What is the clinical evidence from this study?

Study design: Other. Population: Cardiovascular interventional diseases (n=728). Intervention: LSTM-BLS model vs. Traditional LSTM, DNN, and RNN models. Primary outcome: Prediction accuracy of postoperative mortality risk.

What question did this study set out to answer?

The aim is to create a database for cardiovascular interventional diseases and develop a model to predict postoperative mortality risk.

March 9, 2026Open Access

Study on establishment of cardiovascular interventional disease database and prediction of postoperative mortality risk

Key Result

The proposed LSTM-BLS model predicted the risk of death after cardiovascular interventions with an accuracy of 87.46%, precision of 90.74%, and recall of 93.61%.

Key Points

The aim is to create a database for cardiovascular interventional diseases and develop a model to predict postoperative mortality risk.
Constructed a specific cardiovascular disease database over one year at a top-three hospital.
Developed an improved predictive model combining LSTM and BLS to assess mortality risk.
Compared the new model's performance with traditional LSTM to address overfitting and accuracy.
Achieved an accuracy rate of 87.46% with the new predictive model.
Reported precision of 90.74% and recall rate of 93.61%.
The model effectively supports timely medical interventions for high-risk patients.

Structured PICO

Does an improved LSTM-BLS machine learning model accurately predict 6-month postoperative mortality risk in patients undergoing cardiovascular interventions compared to traditional models?

Population

638 adult patients (≥18 years old) who completed cardiovascular interventional surgery at a top-three hospital in Zhejiang Province, China, with complete whole-cycle case data.

Intervention

LSTM-BLS (Long Short-Term Memory - Broad Learning System) hybrid machine learning model for predicting postoperative mortality risk

Comparator

Traditional machine learning models including DNN, RNN, and traditional LSTM

Outcome

Postoperative death (death or non-death) during 6-month follow-uphard clinical

A novel LSTM-BLS machine learning model utilizing a structured cardiovascular interventional database can accurately predict 6-month postoperative mortality, potentially aiding in timely clinical decision-making for high-risk patients.

Abstract

In order to solve the problems of missing, discontinuous, and unstructured data in past cardiovascular interventional studies, this work constructed a database of specific cardiovascular interventional diseases. Within one year of its implementation in a top-three hospital in Zhejiang Province, the database collected a total of 728 cases of cardiovascular interventional patients, realizing the structuring of patient data and 360° whole-cycle management. With the support of a specific disease database, we proposed an improved LSTM-BLS model to predict the risk of death after cardiovascular interventions. Compared with the traditional long-short term memory (LSTM) model, the parallel learning structure-width learning system (BLS) was introduced to calculate the weights directly, which can solve the problems of overfitting and delay caused by the deep structure of LSTM, so as to improve the accuracy of prediction. The experimental results showed that the accuracy rate of the proposed model was 87.46%, the precision was 90.74%, and recall rate was 93.61%, which can objectively reflect the postoperative death risk of patients, and help doctors to make timely medical intervention for patients with high death risk.

Bookmark

View Full Paper