This work presents a novel hybrid deep learning approach for the classification and identification of medicinal leaves, integrating Convolutional Neural Networks (CNNs) and Long Short-Term Memory networks (LSTMs). Medicinal plants are invaluable resources for traditional medicine, and automating their identification process holds significant potential for advancements in healthcare and biodiversity conservation. The proposed approach combines the spatial feature extraction capabilities of CNNs with the sequential learning abilities of LSTMs to enhance the model's understanding of complex leaf structures and variations. The CNN-LSTM hybrid model processes high-resolution leaf images, capturing both local and global contextual information. The sequential nature of LSTMs enables the model to recognize temporal patterns, crucial for distinguishing subtle differences in leaf structures and textures. Experimental evaluations on a diverse dataset of medicinal leaves demonstrate the effectiveness of the hybrid model, outperforming traditional deep learning architectures. The proposed approach contributes to the field of botanical research, providing a robust tool for accurate medicinal leaf classification and identification, thereby facilitating advancements in medicinal plant-based healthcare.