A Multisensor-Based Diagnostic Device for Early Detection of Preterm Birth Risk in Low- Resource Settings

Background Preterm birth (PTB), defined as delivery before 37 weeks of gestation, remains a leading cause of neonatal morbidity and mortality worldwide, especially in low-resource settings. In Ethiopia alone, over 320,000 babies are born prematurely each year, with more than 24,000 deaths attributed to PTB-related complications. Objective This study presents the design and validation of a multisensor-based diagnostic device for early detection of PTB risk by monitoring three key indicators: intra-amniotic infection (chorioamnionitis), bacterial vaginosis, and maternal stress. Methods The device integrates a color sensor for pH analysis, a pulse sensor for heart rate monitoring, an infrared temperature sensor for fever detection, and a galvanic skin response (GSR) sensor to assess stress levels. A keypad interface allows users to input responses to clinical prompts regarding uterine fundal tenderness and foul-smelling vaginal discharge. Data acquisition is managed via an Arduino Mega microcontroller, with a multilingual LCD interface supporting user interaction. The system was simulated using Proteus and AutoCAD and validated across multiple prototypes. Results The final prototype demonstrated high accuracy in detecting physiological changes associated with PTB risk. Temperature readings showed 99.7% accuracy, GSR precision was ± 660 ohms, and pH detection correlated strongly with digital pH meters. The device is portable, cost-effective (less than 5,000 Ethiopian Birr), and suitable for both clinical and community settings. Conclusion This novel diagnostic tool offers a promising solution for early identification of PTB risk in low-resource environments. Its affordability, simplicity, and multilingual support make it suitable for deployment in rural health posts and home settings, with potential to reduce neonatal mortality and improve maternal care.