Vibration Response Imaging in the Diagnosis of Pulmonary Disease

Vibration Response Imaging (VRI) is novel technology which records breath sounds via pizo-electric sensors and produces a digital image using a computer algorithm. It is radiation free and is portable to the patient's bedside. Data exists to show that the recordings from normal individuals differs from those who have pulmonary pathology. There is also evidence that recordings have high levels of inter and intra-observer reliability. However, data on specific VRI patterns for specific pathology is still needed before this can be used as a diagnostic tool. We aim to perform an open label feasibility trial on inpatient and outpatient pulmonary patients. Bedside clinical examination and chest auscultation will be used as the reference gold standard. Other diagnostic modalities that have been used as part of the patient's usual standard of care will also be used for comparison. Specifically breath sound progression, the maximal sound energy shape/distribution and the presence of artifactual sounds will be used to search for patterns that may be used for diagnosis. Sensitivity and specificity will be calculated for each disease (eg. asthma, emphysema, bronchiectasis, pneumonia, effusion, pneumothorax, etc)

Introduction: The measurement of regional distribution of breath sound intensity within the human thorax during respiration using computerized lung sound analysis has generated interest as a possible diagnostic modality. Using normal breath sound distribution as a reference, abnormal distributions of sound energy can be used to identify potential pathological states. Vibration Response Imaging (VRI) utilizes such technology. VRI processes lung sound recordings and generates a dynamic representation of this sound energy distribution on a computer screen using a grey-scale coding. Air turbulence within the airways and the consequent vibrations that are generated are captured by multiple piezoelectric contact sensors, which have been placed on the skin of the patient. The signals are processed by band-pass (150-250 Hz) filtering to eliminate heart sounds and chest wall movements. These filtered signals are then combined, sliced over 0.17 second intervals, and converted into a sequence of dynamic images that reflect regional sound vibration energies against a time axis. Lung vibration energy is presented on the screen in various grey shades from black in maximum energy areas to white in zero energy areas. Pilot data suggest that VRI can distinguish breath sounds in normal patients from those with intra-thoracic pathology based on distinctive patterns of vibration energy. Reproducibility of VRI recordings and consistency of inter-observer interpretation have also been documented. Aim: To assess the diagnostic capability of Vibration Response Imaging (VRI) in the evaluation of pulmonary disease (Subjects presenting as pulmonary inpatients and outpatients). Chest auscultation and other diagnostic modalities that have been used as part of standard of care (eg. chest radiograph, CT scans and thoracic ultrasound) will be used as the reference gold standard to compare against. Methods: All adult (≥ 21 years old) patients (inpatients and outpatients) under the care of the Department of Respiratory and Critical Care Medicine between 1/07/2008 and 31/05/2009 will be considered eligible. Children may have too small body sizes for the current standard sensors and will not be recruited. Other inclusion criteria will be ability to provide informed consent. Exclusion criteria: Conditions that will prevent the placement of sensors oh the patients back such as bony/chest wall deformity and contagious skin conditions. The presence of a pacemakers and pregnancy are also considered contraindications because of the yet undefined safety issues associated with these conditions. Patients who have signed informed consent will be asked to sit upright with their back exposed. Those unable to sit up will have the sensors slid under them while in the supine position. Recordings will be then performed using the VRI device (Deep Breeze™, Or Akiva, Israel). Forty-two sensors that are assembled on 2 planar arrays will be placed on subject's back. Each row of 3 sensors is held in place by silicone cups that are coupled to the patient's back by a computer-controlled low vacuum seal. Prior to the application of the sensors, the patient's back will be wiped clean with an alcohol-free wet wipe to remove perspiration and dirt debris. Patients will be recorded for 12 seconds while asked to breathe deeply in and out. Forced exhalations are discouraged. Each 12-second recording usually include 3-4 respiratory cycles. Most patients need to be coached on the breathing technique for 1-2 minutes by the recording physician. An average of 2-3 recordings will be made at each setting to ensure that a reproducible image could be attained. The principal investigator was trained in the recording and interpretation at Harvard University. The co-investigators have been trained by the principal investigator and are familiar with the machine because it has been on trial with the department. The VRI recordings will be interpreted by the study doctors and by the Deep Breeze (Company manufacturing VRI equipment) scientist. None of the patient's personal or clinical information will be released to Deep Breeze Pte Ltd. No additional tests or investigations will be performed on the patients. They will receive management as per standard of care. The VRI recordings will be compared against standard physical examination and diagnostic investigations that the patients would have had anyway. Therefore, patients will not be paid any monetary compensation for their participation in the trial. Data management Information that will be collected is described in the Case Report Form (Appendix A). All clinical data will be stored in the locked department office of the Principal Investigator for a period of 2 years after the completion of the trial for the purpose of data analysis. The patient's VRI recordings will remain on the VRI machine which is also password protected. For the purpose of analysis, the recordings may be downloaded on to CDs and sent to Deep Breeze. These recordings will contain no personal information except de-identified subject code numbers. Statistics As this is an exploratory trial to test the feasibility of a novel technology, there is no basis for sample size collection. An arbitrary figure of 200 was chosen because this was felt to be achievable within the trial period. We will attempt to correlate the VRI findings with the physical examination and other diagnostic modalities for individual disease states depending on the clinical diagnosis (e.g. asthma, bronchiectasis, pneumonia, emphysema, heart failure, effusion, pneumothorax) with the hope of finding reliable patterns that will enable a diagnosis solely based on VRI. Sensitivity and specificity will be calculated for each condition

Recordings will be then performed using the VRI device (Deep Breeze™, Or Akiva, Israel). Forty-two sensors that are assembled on 2 planar arrays will be placed on subject's back. Each row of 3 sensors is held in place by silicone cups that are coupled to the patient's back by a computer-controlled low vacuum seal.

Inclusion Criteria: All adult ( ≥ 21 years old ) patients (inpatients and outpatients) under the care of the Department of Respiratory and Critical Care Medicine between 1/07/2008 and 31/05/2009 will be considered eligible. Children may have too small body sizes for the current standard sensors and will not be recruited. Other inclusion criteria will be ability to provide informed consent. Exclusion Criteria: Conditions that will prevent the placement of sensors oh the patients back such as bony/chest wall deformity and contagious skin conditions. The presence of a pacemakers and pregnancy are also considered contraindications because of the yet undefined safety issues associated with these conditions.

Maher TM, Gat M, Allen D, Devaraj A, Wells AU, Geddes DM. Reproducibility of dynamically represented acoustic lung images from healthy individuals. Thorax. 2008 Jun;63(6):542-8. doi: 10.1136/thx.2007.086405. Epub 2007 Nov 16.

Dellinger RP, Parrillo JE, Kushnir A, Rossi M, Kushnir I. Dynamic visualization of lung sounds with a vibration response device: a case series. Respiration. 2008;75(1):60-72. doi: 10.1159/000103558. Epub 2007 Jun 4.