A novel methodology for the experimental evaluation of pulmonary ventilator performance drift

An automatic system for the evaluation of pulmonary ventilator drift is proposed. Some statistical indexes commonly utilized in machine diagnostics, such as "Run test for randomness", are briefly described and utilized in drift tests in order to comparatively evaluate their sensitivity. Three ventilators were tested for 20 days by acquiring pressure and flow analog signals by means of an analog-to-digital converter in order to: (1) calculate main respiratory parameters; (2) save data onto magnetic support; and, finally, (3) postprocess the collected data for drift evaluation. The results reported here validate the proposed experimental set-up for objective ventilator performance drift evaluation and show the differing levels of effectiveness of the examined indexes. Therefore, the system can be considered a valuable tool for Clinical Engineering Service (CES) technicians, during maintenance procedures.