Published April 25, 2023 | Version 1.0
Dataset Restricted

A Dataset for Evaluating and Validating Blood Perfusion Monitoring During an Occlusion Protocol: Multi-spectral and Plethysmography data

  • 1. Universidad Autonoma de Aguascalientes
  • 2. Universidad Autonoma de San Luis Potosi

Description

The dataset contains multispectral images and PPG data from 45 subjects who underwent an occlusion protocol  to induce changes in blood perfusion in the dominant hand. All the partipiants signed an informed consent.

Blood perfusion

Blood perfusion refers to the passage of oxygen and other nutrients through the circulatory system. We look to provide data for the evaluation of non-invasive approaches based on multi-spectral images. PPG reference data from the thumb is provided for validation purposes.

Occlusion Protocol

The occlusion protocol for each participant lasts around 10 minutes.

The participants were seated, and their superior limbs were extended on a table.  An automatic blood pressure monitor measured their systolic and diastolic pressures. A blood cuff was placed in their dominant arm, and a PPG sensor MAX30102 (Maxim Integrated, CA, United States) was placed in their thumb finger.  A multispectral camera was positioned above to record the hand palm throughout the protocol. The camera employed is a CMS-V1-C-EVR1M-USB3 (SILIOS Technologies SA., France), which records 9 multispectral channels in the VNIR region of the spectrum.

.

The protocol is divided into 5 stages, each one lasts 2 minutes.

  1. Start: No pressure is applied
  2. Vascular Occlusion: a fixed pressure of 60 mmHg is applied through the blood cuff
  3. Rest: The pressure is liberated
  4. Total Occlusion:  A pressure equal to the initial systolic pressure plus 20 mmHg is constantly applied.
  5. Hyperemia: The pressure is released.

Data available:

The multispectral datasets for each subject are contained in a P#.7z file. Single-channel images are included. The name of each file specifies the participant ID, the time (minutes_seconds_milliseconds) of the acquisition with respect to the start of the occlusion protocol, the multispectral dataset, and the corresponding multispectral channel.

Example: the following file name specifies a file for participant number 21. The data was captured at 9 minutes and 59.5290 seconds after the start of the occlusion protocol. The image corresponds to the first channel (the count starts at 0).

P21_09_59_5290_d02459_channel0.png

Files containing masks for each dataset are available for the whole hand region (hand_masks.7z) and the middle finger (finger_masks.7z). Finger masks for participants with a lot of movement are not available.

The PPG data is stored in a simple csv file with txt extension for each participant. It contains the Red and Infrared channel data as well as the acquisition time of each sample. The PPG data for all the participants are contained in the compressed file PPG.7z.

Reference multispectral images for calibration are also provided in the file Calibration.7z. This file contains multispectral images with the cap on (Dark_Reference) plus three folders with reference images taken during the same session as the participants. 

reference_set_A:  = P1 to P10
reference_set_B:  = P11 to P30
reference_set_C:  = P31 to P45

each folder contains multispectral images of different white reference materials which can be employed for different calibration methods: 

WhiteA = white tile 
WhiteB = white reference bar 
WhiteP = sheet of white paper
WhiteT = Teflon (PTFEsheets

The participant’s data is summarized in the next table.

ID

Age

Gender

Skin type

Systolic Pressure

# Multi-spectral Datasets

P1

20

M

4

109

2,456

P2

20

M

3

129

2,458

P3

22

M

4

117

2,460

P4

19

F

3

110

2,454

P5

18

F

4

106

2,458

P6

18

F

4

111

2,471

P7

18

F

4

91

2,462

P8

18

F

4

102

2,474

P9

21

M

3

113

2,468

P10

19

F

3

108

2,461

P11

18

M

4

137

2,461

P12

18

F

3

132

2,457

P13

18

M

4

119

2,461

P14

18

F

4

100

2,460

P15

18

F

3

104

2,460

P16

18

M

3

126

2,462

P17*

18

F

3

105

2,469

P18

21

F

4

122

2,461

P19

21

F

3

130

2,462

P20

21

F

3

106

2,457

P21

23

M

3

133

2,459

P22

23

M

3

134

2,471

P23

21

F

3

115

2,472

P24

21

M

3

111

2,445

P25

21

M

3

116

2,460

P26

21

M

4

107

2,462

P27

21

F

4

106

2,460

P28

24

F

4

106

2,460

P29

22

M

4

125

2,460

P30

21

M

4

125

2,461

P31

21

F

3

100

2,472

P32

22

M

3

120

2,460

P33

21

M

4

127

2,461

P34

19

F

3

100

2,457

P35

19

F

3

108

2,460

P36

20

F

3

136

2,471

P37

19

M

4

129

2,457

P38*

19

M

3

128

2,463

P39*

18

M

3

111

2,479

P40

23

M

4

110

2,462

P41

21

M

3

130

2,465

P42

20

F

3

126

2,461

P43

22

F

3

125

2,459

P44

22

M

3

138

2,467

P45

22

M

4

104

2,461

Table 1.- Participant's ID and relevant information.Participants with an * presented a lot of movement during the recording.

 

 

Notes

Data formats MSI: png files PPG: csv files with txt extension

Files

Restricted

The record is publicly accessible, but files are restricted. Log in to check if you have access.

Request access

If you would like to request access to these files, please fill out the form below.

You need to satisfy these conditions in order for this request to be accepted:

For database access, please submit the following details: your name, organization, and a brief overview of your primary research topics. Admission will be granted to you promptly.

You are currently not logged in. Do you have an account? Log in here