About ETRO  |  News  |  Events  |  Vacancies  |  Contact  
Home Research Education Industry Publications About ETRO

Master theses

Current and past ideas and concepts for Master Theses.

Explainable clinical coding based on knowledge graph information


Clinical coding refers to the task of assigning clinical codes to the clinical notes. Clinical codes typically refer to the classes or abbreviations referring to unique clinical concepts. Clinical coding has several applications ranging from billing to predictive modeling of patient state. In the proposed work we intend to explore several deep learning methods such as BERT with LSTM memory networks, BigBird model, Longformer, reformer models.

Kind of work

The student will investigate different deep learning architectures to perform this task. We plan to create graph embeddings based on the existing knowledge bases, then train the classification models with these embeddings as inputs. Additionally, together with the transformer based embeddings and knowledge graph embedding we plan to conduct experiments to see if there are any performance improvements. With these experiments we are hopeful to achieve state of art performance for the considered task.

Framework of the Thesis

Mullenbach, J. Wiegreffe, S. Duke, J. Sun, J. and Eisen-stein, J. 2018. Explainable prediction of medical codes from clinical text. InProceedings of the 2018 Conference of theNorth American Chapter of the Association for Computa-tional Linguistics: Human Language Technologies, Volume1 (Long Papers), 1101–1111.

Saha, Ankita, et al. "BIOINTMED: integrated biomedical knowledge base with ontologies and clinical trials." Medical & Biological Engineering & Computing 58.10 (2020): 2339-2354.

Number of Students


Expected Student Profile

The student should have a background in machine learning, NLP and python programming. Knowledge of deep learning and libraries like PyTorch, Transformers, TensorFlow is a plus.


Prof. Dr. Ir. Nikolaos Deligiannis

+32 (0)2 629 1683

more info


Dr. Jasabanta Patro

+32 (0)2 629 2930

more info

- Contact person




- Contact person

- Thesis proposals

- ETRO Courses

- Contact person

- Spin-offs

- Know How

- Journals

- Conferences

- Books

- Vacancies

- News

- Events

- Press


ETRO Department

Tel: +32 2 629 29 30

©2022 • Vrije Universiteit Brussel • ETRO Dept. • Pleinlaan 2 • 1050 Brussels • Tel: +32 2 629 2930 (secretariat) • Fax: +32 2 629 2883 • WebmasterDisclaimer