Non-Invasive Intelligence Of Bladder Filling Based On Electrical Impedance Research On Perceptual Platform Design

Urinary incontinence is a very common involuntary urinary leakage disorder caused by neurological disorders,and approximately 20% of the world’s population suffers from varying degrees of urinary incontinence.How to restore or improve bladder sensation in patients with urinary incontinence has become a hot topic of current research.Traditional bladder filling tests need to be guided and performed by a healthcare professional,and are severely lacking in affordability and portability.In this paper,we design a bladder filling prediction model based on the bioelectrical impedance changes of the human lower abdomen caused by the bladder filling process to achieve a more accurate and reliable prediction of bladder filling.The main works and results of this paper are as follows:(1)The mechanism of the electrical properties of human lower abdomen was investigated.The relationship between bioelectrical impedance and bladder filling in human lower abdomen was analyzed by studying the basic knowledge of bioelectrical impedance,such as the three-element equivalent model of biological cells and the dispersion principle.A high-fidelity 3D model of the human lower abdomen was constructed based on computed tomography(CT)images,and twodimensional and three-dimensional finite element simulation experiments were designed to compare and investigate the relatively optimal electrode placement in the human lower abdomen.By recruiting subjects and using professional LCR equipment for data acquisition,a 2000-group human lower abdominal bioelectric impedance dataset applicable to the bladder filling prediction model was constructed and preprocessed,and features were extracted for the next training task of the bladder filling prediction model.(2)The perceptual strategy of bladder fullness was investigated.A bladder fullness prediction model based on threshold comparison method,Full Connected(FC)network and Long Short Term Memory(LSTM)network was constructed,and the model performance was analyzed by comparison experiments,and the highest correct rate of 97.50% was obtained for the prediction model of LSTM network.(3)A non-interventional intelligent sensing platform for bladder filling based on electrical impedance was designed and developed.The overall design of this system platform was carried out for the functional requirements of the system,and the CS1259 chip was used as the core component of data acquisition for the hardware design,while the design of the upper computer We Chat applet was completed.The reliability and measurement accuracy of the system were evaluated through test equipment evaluation experiments,which proved that the designed system can accurately respond to the change of bladder filling degree through the change trend of bioelectric impedance of human lower abdomen.The simulation experiments were also evaluated by human experiments,which proved that the simulation experiments have good agreement with the human experimental results,and have guiding significance for finding the most sensitive electrode location in the human lower abdomen for bladder filling.