Medical text mining. Also in German language.

We push digitalization in the healthcare and life science sector. We convey unstructured, siloed or scattered clinical or biomedical text documents into structured, actionable data — cleanly represented using controlled vocabularies including MedDRA, ICD10, MeSH, and ATC. Natural language processing in English or German, for texts written by healthcare professionals, researchers, or patients.

Structured drug information.

Our core expertise is in processing medication-related content (adverse drug reactions, drug safety, pharmacovigilance). We leverage modern text and data mining technologies for creating and maintaining structured drug information resources to power Digital Health solutions.

Solutions ◼

datapeutics◼ supports relevant use cases around transforming unstructured data into structured, actionable information. They all are designed to improve your digital processes, reduce manual efforts, and hence cut your costs. Examples:

▶ Deep Digitalization

▶ Improved Clinical Coding

▶ Easier Predictive Analytics

▶ Natural User Interface

▶ Smart Document Search

▶ Competent Consulting

Each use case can be adapted to best serve your needs. Please do contact us about your specific use case to learn more.

Technology ◼

Enabling Artificial Intelligence in Healthcare.

We turn your raw text like case reports, Summary of Product Characteristics (SmPCs), or electronic health records (EHRs) into structured actionable data — by mapping it to controlled vocabularies like MedDRA, ICD-10, OPS, MeSH, or ATC. Because structured data is what you need in order to improve processes, build smart solutions with artificial intelligence (AI), and boost digitalization in your business, be it pharmaceutics, healthcare, or life science.

Turning Text into Knowledge. In German.

We are experts in natural language processing (NLP) and machine learning (ML), and utilize modern text mining technology. Our specialty is processing German language text, which is poorly covered by academic endeavors. We focus on medical language and terminology, and carefully address their particularities and challenges. This way we can achieve highest accuracy, and create and maintain the structured drug information that you need to boost your business.

Methods and Techniques We Use:

Named Enitity Recognition (NER)
- Examples: indication, adverse drug reaction, drug, dosage, patient group
Relation Extraction (REx)

e.g. mapping adverse drug reactions to drug usage

Text Classification

e.g. searching for relevant documents or clustering topics

Automatic Normalization/Mapping to Controlled Vocabulary and Ontologies
- ICD10/OPS
- MedDRA
- ATC
- MeSH
- UMLS
- others...

Team ◼

We are a tightly integrated team of data science and engineering experts with both deep technological expertise in IT systems, Machine Learning including Text Mining/NLP as well as domain knowledge in pharmacovigilance and drug data incl. clinical trials. From different backgrounds and decades of experience in both academia and industry we combine the required skills necessary to push digitalization in healthcare and solve customer-specific data challenges in the biomedical domain.

Dr. Alexander Zien ◼

Machine Learning, Personalised medicine,
Bioinformatics

alexaarander.intuzien@datesarapeutics.com

[ show details ]

Dipl.-Inf. Christian Hachenberg ◼

Text and Data Mining, NLP, Software Development