MedAid Core is an engine for auto-indexation of biomedical documents regardless of their format.

Start getting the most out of your work no matter its format or where it is, and efficiently collaborating with your colleagues reusing the knowledge of your group, and inspiring your teammates to engage in new and different ways.

  • Flexible Framework.

    Medaid is a natural language processing framework specialized in bionpl that uses deep neural networks to facilitate tasks such as search and recommendation ranking, multiclass classification and medical query understanding. MedAid pipeline has been designed with flexibility in mind, so that different NLP models can be “swapped” in as needed, depending on the requirements of different production processes.

  • Expandible Capabilities.

    Medaid will be offered as open source software, meaning that others can download it and easily adapt it for their own use. In the future, we are looking to extend MedAid as a general language understanding and generation library so that it can perform other important BioNLP tasks like Automated Question Answering.

  • Fully upgradeable.

    MedAid core layer combines the best state-of-the-art algorithms such as Transformers for Natural Language Processing or Contextualized word embeddings from BERT. As a future direction, we plan to use lighter versions of Transformer-based architectures to decrease training time periods while keeping prediction performances high.

How it works?.

Click to enlarge.

  Some examples of practical applications

With the adoption of Electronic Health Management Information Systems, hospitals and clinical centers have access to large amounts of heterogeneous patient data and the area of eHealth is constantly evolving in the last years. However, many entries on hospital records and medical studies are still registered in a completely unstructured free form. Practitioner and patient encounters are commonly recorded in free-form clinical notes as summarized by practitioners during the medical triage, diagnosis, and different assessments. Our software will allow you to access your data wherever it is, and find patterns and relationships between your documents and the work of your collaborators and research partners.

Highlights really important information.

Gain time and security in your work by getting what you are looking for faster.

Performs semantic searches over all types of sources.

Identify concepts, entities and semantic relationships in your content and use them to perform advanced searches.

Support for the most common formats.

Your documents are automatically indexed regardless of their source format. Automatically recognize your DOC, ODT, XLS, PDF, TXT documents on Pc, Android or Mac.

Document indexing now is simple and does not require user intervention.

You continue working as usual while we explore your files, guaranteeing your privacy.

Keep your documents safe.

If you have a good catalog and a complete indexing of your content, then it is easy to manage changes in your files made by your collaborators and carry out version control to keep your work safe.

Navigate your document database while you work.

Don't waste time and browse your document database and obtain relevant documents while you continue working.

Realtime re-analysis.

Each document is compared with your knowledge database to extract its unique fingerprint, which allows us to classify it according to different aspects.

Smart & non-intrusive interactive recommendations

Get recommendations for articles, contributors and technical information in realtime while always working in a non-intrusive way and always without interrupting your work.

Build your knowledge database

Explore your file history in a unified way by searching clinical documents for semantic similarity.

Improved teamwork

When medical teams focus on collaboration, staff is reorganized to focus on patients, rather than individual skills and expertise. Regularly seek out a diversity of opinions and ideas among teammates to build strategies and solve problems. As a result, team participants are more engaged, feel trusted and are more likely to take ownership of their work.

Other ideas.

Enriching with open data repositories.

Creating datasets compatible with open standards and repositories.

- Obtain summaries automatically. Get bibliography recommendations related to content summaries

- Obtain definitions of concepts, images and more. Get definitions of concepts, images, figures, and related data tables.

- Quickly see dosages, drugs and interactions. Consult drug doses, incompatibilities and pharmacological protocols.

Send us your opinions