Author :
Ms. Jeya Sudikshaa M M, Ms. Nivetha S, Ms. Priyanka R, Assistant Professor Mr. Mustafa Nawaz SJourna Name:
International Journal of Science, Engineering and Technology Country :
IndiaVolume:
12 issue:2 Year:2024 Views : 360
Abstract:
Abstract: In recent years, assistive solutions to difficulties in the medical arena have been made possible in large part by technologies like deep learning and machine learning. Through the use of audio analysis and medical imaging, they also increase the predictive accuracy for prompt and early disease detection. Given a scarcity of skilled human resources, medical professionals are grateful for this technological assistance since it helps them handle an increasing amount of patients. In addition to life-threatening diseases like diabetes and cancer, the impact of respiratory illnesses is progressively increasing and compromising people’s lives in society. X-rays of the chest and audio recordings of respiratory sounds have been shown to be extremely beneficial in the diagnosis and treatment of respiratory ailments since early detection and immediate intervention are crucial. The aim of the research that is being discussed is to use methods of deep learning based on convolutional neural networks in order to support physicians by providing a thorough and precise examination of medical respiratory audio recordings for identifying the presence of chronic obstructive pulmonary disease. We have used Librosa machine learning library features, comprising MFCC, Mel-Spectrogram, Chroma, Chroma (Constant-Q), and Chroma CENS, in the conducted evaluations. The severity of the diagnosed disease, such as mild, moderate, or acute, might also be determined by the system through which the diagnosis was given. The results of the study confirm that the specified deep learning strategy works. An increased 93% ICBHI score has been obtained for the system’s classification accuracy. Additionally, we utilized K-fold Cross-Validation with ten splits in the experiments to achieve the greatest the effectiveness of the deep learning strategy that was mentioned.
Chicago:Ms. Jeya Sudikshaa M M, Ms. Nivetha S, Ms. Priyanka R, Assistant Professor Mr. Mustafa Nawaz S. "Prediction of COPD Using Deep Learning" Example, Volume-12-issue-2-Year-2024-2348-4098. https://www.ijset.in/wp-content/uploads/IJSET_V12_issue2_642.pdf.