An ultrasonic sensing system for colonic content discrimination in ostomy patients: Development and experimental validation

Background Enterostomy leads to excretion incontinence, complications, and impaired quality of life, while existing interventions lack colonic content sensing capabilities. Objective The aim of this validation-based study was to develop and evaluate the technical performance of a smart ultrasonic sensing system for discriminating colonic content states to improve ostomy management. Method A modified 2 MHz reflective ultrasonic sensing system was developed, integrating an ultrasonic probe, transceiver circuit, signal amplification module, single-chip microcomputer, RGBW LEDs, and a buzzer alarm. Functional testing was performed under three immersion conditions (non-immersed, semi-immersed, and fully immersed) in normal saline. Voltage thresholds were established in vitro using excised colonic segments from 20 Oryctolagus cuniculus ( O. cuniculus ), with voltage signals measured under four colonic content states (gaseous, solid, fluid, and empty). In vivo validation was then conducted in 20 O. cuniculus and 12 Canis lupus familiaris ( C. lupus familiaris ). Data were analyzed using one-way ANOVA with LSD post-hoc tests (SPSS 27.0, p < 0.05). System performance was evaluated by the concordance between LED and buzzer outputs and reference colonic content states. Results Voltage output increased progressively with immersion depth in normal saline: 343.85 ± 2.11 mV (non-immersed), 637.60 ± 8.01 mV (semi-immersed), and 943.35 ± 3.86 mV (full- immersed) ( p < 0.001). In vitro, distinct voltages were recorded across states: gaseous (257.12 ± 10.51 mV), solid (398.12 ± 32.33 mV), fluid (507.24 ± 25.01 mV), and empty (658.56 ± 14.85 mV) (all p < 0.001). In vivo, the system achieved 98.0% overall accuracy in O. cuniculus (392/400) and 97.0% in C. lupus familiaris (419/432), with the solid state showing the lowest accuracy. Conclusion This study developed an integrated ultrasonic sensing system for artificial anal plugs, which achieved high classification accuracy and can provide reliable real-time feedback on colonic content, demonstrating strong potential for artificial anal plugs.