Accurate recognition of puncturing events in heterogeneous tissue is pivotal for improving the interaction between humans and robotic systems in biomedical applications, such as clinical process and surgical analysis. This study presents a novel Human-Robot Interface (HRI) framework that leverages Long Short-Term Memory (LSTM) networks to detect tissue penetration events in real time. The system captures time-series force feedback data via sensors embedded in a 1-DOF prismatic robot during tissue insertion. This feedback serves as a communication channel between the user and the robotic interface, enabling the LSTM model to interpret subtle force patterns indicative of transitions between tissue layers. By translating complex physical interactions into interpretable feedback for the operator, the proposed HRI system enhances situational awareness and control during procedures. Experimental results confirm the model effectiveness in providing accurate, real-time feedback, marking a significant advancement in intuitive and safe human-robot interaction for medical environments.