| | Phase | Description |
| | Recruitment | The recruitment of the participants was performed by means of publicizing of the trial process, involving a suitable number of people, through information campaigns, using posters, brochures and meetings. Subjects who met the criteria for inclusion were invited to participate in the process. |
| | Registration of participants | In this phase three documents were delivered to the participants. The first one was the information sheet to indicate the aim of the study, the procedures involved, the benefits and the possible risks of the examinations. Additionally, two forms were signed by the participants, the informed consent and the authorization for the processing of personal data, in order to protect the participants’ privacy. |
| | Medical phoniatric examinations | All recruited subjects were examined by medical experts involved in the study to diagnose the presence or not of a voice disorder. In accordance with the guidelines of the SIFEL protocol, they performed a scrupulous anamnesis to collect information relating to lifestyle (for example smoking status, alcohol consumption, voice use and professional activity) and previous or concomitant diseases, such as gastro-esophageal reflux, together with other data of interest to perform a correct diagnosis. Subsequently, the specialists performed a laryngoscopy, an invasive examination necessary to investigate the anatomical structure and any possible alterations of the larynx [44–46]. Figure 2 shows an image of a doctor performing this examination. The instruments employed were a Storz Laryngoscope, the 6.0 mm autoclavable 70° stiff model, used in most cases, and the 2.8 mm flexible model to perform laryngoscopies via the nose for more sensible subjects.
Finally, the participants were invited to compile two self-assessment questionnaires required by the SIFEL protocol, the Voice Handicap Index (VHI) [47] and Reflux Symptom Index (RSI) [48], to evaluate, respectively, the psycho-social consequences of any voice disorders and the presence of gastro-esophageal reflux. |
| | Vox4Health testing | The tests, conducted individually, were carried out in a quiet room (<30 dB of background noise) which was not too dry (humidity greater than 30-40%). The subjects recorded the vowel /a/ with a constant intensity, having been trained two or three times before the vocal acquisition. After the audio capture the estimation of each acoustic parameter was shown.
The app was installed on a Samsung Galaxy S4, Android version 5.0.1. The distance between the microphone of the mobile device and patient was about 20 cm with an inclination angle of about 45°.
All the voice samples were saved in Wave files, sampled at 8000 Hz, and their resolution was 32 bits, recorded in a mono-channel format. |
|
|
