Résumé: Poor feminine hygiene leads to various infections such as hepatitis B, fungal infections, reproductive tract infections, urinary infections, etc. The main concern associated with the menstrual cup is the problem of disposal and hygiene management which is still a major concern in the whole world. To overcome this problem, this paper proposed a novel technique integrating the Internet of Things (IoT) and machine learning methodology. The chaos game (CG) optimized adaptive convolutional neural network (ACNN) architecture is used to detect the menstrual flow when it exceeds a specific mark (17 ml) in the menstrual cup. When the menstrual cup is prone to leakage the participant is alerted via a text message. The efficiency of the proposed methodology is evaluated by using the voltage data obtained from the different IoT devices of participants. The efficiency of the CG optimized ACNN architecture is evaluated using different performance metrics such as accuracy, Root means Square Error, sensitivity, specificity, and recall. The proposed methodology is both capable of identifying the abnormal menstrual flow and alerting the user for disposal. Our experimental findings illustrate the efficiency of this methodology in improving women’s menstrual hygiene in real time by analyzing the flow.

Langue: Anglais