Your browser doesn't support javascript.
loading
Predicting the 90-day prognosis of stereotactic brain hemorrhage patients by multiple machine learning using radiomic features combined with clinical features.
Li, Jinwei; Liang, Cong; Dang, Junsun; Zhang, Yang; Chen, Hongmou; Yan, Xianlei; Liu, Quan.
Afiliación
  • Li J; Department of Neurosurgery, Liuzhou Workers Hospital, Liuzhou, Guangxi, China.
  • Liang C; Department of Neurosurgery, West China Hospital, Sichuan University, Chengdu, Sichuan, China.
  • Dang J; Department of Pharmacy, Liuzhou Workers Hospital, Liuzhou, Guangxi, China.
  • Zhang Y; Department of Neurosurgery, Liuzhou Workers Hospital, Liuzhou, Guangxi, China.
  • Chen H; Department of Vascular Surgery, Fuwai Yunnan Cardiovascular Hospital, Affiliated Cardiovascular Hospital of Kunming Medical University, Kunming, Yunnan, China.
  • Yan X; Department of Neurosurgery, Liuzhou Workers Hospital, Liuzhou, Guangxi, China.
  • Liu Q; Department of Neurosurgery, Liuzhou Workers Hospital, Liuzhou, Guangxi, China.
Front Surg ; 11: 1344263, 2024.
Article en En | MEDLINE | ID: mdl-38389861
ABSTRACT
Hypertensive Intracerebral Hemorrhage (HICH) is one of the most common types of cerebral hemorrhage with a high mortality and disability rate. Currently, preoperative non-contrast computed tomography (NCCT) scanning-guided stereotactic hematoma removal has achieved good results in treating HICH, but some patients still have poor prognoses. This study collected relevant clinical and radiomic data by retrospectively collecting and analyzing 432 patients who underwent stereotactic hematoma removal for HICH from January 2017 to December 2020 at the Liuzhou Workers Hospital. The prognosis of patients after 90 days was judged by the modified Rankin Scale (mRS) scale and divided into the good prognosis group (mRS ≤ 3) and the poor prognosis group (mRS > 3). The 268 patients were randomly divided into training and test sets in the ratio of 82, with 214 patients in the training set and 54 patients in the test set. The least absolute shrinkage and selection operator (Lasso) was used to screen radiomics features. They were combining clinical features and radiomic features to build a joint prediction model of the nomogram. The AUCs of the clinical model for predicting different prognoses of patients undergoing stereotactic HICH were 0.957 and 0.922 in the training and test sets, respectively, while the AUCs of the radiomics model were 0.932 and 0.770, respectively, and the AUCs of the combined prediction model for building a nomogram were 0.987 and 0.932, respectively. Compared with a single clinical or radiological model, the nomogram constructed by fusing clinical variables and radiomic features could better identify the prognosis of HICH patients undergoing stereotactic hematoma removal after 90 days.
Palabras clave
Texto completo
Añadir a Mi BVS
Imprimir
XML
PubMed Links
Buscar en Google

Texto completo: 1 Colección: 01-internacional Base de datos: MEDLINE Idioma: En Revista: Front Surg Año: 2024 Tipo del documento: Article País de afiliación: China Pais de publicación: Suiza
Texto completo
Añadir a Mi BVS
Imprimir
XML
PubMed Links
Buscar en Google

Texto completo: 1 Colección: 01-internacional Base de datos: MEDLINE Idioma: En Revista: Front Surg Año: 2024 Tipo del documento: Article País de afiliación: China Pais de publicación: Suiza