A Pressure Map Dataset for In-bed Posture Classification

Data Collection

In this project, we collected in-bed posture pressure data from multiple adult participants using two different types of pressure sensing mats. To the best of our knowledge, our dataset, PmatData is the first publicly-available dataset of pressure sensor data which includes various sleeping postures. PmatData contains pressure data from two separate experiments. These two experiments are:

• Experiment I: Pressure data from 13 participants in 8 standard postures and 9 additional states.
  • Data is collected using Vista Medical FSA SoftFlex 2048
  • Size of pressure mat is 32*64. This is the raw data collected reporting numbers in range of [0-1000] for each sensor.
  • Sampling rate is 1Hz.
  • Each file includes the data frames around 2-mins (around 120 frames).
• Experiment II: pressure data from 8 participants in 29 different states of 3 standard postures. Furthermore, experiment II was collected separately for both regular and air-alternating pressure mattresses.
  • Data is collected using Vista Medical BodiTrak BT3510
  • Size of pressure mat is 27*64. This is the raw data collected reporting numbers in range of [0-500] for each sensor.
  • Sampling rate is 1Hz.
  • Each file contains the average of around 20 frames.
  • The data is collected for both sponge and air mattresses separately.

Data Files

The experiment-info.docx file provides information about subjects and postures, including visual diagrams. There are also subject-info-*.csv files in each experiment subdirectory describing subject information.

The data files are provided in tab delimited text format, listed in subdirectories named after the experiment number and subject number.

For experiment i, the text files are named after the position number from 1-17. Each file contains multiple frames of recordings, with the 2048 columns representing the 32x64 pressure sensor measurements, and each row giving a new set of samples.

For experiment ii, the text files are named after the 29 position numbers. Each file contains one average of multiple frames of recordings. In contrast with the experiment i text files, the 27 columns and 64 rows represent the 27x64 sensor matrix.

Contributors

A team of researchers from Quality of Life Technology (QoLT) Laboratory at the University of Texas at Dallas created and contributed this database to PhysioNet. The contributors include Maziyar Baran Pouyan, Javad Birjandtalab, Mehrdad Heydarzadeh, Sarah Ostadabbas, Rasoul Yousefi, Masoud Farshbaf and Mehrdad Nourani (Faculty Advisor) and Matthew Pompeo, MD (clinical consultant).