DB3 Data

Subject Hand Handedness Age Height Weight Remaining Forearm (%) Years passed by the amputation Amputation cause Phantom Limb Sensation Intensity DASH Score Use of cosmetic prosthesis (Years) Use of kinematic prosthesis (Years) Use of myoelectric prosthesis (Years) Pre-Processed Files
01 Right Hand Amputated Right 32 172 86 50 13 Accident 2 1.67 0 0 13
02 Left Hand Amputated Right 35 183 81 70 6 Accident 5 15.18 6 0 0
03 Right Hand Amputated Right 50 178 82 30 5 Accident 2 22.5 0 8 8
04 Right Hand Amputated Right 34 166 68 40 1 Accident 1 86.67 0 0 0
05 Left Hand Amputated Left 67 175 75 90 1 Accident 2 11.67 0 0.4 0
06 Left Hand Amputated Right 32 172 66 40 13 Accident 4 37.5 0 12 0
07 Right Hand Amputated Right 35 185 75 0 7 Accident 0 31.67 0 0 6
08 Right Hand Amputated Right 33 175 80 50 5 Accident 2 33.33 0 0 4
09 Right Hand Amputated Right 44 180 95 90 14 Accident 5 3.33 0 0 14
10 Right Hand Amputated Right 59 177 86 50 2 Accident 5 11.67 0 1.66 0
11 Right Hand Amputated Right 45 183 75 90 5 Cancer 4 12.5 0 5 5

DB3 Guidelines


This Ninapro dataset includes sEMG, inertial, kinematic and force data from 11 transradial amputees while repeating up to 49 hand movements plus the rest position.


The dataset is described in detail in the following scientific paper.
Please, refer to it for detailed guidelines and cite it in case you use this dataset.



11 trans-radial amputees.

Acquisition Setup

The sEMG data are acquired using up to 12 Delsys Trigno electrodes, while kinematic data are acquired using a Cyberglove 2 data glove.

Acquisition Protocol

The dataset includes up to 10 repetitions of 49 hand movements. The subjects were asked to think to repeat as naturally as possible movements which were shown as movies on the screen of a laptop. The experiment is divided in three exercises:
  1. Basic movements of the fingers.
  2. Grasping and functional movements.
  3. Force patterns.
In the first two exercises, the subjects are asked to think to repeat as naturally as possible several movements represented by movies that are shown on the screen of a laptop; the kinematic information is recorded with a dataglove (cyberglove 2) and an accelerometer located on the wrist. In the last exercise the subjects were asked to think to press combinations of fingers with an iccreasing force on a custom made device called Finger Force Linear Sensor (Castellini 2012). The muscular activity is gathered using 12 active double differential wireless electrodes from a Delsys Trigno Wireless EMG system. The electrodes are positioned as shown in the figure: eight electrodes are equally spaced around the forearm in correspondence to the radio humeral joint; two electrodes are placed on the main activity spots of the flexor digitorum and of the extensor digitorum as described in; two electrodes are placed on the main activity spots of the biceps and of the triceps. The described locations have been chosen in order to combine a dense sampling approach with a precise anatomical positioning strategy. The electrodes were fixed on the forearm using their standard adhesive bands. Moreover, a hypoallergenic elastic latex free band was placed around the electrodes to keep them fixed during the acquisition. The sEMG signals are sampled at a rate of 2 kHz. During the acquisition, the subjects were asked to repeat the movements with the amputated hand. Each movement repetition lasted 5 seconds and was followed by 3 seconds of rest. Movements were selected from the hand movements taxonomy and robotics literature.

Dataset variables

For each exercise, for each subject, the database contains one matlab file with synchronized variables. The variables included in the matlab files are:

Acquisition procedure. The subjects are asked to mimic movies of movement shown on the screen of the laptop. sEMG signal is recorded with 10 electrodes (red, blue; the electrode on the flexor digitorum superficialis and part of the equally spaced electrodes are not represented due to perspective reasons).

Hand movements. Performed exercises: B, C, D.