture Notes in Computer Science, volume 12299 (pp. 325–335). Springer, Cham. https://doi.org/10.1007/978-3-030-59137-3_29. CONTENT ------------------ - Sensor (PPG & Blood Pressure) and clinical data from 9 hypertensive subjects in two experiments (stress test and 24 hours) - PPG data from Empatica E4: photoplethysmography (PPG) data. - Spacelabs (SL 90217): blood pressure data. - Demographics and self-reported data during 24 hours (exercise, medication, etc.). AVAILABILITY ------------------ - Available to the scientific community through a data agreement. - The requester must be affiliated to a research institution.

This database includes 22 half-hour ECG recordings of subjects who experienced episodes of sustained ventricular tachycardia, ventricular flutter, and ventricular fibrillation. The reference annotation (.atr) files contain only rhythm labels (no beat labels).

The 5th Ninapro database includes 10 intact subjects recorded with two Thalmic Myo (https://www.myo.com/) armbands. The database can be used to test the Myo armbands separately as well. The 5th Ninapro database is thoroughly described in the paper: "Stefano Pizzolato, Luca Tagliapietra, Matteo Cognolato, Monica Reggiani, Henning Müller, Manfredo Atzori, Comparison of six electromyography acquisition setups on hand movement classification tasks, PLOS One, 2017" ### Acquisition Protocol The subjects have to repeat several movements represented by movies that are shown on the screen of a laptop. The experiment is divided in three exercises: 1. Basic movements of the fingers 2. Isometric, isotonic hand configurations and basic wrist movements 3. Grasping and functional movements During the acquisition, the subjects were asked to repeat the movements with the right hand. Each movement repetition lasted 5 seconds and was followed by 3 seconds of rest. The protocol includes 6 repetitions of 52 different movements (plus rest) performed by 10 intact subjects. The movements were selected from the hand taxonomy as well as from hand robotics literature. ### Acquisition Setup The muscular activity is gathered using 2 Thalmic Myo armbands. The database can be used to test the Myo armbands separately as well. The subjects in this database wore two Myo armbands one next to the other, including 16 active single–differential wireless electrodes. The top Myo armband is placed closed to the elbow with the first sensor placed on the radio humeral joint, as in the standard Ninapro configuration for the equally spaced electrodes; the second Myo armband is placed just after the first, nearer to the hand, tilted of 22.5 degrees. This configuration provides an extended uniform muscle mapping at an extremely affordable cost. The Myo sensors do not require the arm to be shaved and after few minutes the armband tighten very firmly to the arm of the subject. The sEMG signals are sampled at a rate of 200 Hz.

Facial electromyography recordings during both silent and vocalized speech. Source: Silent Speech EMG

This is a comprehensive dataset of 6,388 surgical patients composed of intraoperative biosignals and clinical information. The biosignal data included in the dataset is high quality data such as 500 Hz waveform signals and numeric values at intervals of 1-7 seconds. More than 60 surgery related clinical information is also provided to help interpret the signals. The dataset is provided free of charge to help researchers who want to study and develop new medical AI algorithms using monitoring signals from surgical patients. We expect that the distribution of this world's largest biosignal dataset will greatly contribute to the advancement of medical AI research.

Description: <a href="https://gts.ai/dataset-download/qr-code-detection-yolo/" target="_blank">👉 Download the dataset here</a> This dataset provides a comprehensive platform for enhancing QR code detection models, leveraging both synthetic and real-world data. Dataset Components Training Set: Includes thousands of auto-generated QR codes. These QR codes are isolated with no background, allowing the model to focus purely on shape and structure recognition. Test Set: Consists of real-life images featuring QR codes in varied contexts, from different angles, lighting conditions, and cluttered environments. Background Images: A curated set of images taken from malls, retail stores, streets, etc., allowing developers to blend QR codes into realistic scenarios, simulating real-world conditions for model training. Download Dataset Data Usage and Goals The primary objective is to train object detection models, such as YOLO (You Only Look Once), using synthetic QR codes and background images to improve generalization on unseen, real-world data. Training on the synthetic set avoids overfitting, while the test set evaluates the model’s performance in natural environments with QR codes placed in cluttered, busy scenes. Data Augmentation and Variety To make the dataset more robust, various augmentations are included: Diverse QR Code Sizes: Codes of varying sizes are use, enabling the model to detect small and large codes in the same setting. Rotations and Distortions: QR codes are display from multiple angles and subjected to distortions, mimicking real-life conditions where codes may not be perfectly align. Occlusion Handling: Some QR codes are partially obscured by objects, providing additional complexity for the detection task. Potential Applications This dataset is suit for AI developers working on: Mobile scanning apps Inventory management systems Augmented reality applications that rely on QR codes for object recognition. This dataset is sourced from Kaggle.