Acoustic BioPower Training in Aging Dataset

Experimental Design This dataset was collected as part of a 6-week intervention designed to investigate the effects of different auditory stimuli on exercise performance in older adults. Participants were allocated to one of two auditory stimulus groups: Acoustic Biofeedback (ABF) or Gymnasium Music (GM). Throughout the training sessions, participants listened to their respective auditory stimulus while performing multiple repetitions of a predetermined load of each exercise, at the maximum velocity in the concentric phase of the exercise. Physiological and kinetic data were collected throughout the intervention to characterize participants' muscle activity and exercise performance. At baseline, participants underwent an initial assessment that included the measurement of handgrip strength, body composition, and height, as well as the completion of the International Physical Activity Questionnaire (IPAQ) and the Hearing Handicap Inventory for the Elderly-Screening (HHIE-S) questionnaire. An estimate of the one-repetition maximum (1-RM) was also obtained for each exercise. To estimate the 1-RM, participants performed the bench press and squat exercises using progressively increasing loads. The load was increased until the participant was unable to perform more than 10 repetitions. Therefore, the 1-RM was estimated based on the maximum load that the participant could successfully perform for no more than 10 repetitions. The same assessment procedure was repeated at endline following completion of the intervention. The training intervention consisted of eight muscle power-training sessions performed using the Speediance® GYM Monster 2 Digital Weight Machine (Shanghai, China) to provide resistance. Participants were allocated to one of two auditory stimulus conditions: Acoustic Biofeedback (ABF) or Generic Gym Music (GM), and listened to their assigned auditory stimulus throughout the training sessions. During each training session, participants performed the bench press, squat, and dorsal row exercises using predetermined loads based on their individual 1-RM. Multiple repetitions were performed at the same relative load, with participants instructed to perform the concentric phase of each repetition at maximum intended velocity. A total of 23 participants (9 males and 14 females; age = 70.12 ± 0.92 years) were recruited from a Local Senior University. Data collection To collect physiological and kinematic signals during the power training, an 8-channel biosignalsplux wireless hub device with a 16-bit resolution (PLUX Wireless Biosignals, Lisbon, Portugal) was used at a sampling rate of 1000 Hz. Physiological signals included muscle activity, recorded using four electromyography (EMG) sensors (PLUX Wireless Biosignals S.A., Portugal) with Ag/AgCl adhesive electrodes. The electrodes were placed according to standard anatomical landmarks for each target muscle: Triceps Brachii: electrodes were positioned at the midpoint of the line between the posterior aspect of the acromion and the olecranon process, approximately two finger widths medial to this line. Biceps Brachii: electrodes were positioned along the line between the medial acromion and the cubital fossa, at approximately one-third of the distance from the cubital fossa. Rectus Femoris: electrodes were positioned at the midpoint between the anterior superior iliac spine and the superior border of the patella, oriented parallel to the line connecting these anatomical landmarks. Biceps Femoris: electrodes were positioned at the midpoint of the line between the ischial tuberosity and the lateral epicondyle of the tibia. Kinematic signals were acquired using a triaxial accelerometer (ACC) (PLUX Wireless Biosignals S.A., Portugal), while force data were recorded using a load cell (PLUX Wireless Biosignals S.A., Portugal). Both sensors were attached to the barbell of the weight machine, with the accelerometer positioned so that its Y-axis (ACCy) was aligned with the direction of movement.All these signals were acquired only from the right side (unilateral measurements) that was the dominant hand's side for all the participants in this study. Data Structure and Description The main folder, Data Acoustic BioPower, is organized into four main data folders: Baseline_Data, ABF, GM, and Final_Data. The Baseline_Data and Final_Data folders contain the baseline and endline measurements, respectively, for each participant. The measurements in these folders were performed using loads corresponding to predefined percentages of each participant's previously determined one-repetition maximum (1-RM). For the bench press, the loads correspond to percentages ranging from 30% to 70% of the individual 1-RM, whereas for the squat, the loads range from 40% to 80% of the individual 1-RM. The ABF and GM folders contain the experimental data collected during the ABF and GM conditions, respectively. Within these folders, the data are organized according to participant ID and, for the experimental data, according to the different training sessions (T1–T8). The main folder also contains two text files: ACC_calib.txt, containing the accelerometer calibration data, files_notes.txt containing notes about specific files in the dataset. and subject_info.txt, containing the characterization and relevant outcome measures for each participant. The subject information includes the participant ID, stimulus condition (ABF or GM), sex (1 – Male; 2 – Female), age (years), body mass (kg), height (m), global body fat (%), global muscle (%), right arm muscle mass (kg), right leg mucle mass (kg), baseline and endline one-repetition maximum (1-RM) values for the bench press and squat (kg), baseline and endline handgrip strength (kg), IPAQ level, and HHIE-S score. A schema of the dataset organization is bellow: Acoustic BioPower Training in Aging Dataset │├── Baseline_Data│ ││ ├── ID1│ │ ├── Baseline_40_SQ_BP.txt│ │ ├── ...│ ││ └── IDy│ └── ...│├── ABF│ ││ ├── ID1│ │ ││ │ ├── T1│ │ │ ├── ID1_T1_BP_rep1.txt│ │ │ ├── ID1_T1_SQ_rep2.txt│ │ │ └── ...│ │ ││ │ ├── T2│ │ │ ├── ID1_T2_BP_rep1.txt│ │ │ ├── ID1_T2_SQ_rep1.txt│ │ │ └── ...│ │ ││ │ ├── ...│ │ ││ │ └── T8│ │ └── ...│ ││ ├── IDy│ │ └── (same structure)│├── GM│ ││ ├── ID2│ │ ││ │ ├── T1│ │ │ ├── ID2_T1_BP_rep1.txt│ │ │ ├── ID2_T1_BP_rep2.txt│ │ │ └── ...│ │ ││ │ └── T8│ │ └── ...│ ││ ├── ID6│ │ └── (same structure)│ ││ └── ...│├── Final_Data│ ││ ├── ID1│ │ ├── Final_40_SQ_BP.txt│ │ └── ...│ ││ ├── ...│ ││ └── ID26│ └── ...│├── Subject_info.txt│├── Files_notes.txt│└── Acc_calib.txt For all the recordings the columns should be: Nseq - Number of SequenceLOADCELL - data from the loadcell sensor.ACCx - data from the x axis of the accelerometer sensor.ACCy - data from the y axis of the accelerometer sensor.ACCz - data from the z axis of the accelerometer sensor.EMG_BB – EMG data from the Biceps Brachii muscle.EMG_TB – EMG data from the Triceps Brachii muscle.EMG_RF – EMG data from the Rectus Femoris muscle.EMG_BF – EMG data from the Biceps Femoris muscle. Data Usage and Processing All the sensor data is in ADC values and not in physical units, to transform these please check the datasheet available for each sensor: Sensor Datasheet EMG https://support.pluxbiosignals.com/wp-content/uploads/2021/10/biosignalsplux-Electromyography-EMG-Datasheet.pdf ACC https://support.pluxbiosignals.com/wp-content/uploads/2021/11/Accelerometer_ACC_Datasheet.pdf For the Accelerometer, a Callibration procedure was followed according to the ACC datasheet, the callibration values can be extracted from the Acc_callib.txt file. For the loadcell, the following equation was used after sensor callibration using known weights: Force = -0.003x + 96.6337 (kg) where x is the ADC value.An example of data usage follows, a code snippet for data extraction. # Loading bench press data from subject ID3, Session 2, at 40% of 1-Repetition maximum, # under the ABF Stimulus. file_path = "ABF\ID5\T2\ID5_T2_BP_rep1.txt" with open(file_path, 'r') as f: lines = f.readlines() # Columns header_line = lines[0].strip("\t") column_labels = header_line.split() # Data data_lines = [line.strip().split("\t") for line in lines[1:]] df = pd.DataFrame(data_lines, columns=column_labels) for c in list(df.columns): df[c] = pd.to_numeric(df[c], errors="coerce") Ethics The study protocol was approved by the Ethics Committee of the NOVA School of Science and Technology (CE_FCT_021-2025). Informed consent was obtained from all subjects involved in the study. Funding This work was done under the ASTROPOWER project that explores the analogy between the effects of microgravity and aging. The ASTROPOWER project is financed by the European Space Agency through the Programme de Développement d'Expériences scientifiques (PRODEX). D.Furk was supported by the individual doctoral grant 2025.00753.BD financed by the Portuguese Foundation for Science and Technology (FCT).