Smart Digital Stethoscope Chip Simulation Design

Abstract

Auscultation is an important key for the physical (respiratory and circulatory) examination and is helpful in diagnosing various disorders. Auscultation is performed for the purposes of examining the circulatory and respiratory sounds and gastrointestinal system (bowel sounds). Besides inconsistencies in the propagation of the normal sounds, there are also several types of specific irregularities that can be heard in respiratory sounds. However, detection of abnormal sounds during auscultation needs extensive training and experience. So, the separation of these heart sound signals (HSS) and the lung sound signals (LSS) is of great research interest. Heart sounds (HS) and lung sound (LS) separation is a challenging research task for respiratory specialists and cardiologists.Independent Component Analysis (ICA) over auscultation separation is a challenging signal processing problem. So in this study, we successfully evaluated and compared various performance parameters for heart and lung sound signal separation based on Independent Component Analysis (ICA) algorithms. The empirical results demonstrate the effectiveness of various ICA algorithms with a performance superiority over these reference techniques for various performance metrics.Currently, the growth of micro and nano (very large-scale integration-ultra large-scale integration) electronics technology has greatly impacted biomedical signal processing devices. These high-speed micro and nano technology devices are very reliable despite their capacity to operate at tremendous speed, and can be designed to consume less power in minimum response time, which is particularly useful in biomedical products like portable digital stethoscope. Stethoscope signal interpretation is difficult due to the presence of interference generated by the heart sound. This may lead to some noise and errors during the listening of lung sounds.The digital filters with fixed coefficients demonstrate acceptable performance to cancel the noise when the desired signal has variable stationary characteristics. However, when the desired signal has fluctuating characteristics, digital filters suffer performance degradation in eliminating the noise. To overcome the error rate at an optimum level, in this work, the authors proposed a novel non-linear artificial neural network (ANN) based adaptive line enhancer (least mean square (LMS) and normalized least mean square (NLMS)) architecture to separate the real-time auscultation sound signals effectively.Adaptive line enhancer (ALE) design with LMS filter and ALE design with NLMS filter are implemented in Verilog hardware description language (HDL) language to obtain both the network and adaptive algorithm in cadence Taiwan Semiconductor Manufacturing Company (TSMC) 90 nm standard cell library environment for ASIC level implementation. Native compiled simulator (NC) sim and RC lab were used for functional verification and design constraints and the physical design is implemented in Encounter to obtain the Geometric Data Stream (GDS II). In this ALE (LMS/NLMS) architecture, the area occupied is 0.08 m, the total power consumed is 5.05 mW and the computation time is 0.82 μs for ALE LMS design and the area occupied is 0.14 m, the total power consumed is 4.54 mW and the computation time is 0.03 μs for ALE-NLMS design.Finally, for our proposed ANN ALE LMS architecture, the area occupied is 0.12m, the total power consumed is 6.65 mW and the computation time of the proposed system is 0.3 μseconds. For ANN ALE NLMS architecture, the area occupied is 0.18m, the total power consumed is 6.16 mW and the computation time of the proposed system is 0.8 μseconds that will pave a better way in future electronic stethoscope design.The rapid technological scaling of the metal-oxide-semiconductor (MOS) devices aids in mapping multiple applications for a specific purpose on a single chip which motivates us to design a sophisticated, small and reliable application specific integrated circuit (ASIC) chip for future real time medical signal separation and processing (digital stethoscopes and digital microelectromechanical systems (MEMS) microphone).Keywords: