logo

Hercules Dalianis

Bio

Dalianis has been a professor in Computer and Systems Sciences at Stockholm University, Sweden since 2011. Dalianis graduated as a civil engineer (Civilingenjör) in Electrotechnical Engineering, 1984 from KTH, Stockholm, Sweden with a specialisation in telecommunications. Thereafter he focused on language engineering. Dalianis became a PhD (Teknologie doctor) at KTH in 1996 and associate professor, (docent), in Computer and Systems Sciences at Stockholm University, Sweden, 2001.

Dalianis works in the interface between industry and university with the aim to make research results useful for society. He has been the project leader for a large number of research projects both in academia and in industry at national, Nordic and international levels. Dalianis is specialised in the area of human language technology, to make computers understand and process human language text, but also to make a computer produce text automatically. The last 13 years he has worked in the area of clinical text mining focused on processing electronic patient records in Swedish; specifically in finding adverse drug events as well as health care associated infections but also with automatic de-identification and pseudonymisation pf electronic patient records.

Hercules's projects
Project title Year Theme Project management
NorKlinTekst: Natural language processing to extract knowledge from clinical notes in electronic health records 2018 - 2020 Health data
Taridzo Chomutare
ClinCode - Computer-Assisted Clinical ICD-10 Coding for improving efficiency and quality in healthcare 2021 - 2024 Health data
Hercules Dalianis
Hercules's publications in Cristin
Title Year Category
Easy-ICD - a tool to predict ICD-10 gastro codes from Swedish discharge summaries 2024 Poster
Deidentifying a Norwegian clinical corpus - An effort to create a privacy-preserving Norwegian large clinical language model 2024 Academic article
Improving Quality of ICD-10 (International Statistical Classification of Diseases, Tenth Revision) Coding Using AI: Protocol for a Crossover Randomized Controlled Trial 2024 Academic article
De-identifying Norwegian Clinical Text using Resources from Swedish and Danish 2023 Academic article
De-identifying Norwegian clinical text using resources from Swedish and Danish 2023 Academic lecture
Using a large open clinical corpus for improved ICD-10 diagnosis coding 2023 Academic article
Using a large open clinical corpus for improved ICD-10 diagnosis coding 2023 Academic lecture
Kunstig intelligens kan tolke din pasientjournal 2022 Interview
ClinCode - Computer-Assisted Clinical ICD-10 Coding for improving efficiency and quality in healthcare 2022 Academic lecture
ClinCode - Computer-Assisted Clinical ICD-10 Coding for improving efficiency and quality in healthcare 2022 Poster
Evaluation of LIME and SHAP in Explaining Automatic ICD-10 Classifications of Swedish Gastrointestinal Discharge Summaries 2022 Academic article
Combining deep learning and fuzzy logic to predict rare ICD-10 codes from clinical notes 2022 Academic chapter/article/Conference paper
The Influence of NegEx on ICD-10 Code Prediction in Swedish: How is the Performance of BERT and SVM Models Affected by Negations? 2022 Academic lecture
The Influence of NegEx on ICD-10 Code Prediction in Swedish: How is the Performance of BERT and SVM Models Affected by Negations? 2022 Academic article
Implementation of specialised attention mechanisms: ICD-10 classification of Gastrointestinal discharge summaries in English, Spanish and Swedish 2022 Academic article
Multi-label Diagnosis Classification of Swedish Discharge Summaries – ICD-10 Code Assignment Using KB-BERT 2021 Academic article
On the Contribution of Per-ICD Attention Mechanisms to Classify Health Records in Languages With Fewer Resources than English 2021 Academic article
Challenges and opportunities beyond structured data in analysis of electronic health records 2021 Academic literature review
De-identifying Swedish EHR text using public resources in the general domain 2020 Academic article
Building a De-identification System for Real Swedish Clinical Text Using Pseudonymised Clinical Text 2019 Academic chapter/article/Conference paper
Negation detection in Norwegian medical text: Porting a Swedish NegEx to Norwegian. Work in progress 2018 Poster
Hercules's posters
Poster no. Title Author(s) Size
2022 06 ClinCode - Computer-Assisted Clinical ICD-10 Coding for improving efficiency and quality in healthcare Hercules Dalianis, Taridzo Chomutare, Andrius Budrionis, Therese Svenning A1
2018 14 Negation detection in Norwegian medical text: Porting a Swedish NegEx to Norwegian Work in progress Andrius Budrionis, Hercules Dalianis, Kassaye Yitbarek Yigzaw, Alexandra Makhlysheva, Taridzo Chomutare A0
Hercules's presentations
Date Title Event Author(s)
2019.06.18 Clinical text mining – Secondary use of electronic patient records to improve healthcare Kunstig intelligens i helsetjenesten, konferanse i Bodø Hercules Dalianis, Stockholm University