| Title |
Bridging structured and unstructured data in stress echocardiography survival analysis |
| Authors |
Satkauskienė, Saulė ; Petkevičius, Linas ; Balkevičienė, Laura ; Čelutkienė, Jelena |
| Full Text |
|
| Is Part of |
Proceedings of the 30th international conference on Information Society and University Studies Kaunas, Lithuania, May 15, 2025.. Aachen : CEUR-WS. 2026, p. 315-324 |
| Keywords [eng] |
survival analysis ; echocardiography data ; BERT ; canonical correlation ; unstructured data |
| Abstract [eng] |
The research innovatively incorporates unstructured patients diagnosis data processing using the BERT language model, subsequently analysing it with logistic regression and canonical correlation analysis to understand its relationship with structured data. This approach enriches the predictive model and underscores the value of integrating diverse data types. This processed data is then integrated alongside structured data, significantly enhancing the accuracy of the survival models. This incorporation demonstrates the potent impact of combining diverse data types on improving predictive capabilities in medical prognostics. |
| Published |
Aachen : CEUR-WS |
| Type |
Conference paper |
| Language |
English |
| Publication date |
2026 |
| CC license |
|
