IoT cihazlari i̇çin aktif gürültü önleme özellikli bağırsak sesi tespit algoritmasi

Internet of Things (IoT) enables a variety of applications in healthcare. A key application for IoT based technologies is wireless medical sensors that can be used to monitor patients' physiological information such as heartbeat, bowel activity, lung sound. Real-time detection of bowel motility after major abdominal surgery has significant importance in the patients' healing process. Because intestinal motility temporarily stops after the surgery, a period of fasting is commonly practiced, and patients are fed with fluids following the recovery of bowel motility. Many studies have been conducted to monitor intestinal motility and automatically detect bowel activity. But bowel activity detection suffers from ambient noise occurs in clinics. In this study, a bowel activity detection algorithm with active noise canceling feature was developed for our custom design IoT-driven electronic stethoscope, which was developed in our previous studies. This study's focus is to develop an effective active noise cancellation application by using both microphones of the electronic stethoscope. Experiments were conducted using both synthetic and real auscultation data. Five different adaptive filters were tested for active noise cancellation, and an ideal adaptive filter algorithm was determined. Then, a bowel activity detection algorithm was developed using real auscultation data. Active noise cancellation, bandpass filter, signal normalization, and Hilbert transform techniques were applied for signal enhancement. Bowel sounds were detected using an amplitude threshold value, which is determined from the generated ROC curve. The algorithm was tested on ten different auscultation recordings, and a sensitivity of 95.09% and a specificity of 95.83% were obtained in bowel sound recognition.