Penilaian Kebisingan Rumah Sakit Berbasis IoT dengan Pelabelan Self-Supervised Menggunakan Klasterisasi PSO-K-Means dan Spectral Flatness
Abstract
Kebisingan kebisingan di lingkungan rumah sakit dapat mengganggu kenyamanan pasien dan kinerja tenaga medis. Pemantauan kebisingan masih banyak dilakukan secara manual sehingga tidak mampu memberikan informasi yang akurat dan terus menerus. Penelitian ini menggunakan data audio untuk mengelompokkan tingkat kebisingan secara otomatis.
Penelitian ini semakin penting karena rumah sakit memiliki banyak sumber suara, seperti alarm alat medis, aktivitas staf, dan aktivitas operasional. Jika pola kebisingan tidak dikenali dengan baik, komunikasi bisa terganggu dan kualitas layanan kesehatan turut menurun. Oleh karena itu, dibutuhkan metode yang dapat membaca karakteristik suara secara objektif.
Penelitian ini mengusulkan model pengelompokan tingkat kebisingan menggunakan fitur spectral flatness, RMS, RMSE, SNR, dan dB sebagai fitur akustik. Proses optimasi dilakukan untuk menentukan konfigurasi klaster terbaik, diikuti dengan pengelompokan K-Means untuk menghasilkan kelompok akhir. Kontribusi penelitian ini berupa pemilihan fitur akustik, serta rancangan model yang dapat diintegrasikan sistem IoT untuk pemantauan kebisingan.
Hasil Menunjukkan bahwa konfigurasi tiga klaster merupakan pilihan paling optimal dengan nilai silhouette score sebesar 0.54. Visualisasi hasil klaster menunjukkan perbedaan masing-masing kelompok, dengan nilai SNR menjadi penanda utama tingginya tingkat kebisingan. Secara keseluruhan, model yang dibangun bekerja secara efektif dan dapat diadaptasikan dalam sistem pemantauan kebisingan rumah sakit.
Kata kunci: spectral flatness, internet of things, particle swarm optimization, K-Means clustering, self-supervised learning, kebisingan di rumah sakit
References
G. Chatzilampri and F. Stevens, "Are Noisy Hospitals Making Us Sick?," Frontiers for Young Minds, vol. 9, 2022.
E. e Lima Andrade, D. C. da Cunha e Silva, E. A. de Lima, R. A. de Oliveira, P. H. T. Zannin and A. C. G. Martins, "Environmental noise in hospitals: a systematic review," Environmental Science and Pollution Research, vol. 28, 2021.
A. S. Soni, "Exploring the Therapeutic Potential of Acoustic Interventions in Healing Environments," Indian Scientific Journal Of Research In Engineering And Management, 2024.
S. S. Nair, V. S. Meghashree and A. Mukhopadhyay, "Real-time Hospital Noise Detection and Classification Using IoT Devices, MFCC, and Artificial Neural Networks," in 2024 International Conference on Computing, Sciences and Communications (ICCSC), 2024.
W. B. W. Ismail, R. Rosdi and S. K. Zulkefli, "A Real-time Noise Detector for Patient Monitoring at Hospital Wards," in 2024 IEEE 14th Symposium on Computer Applications & Industrial Electronics (ISCAIE), 2024.
D. Hu, Z. Chen and F. Yin, "Frequency response analytical calibration based on spectral flatness for microphone arrays," IEEE Transactions on Instrumentation and Measurement, vol. 70, 2020.
E. R. Krishna, N. Devarakonda, M. Y. H. Al-Shamri and D. Revathi, "A novel hybrid clustering analysis based on combination of k-means and pso algorithm," in Data Intelligence and Cognitive Informatics: Proceedings of ICDICI 2021 , Springer , 2022.
Q. Pu, J. Gan, L. Qiu, J. Duan and H. Wang, "An efficient hybrid approach based on PSO, ABC and k-means for cluster analysis," Multimedia Tools and Applications , vol. 81, no. 14, 2022.
W. Yang, Y.-s. Liu and F.-m. Wu, "Unsteady Narrowband Spectrum Enhancement Method of Ship Noise Using PSO and Stochastic Resonance Processing," in 2024 OES China Ocean Acoustics (COA) , 2024.
A. A. Rafiq, S. D. Riyanto and H. Susanti, "Noise Detection System in The Classroom Using Sound Sensors and NodeMCU ESP6288," Journal of Electronics Technology Exploration, vol. 1, no. 1, pp. 1-14, 2023.
R. Rifqah, S. Suciyati, A. Surtono and A. Pauzi, "Design of a Classroom Noise Monitoring Tool Using a KY-037 Sound Sensor Based on Wemos D1R1," Journal of Energy, Material, and Instrumentation Technology, vol. 4, p. 125–135, 2023.
E. Segerstrom, L. Vanfretti, C. Mishra, X. Xu, K. D. Jones and R. M. Gardner, "Using Spectral Flatness to Detect and Label Power System Oscillations in the Presence of Intermittent Broadband Noise," in 2022 IEEE Power & Energy Society General Meeting (PESGM), 2022.
S. Harris and R. C. De Amorim, "An extensive empirical comparison of k-means initialization algorithms," IEEE Access, vol. 10, 2022.
S. Riyadi, A. S. Salsabila and A. R. P. Dewi, "Analysis of Music Genre Based on Similar Audio Features Using K-Means Algorithm: A Case of Spotify Top Hits," in 2024 International Conference on Information Technology and Computing (ICITCOM), 2024.
J. Devaney, "Digital Audio Processing Tools for Music Corpus Studies," in The Oxford Handbook of Music and Corpus Studies , Oxford Academic , 2023.
S. Ozsoy, S. Hamdan, S. Arik, D. Yuret and A. Erdogan, "Self-supervised learning with an information maximization criterion," Advances in Neural Information Processing Systems , vol. 35, 2022.
