Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2601.07385 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908759612194816 |
|---|---|
| author | Zelina, Petr Řeháček, Marko Halámková, Jana Bohovicová, Lucia Rusinko, Martin Nováček, Vít |
| author_facet | Zelina, Petr Řeháček, Marko Halámková, Jana Bohovicová, Lucia Rusinko, Martin Nováček, Vít |
| contents | Clinical notes hold rich yet unstructured details about diagnoses, treatments, and outcomes that are vital to precision medicine but hard to exploit at scale. We introduce a method that represents each patient as a matrix built from aggregated embeddings of all their notes, enabling robust patient similarity computation based on their latent low-rank representations. Using clinical notes of 4,267 Czech breast-cancer patients and expert similarity labels from Masaryk Memorial Cancer Institute, we evaluate several matrix-based similarity measures and analyze their strengths and limitations across different similarity facets, such as clinical history, treatment, and adverse events. The results demonstrate the usefulness of the presented method for downstream tasks, such as personalized therapy recommendations or toxicity warnings. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_07385 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Computing patient similarity based on unstructured clinical notes Zelina, Petr Řeháček, Marko Halámková, Jana Bohovicová, Lucia Rusinko, Martin Nováček, Vít Machine Learning Clinical notes hold rich yet unstructured details about diagnoses, treatments, and outcomes that are vital to precision medicine but hard to exploit at scale. We introduce a method that represents each patient as a matrix built from aggregated embeddings of all their notes, enabling robust patient similarity computation based on their latent low-rank representations. Using clinical notes of 4,267 Czech breast-cancer patients and expert similarity labels from Masaryk Memorial Cancer Institute, we evaluate several matrix-based similarity measures and analyze their strengths and limitations across different similarity facets, such as clinical history, treatment, and adverse events. The results demonstrate the usefulness of the presented method for downstream tasks, such as personalized therapy recommendations or toxicity warnings. |
| title | Computing patient similarity based on unstructured clinical notes |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2601.07385 |