Adequate compression depth is a main quality parameter during cardiopulmonary resuscitation (CPR). Current CPR feedback devices can guide adult target depth which is fixed between 5 and 6 cm. For pediatric patients, conversely, target depth should be one third of the antero-posterior diameter of the chest. The aim of this study was to develop an algorithm to estimate chest diameter in pediatric patients using accelerometers. Using a tri-axial accelerometer, we measured the accelerations generated when moving the sensor from the floor to five different heights that simulated chest diameter. Five volunteers generated two records each per height. A total of fifty records were acquired. Chest diameter was measured by discrete integration of the z-axis acceleration signal. Velocity signal was band-pass filtered before computing the displacement signal. Chest diameter was identified as the displacement value at the instant in which the movement finished. Median (P<inf>25</inf>, P<inf>75</inf>) unsigned absolute and relative errors were 0.9 cm (0.3, 1.9) and 9.2 % (2.5, 14.6), respectively. Error in estimation of pediatric target compression depth was below 6.5 mm in 75 % of the cases. The proposed algorithm could be used to calibrate target chest compression depth in CPR feedback devices to be adapted for pediatric patients.