In this paper, we presented an efficient deep learning based approach to extract technology-related topics and keywords within scientific literature, and identify corresponding technologies within patent applications. We illustrated the workflow as well as results obtained by mapping publications within the field of neuroscience to related patent applications, aiming at the mapping of neurotechnology, and particularly the identification of emerging ones. Specifically, we utilize transformer based language models, tailored for use with sci- entific text, to detect coherent topics over time and describe these by relevant keywords that are automatically extracted from a large text corpus. We identify these keywords using Named Entity Recognition, distinguishing between those describing methods, applications and other scientific terminology. These topics are created via density- based clustering of transformer based embeddings, which are fine-tuned to scientific literature. In contrast to traditional topic modelling techniques, our approach pro- duces topics focused on the description of technologies and their applications rather than general themes in the corpus. We create a large amount of search queries based on combinations of method- and application-keywords, which we use to conduct semantic search and identify related patents. By doing so, we aim at contributing to the growing body of research on text-based technology mapping and forecasting that leverages latest advances in natural language processing and deep learning. We demonstrate at the case of neuroscience research, that the developed approach is able to extract technology topics in broad, interdis- ciplinary, and dynamic field, and map these topics to patent data. Enabling the semi-automatized mapping of technologies identified in scientific literature to patent applications, we are thereby providing an empirical foundation for the study of science- technology linkages. The presented method as well as the obtained preliminary results are at its current stage subject to a number of limitations.Our main aim is to detect technology topics in scientific publications, use them to identify science- technology linkages, and finally map the development of neurotechnologies in patent data. To do so we create a neuroscience related corpus of scientific literature by filtering Scopus by subject area. Subject areas are assigned on journal level, and rather broad, in this case ranging from computer science, over chemistry, biology, and psychology. While this is an inherent feature of neuroscience research, it complicates the identification of technology topics from the publications text data. While our approach is geared towards selecting terms related to scientific methods, many of them cannot be related to an actual technology, but rather a method of scientific inquiry with in academia. A more focused and technology-targeted selection of publications could limit this effect. However, within the presented approach, the manual selection of relevant technology topics by a domain expert remains necessary. Furthermore, the presented approach assumes technologies to initially emerge in scientific literature, and then later being further developed to commercial technologies. Yet, it is reasonable that distinct technologies are emerging without former traces in science, or in commercial applications develop far away from their scientific origin in terms of methods and techniques as well as issues addressed. This could to some extent be addressed by extending the approach presented in this paper with additional iterative steps to identify additional technology topics within the selected patents, and expanding this selection to similar patents outside the current selection.