Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Video

Updated: May 24, 2026

TBase - an Integrated Electronic Health Record and Research Database for Kidney Transplant Recipients

TBase - an Integrated Electronic Health Record and Research Database for Kidney Transplant Recipients

Published on: April 13, 2021

Transformer-Based Architecture for Predicting Surgical Complications from EHR Data.

Eduardo Alonso^1,2, Naroa Mendez¹, Xabier Calle^1,3

¹Vicomtech Foundation, Basque Research and Technology Alliance, San Sebastián, Spain.

Studies in Health Technology and Informatics

|May 23, 2026

Summary

Related Concept Videos

Issues And Trends In Healthcare Delivery System

Issues And Trends In Healthcare Delivery System

The issues and trends in healthcare delivery are constantly changing. The COVID-19 pandemic is one recent issue that wreaked havoc on healthcare systems, causing a shortage of healthcare workers, high demand for medicines and supplies, and increased medical expenditure due to a lack of insurance. Other issues include rising healthcare costs and care fragmentation.
Cost Containment
Payment for healthcare services has historically promoted adoption of costly and often unnecessary or inefficient...

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Regenerative Peripheral Nerve Interface (RPNI) and vascularized Denervated Muscle Targets (VDMT): a preclinical rabbit model as a translational feasibility and methodological platform.

Journal of translational medicine·2026

Same author

Concentric Versus Delta Bipolar Probes for Intraneural Fascicle Selection: A Rabbit Model Study.

Plastic and reconstructive surgery. Global open·2026

Same author

Pharmacovigilance Assistant: An Agentic Workflow for Reproducible Drug Safety Summaries.

Studies in health technology and informatics·2026

Same author

Non-Invasive Prediction of Embryo Ploidy from Time-Lapse Videos Using Video Vision Transformers (ViViT).

Studies in health technology and informatics·2026

Same author

Understanding LUng Cancer risk factors and their Impact Assessment (LUCIA): protocol for multicentre observational cohort study.

BMJ open·2026

Same author

A representational framework for learning and encoding structurally enriched trajectories in complex agent environments.

Neural networks : the official journal of the International Neural Network Society·2026

This summary is machine-generated.

A new Transformer-based model, STraTS, accurately predicts surgical complications using electronic health records (EHRs). This AI approach enhances preoperative risk assessment by analyzing temporal patient data, improving patient outcomes.

Area of Science:

Medical Informatics
Artificial Intelligence in Healthcare
Clinical Decision Support

Background:

Surgical complications significantly increase patient morbidity and healthcare expenses.
Existing risk calculators often fail to utilize the temporal information present in electronic health records (EHRs).

Purpose of the Study:

To evaluate STraTS, a Transformer-based deep learning architecture, for predicting surgical complications.
To assess the model's ability to leverage longitudinal EHR data for enhanced preoperative risk prediction.

Main Methods:

Utilized a real-world EHR dataset comprising 54,395 surgical procedures.
Employed a Transformer-based architecture (STraTS) with a self-attention mechanism to capture temporal clinical data dynamics.
Evaluated model performance using Area Under the Receiver Operating Characteristic curve (AUROC) and Area Under the Precision-Recall curve (AUPRC).

Keywords:

Electronic Health Records risk prediction surgical complications

Related Experiment Videos

Last Updated: May 24, 2026

TBase - an Integrated Electronic Health Record and Research Database for Kidney Transplant Recipients

TBase - an Integrated Electronic Health Record and Research Database for Kidney Transplant Recipients

Published on: April 13, 2021

Main Results:

STraTS achieved an AUROC of 0.882 and an AUPRC of 0.406 in predicting surgical complications.
The model demonstrated robust and consistent performance across diverse patient subgroups, including those stratified by age and sex.
The self-attention mechanism effectively captured complex temporal relationships within sparse EHR data.

Conclusions:

Transformer-based models, like STraTS, can effectively utilize longitudinal EHR data for surgical risk prediction.
STraTS offers a promising approach for generating individualized perioperative risk assessments.
This AI-driven method has the potential to improve surgical safety and reduce healthcare costs.