Prediction of Surgical Approach in Mitral Valve Disease by XGBoost Algorithm Based on Echocardiographic Features

Lin, Xiaoxuan1; Chen, Lixin1; Zhang, Defu1; Luo, Shuyu1; Sheng, Yuanyuan1; Liu, Xiaohua1; Liu, Qian1; Li, Jian1; Shi, Bobo1; Peng, Guijuan1; Zhong, Xiaofang1; Huang, Yuxiang1; Li, Dagang2,3; Qin, Gengliang2,3; Yin, Zhiqiang2,3; Xu, Jinfeng1; Meng, Chunying2,3; Liu, Yingying1

2023-02-01

10.3390/jcm12031193

JOURNAL OF CLINICAL MEDICINE   Impact Factor & Quartile

2077-0383

In this study, we aimed to develop a prediction model to assist surgeons in choosing an appropriate surgical approach for mitral valve disease patients. We retrospectively analyzed a total of 143 patients who underwent surgery for mitral valve disease. The XGBoost algorithm was used to establish a predictive model to decide a surgical approach (mitral valve repair or replacement) based on the echocardiographic features of the mitral valve apparatus, such as leaflets, the annulus, and sub-valvular structures. The results showed that the accuracy of the predictive model was 81.09% in predicting the appropriate surgical approach based on the patient's preoperative echocardiography. The result of the predictive model was superior to the traditional complexity score (81.09% vs. 75%). Additionally, the predictive model showed that the three main factors affecting the choice of surgical approach were leaflet restriction, calcification of the leaflet, and perforation or cleft of the leaflet. We developed a novel predictive model using the XGBoost algorithm based on echocardiographic features to assist surgeons in choosing an appropriate surgical approach for patients with mitral valve disease.

mitral valve disease echocardiography prediction XGBoost mitral valve repair mitral valve replacement

SCI

English

Corresponding

General & Internal Medicine

Medicine, General & Internal

WOS:000931072000001

MDPI

Web of Science

1.Jinan Univ, Shenzhen Peoples Hosp, Clin Med Coll 2, Dept Ultrasound, Shenzhen 518020, Peoples R China
2.Jinan Univ, Shenzhen Peoples Hosp, Dept Cardiovasc Surg, Clin Med Coll 2, Shenzhen 518020, Peoples R China
3.Southern Univ Sci & Technol, Affiliated Hosp 1, Shenzhen 518020, Peoples R China

Lin, Xiaoxuan,Chen, Lixin,Zhang, Defu,et al. Prediction of Surgical Approach in Mitral Valve Disease by XGBoost Algorithm Based on Echocardiographic Features[J]. JOURNAL OF CLINICAL MEDICINE,2023,12(3).

Lin, Xiaoxuan.,Chen, Lixin.,Zhang, Defu.,Luo, Shuyu.,Sheng, Yuanyuan.,...&Liu, Yingying.(2023).Prediction of Surgical Approach in Mitral Valve Disease by XGBoost Algorithm Based on Echocardiographic Features.JOURNAL OF CLINICAL MEDICINE,12(3).

Lin, Xiaoxuan,et al."Prediction of Surgical Approach in Mitral Valve Disease by XGBoost Algorithm Based on Echocardiographic Features".JOURNAL OF CLINICAL MEDICINE 12.3(2023).