Introduction

Nowadays auscultation is still the first and most common examination carried out by every general practitioner (GP) or family doctor. Most of the diagnoses there are made based on it. It is fast, easy and does not need advanced technology. Nevertheless, one must keep in mind that auscultation is a medical examination that has been known since the time of Hippocrates. Its popularity was made possible thanks to Leannec, who invented the stethoscope in 1816. His invention was a rigid, cylindrical-shaped wooden headpiece with a central groove (as a sound tube) for listening to the lungs and heart. Thanks to this solution the doctor no longer had to put his ear to the patient's body, and examination became simpler and more hygienic. Apart from some minor changes, the form of the acoustic stethoscope has survived to this day. One  modification was the introduction of bowls on both sides to adjust acoustic impedance, thereby increasing the volume. The British physician Bird (1840) introduced an elastic tube that connected the stethoscope chestpiece with a single earpiece. Since 1964 (Littmann, 1964) the stethoscope has also been equipped with an acoustic diaphragm. There are also various types of electronic stethoscopes nowadays, which operate different in different ways. They have a transducer (a piezo microphone) that converts the vibration on the chest surface to an electrical signal that is then reproduced to the physician via headphones and can be recorded for later analysis. Regardless of the type of stethoscope, the most important features of auscultation are non-invasiveness, simplicity, ease to carry out, and the low cost associated with the device. However, the results of this examination are subject to a relatively large rate of error due to the subjectivity of the assessment and the influence of many additional factors. First among these are  the experience of doctors and their perceptual abilities. Physicians often differ in their assessments. Mangione and Nieman (1999) examined the pulmonary auscultatory skills of medical students, pulmonologists and interns in internal medicine and family practice. Research has shown that internal medicine and family practice doctors have not been statistically superior to medical students. Only pulmonologists have achieved statically better results than others.
Furthermore, differences also arise from the different nomenclature used and the division of different sound classes in the existing medical literature. This is an international problem, as has been shown by numerous authors (Bunin et al., 1979; Cugell, 1987; Francis et al., 2013; Wilkins et al., 1990). Pasterkamp et al. (2016), proposed a unified nomenclature for 6 languages and suggested a standardized terminology. Unfortunately, it is not wccepted worldwide in education nor everyday practice. As a consequence, there is still no uniformly standardized classification of the types of phenomena characteristic for the human respiratory system in the whole medical environment. Depending on the handbook and university, physicians often use other words with a completely different semantic meaning to describe the perceived pathology in the respiratory system. This leads to problems in preliminary auscultation courses, but primarily during later professional work, when doctors exchange or consult diagnoses, as they use different descriptions of the same sounds and semantically similar or  even identical terms to describe different types of phenomena, which results in ambiguous descriptions or even makes the descriptions incomprehensible to other doctors, and the examination must be repeated. It is also worth noting that the sounds  are not stored and there is no way to go back to the recordings or compare them to other sounds, so there is no way of verifying the description, which increases the ambiguity even more.

Aim

The goal of this work is to answer a few problems. First, the questionnaire was ascertain whether the medical community is aware of the problem of auscultation sounds classification, how they evaluate education in the area of auscultation, and how they assesses their abilities in it. The main goal of this paper, however, is to answer the question: How correctly and consistently do physicians and medical students evaluate respiratory sounds, and do they categorize them in the same way? In this context, other aspects of the problem have also been analyzed in details: Do pulmonologists perform better than other groups of physicians? How does a group of students fit into this community? And finally we try to conclude how to solve the problem of low efficiency of doctors by creating respiratory signal database and unifying nomenclature. This problem is especially crucial because most of patients that feel ill go to the doctor’s office where they are examined by GPs or family doctors and the diagnosis and further treatment are mainly based on auscultation. The study was approved by the Bioethical Commision of the Poznań University of Medical Science.

Method

The test was distributed among the academic medical community and in hospitals. It contained both signals and question about specialization of the participants. The test was anonymous and was conducted online via the Internet using Questionpro Professional. This program was chosen because enables uncompressed and high quality audio to be presented. Moreover, before the experiment was made available to the participants, the quality of the signals in this software was subjectively verified by two experienced acousticians (sound engineers) independently (without hearing loss) in terms of distortions and possible artifacts. No difference was found between direct and on-line listening. In addition, to minimize the possibility of a layperson completing the survey, the survey was distributed among the academic medical community and in hospitals. It was also passed on to interested people through lecturers at medical universities and through direct contact with doctors asking them to complete the test.
The experiment consisted of two parts.