This research formulates and tests an AI-driven interactive e-book to increase English for Specific Purposes (ESP) learning by healthcare students in order to enhance speaking performance regarding fluency, pronunciation, vocabulary, grammar, and cohesion of content. Existing ESP materials in healthcare settings are non-interactive and too static to facilitate communicative practice in real time. Following a Research and Development (R&D) methodology, the research moved from needs analysis, prototyping of AI-based functionalities such as adaptive learning modules, NLP-based clinical dialogue simulation chatbots, and speech recognition to provide pronunciation feedback, to empirical testing. Thirty medical students completed pre- and post-tests, measured on a valid speaking rubric for the five performance areas and user perception surveys. Quantitative outcomes recorded considerable improvement across every area of speaking with maximum progress in fluency and vocabulary, while qualitative feedback indicated growing confidence and motivation for applying English for work purposes. The concluding AI-facilitated e-book, ISBN and copyrighted, provides a model for the scaling up of AI into domain language acquisition. The results have practical significance for ESP instructors and instructional designers who seek to integrate pedagogy and technology in ESP, particularly in healthcare training. Keywords: ai-integrated e-book, English for specific purposes, healthcare communication, speaking performance, interactive learning.