Digital auscultation simulator with neural network language model support for the educational process: approbation

Despite the intensive implementation of advanced diagnostic methodologies, training in the traditional skill of cardiac and pulmonary auscultation retains fundamental significance in the preparation of medical specialists, thereby stimulating the development and implementation of innovative digital simulation platforms integrated with neural network technologies to enhance the effectiveness of educational processes in medical training. Objective. To conduct an approbation of the developed digital auscultation simulator in a real academic setting, to evaluate its practical utility and to determine areas for further improvement. Materials and Methods. The simulator uses real audio recordings of normal and pathological respiratory and cardiac sounds, which are systematized according to clinical classifications. Two modes are implemented: a training mode with repeated listening and brief explanations, and a testing mode with multiple-choice questions and automatic scoring. The study included 20 physicians and 109 students in their 4^th to 6^th years, with 38 students undergoing repeated testing one week later. The primary outcome was the total score for recognizing respiratory and cardiac sounds. Non-parametric methods were used at a significance level of  p < 0.05. Results. No significant intergroup differences were found between physicians and students in recognizing either respiratory sounds (p = 0.526) or cardiac sounds (p = 0.822). In the subgroup of students who were retested, no statistically significant dynamics were noted for respiratory sounds, with the median changing from 6[6;7] to 7[5;8.75] with p = 0.729. However, a significant improvement in cardiac sounds was recorded, with the median changing from 2[2;2] to 3[3;4] with p = 0.002, confirming the simulator's effectiveness. Overall, physicians and students were better at recognizing pulmonary sounds than cardiac sounds. Users noted the ease of starting and mastering the simulator. Conclusion. The digital auscultation simulator is suitable for inclusion in the educational process, does not require expensive equipment, and provides short-term improvement in cardiac sound recognition among students, serving as an accessible supplement to traditional training. The study's limitations, including a single institution, a small sample size, and a short re-evaluation interval, necessitate the confirmation of results in larger and more randomized studies; further development is related to expanding the sound library and improving the evaluation algorithms. 

1. Bohadana A., Izbicki G., Kraman S. S. Fundamentals of lung auscultation. N Engl J Med. 2014; 370 (8): 744–751. doi: 10.1056/NEJMra1302901.

2. Kligfield P. The bicentennial of the stethoscope: 1816 to 2016. Am J Cardiol. 2016; 118 (10): 1601–1602. doi: 10.1016/j.amjcard.2016.08.033.

3. Pasterkamp H., Kraman S. S., Wodicka G. R. Respiratory sounds. Advances beyond the stethoscope. Am J Respir Crit Care Med. 1997; 156 (3): 974–987. doi: 10.1164/ajrccm.156.3.9701115.

4. Kim Y., Hyon Y., Lee S., Woo S.D., Ha T., Chung C. The coming era of a new auscultation system for analyzing respiratory sounds. BMC Pulm Med. 2022; 22 (1): 119. doi: 10.1186/s12890-022-01896-1.

5. Hafke-Dys H., Bręborowicz A., Kleka P., Kociński J., Biniakowski A. The accuracy of lung auscultation in the practice of physicians and medical students. PLoS One. 2019; 14 (8): e0220606. doi: 10.1371/journal.pone.0220606.

6. Pravkina E. A., Pereverzeva K. G., Budanova I. V., Yakushin S. S. Telemedicine: definition, features of implementation in practice, effectiveness and prospects in cardiology. Nauka Molodykh (Eruditio Juvenium). 2023; 11 (3): 435–446. doi: 10.23888/HMJ2023113435-446. (In Russ.).

7. Belova A. N., Kuznetsov A. N., Sushin V. O., Rezenova A. M., Shabanova M. A., Sheiko G. E., Ananyev R. D. Teleneurorehabilitation in neurological disorders and diseases: opportunities, effectiveness and barriers. I. P. Pavlov Russian Medical Biological Herald. 2024; 32 (1): 159–170. doi: 10.17816/PAVLOVJ364502. (In Russ.).

8. Glotov S. I., Uryasev O. M., Belskikh E. S., Ponomareva I. B., Kotlyarov S. N. Trenazher auskultatsii s integratsiei neirosetevoi modeli dlya opisaniya shumov [Auscultation trainer with a neural-network model for sound description]. Certificate of state registration of computer program No. 2025661319, Russian Federation; filed 2025-0422; published 2025-05-05 (In Russ.).

9. Murphy R. L. H. In defense of the stethoscope. Respir Care. 2008; 53 (3): 355–369.

10. Vyshedskiy A., Alhashem R. M., Paciej R., Ebril M., Rudman I., Fredberg J. J., Murphy R. Mechanism of inspiratory and expiratory crackles. Chest. 2009; 135 (1): 156–164. doi: 10.1378/chest.07-1562.

11. Doroshow R. W., Aldrich J., Dorner R., Lyons L., McCarter R. A randomized, controlled trial of an innovative, multimedia instructional program for acquiring auditory skill in identifying pediatric heart murmurs. Front Pediatr. 2024; 11: 1283306. doi: 10.3389/fped.2023.1283306.

12. Favrat B., Pécoud A., Jaussi A. Teaching cardiac auscultation to trainees in internal medicine and family practice: Does it work? BMC Med Educ. 2004; 4: 5. doi: 10.1186/1472-6920-4-5.

13. Zhang N. S., Yang J. Y., Goldhaber J. I., Phan B. A. P., Cheitlin M. D. Cardiac auscultation skills among medical trainees. Am Heart J. 2025; 286: 14–17. doi: 10.1016/j.ahj.2025.03.006.

14. Osborne C., Brown C., Mostafa A. Effectiveness of high- and low-fidelity simulation-based medical education in teaching cardiac auscultation: a systematic review and meta-analysis. Int J Healthc Simul. 2022; 1 (3): 75–84. doi: 10.54531/NZWS5167.