Clinical Features, Non-Contrast CT Radiomic and Radiological Signs in Models for the Prediction of Hematoma Expansion in Intracerebral Hemorrhage

被引:2
|
作者
Chen, Zejia Frank [1 ]
Zhang, Liying [1 ]
Carrington, Andre M. [1 ,2 ,3 ]
Thornhill, Rebecca [2 ]
Miguel, Olivier [1 ,2 ]
Auriat, Angela M. [1 ,2 ]
Omid-Fard, Nima [2 ]
Hiremath, Shivaprakash [2 ]
Tshemeister Abitbul, Vered [1 ,2 ]
Dowlatshahi, Dar [1 ,4 ]
Demchuk, Andrew [5 ]
Gladstone, David [6 ]
Morotti, Andrea [7 ]
Casetta, Ilaria [8 ]
Fainardi, Enrico [9 ]
Huynh, Thien [10 ,11 ]
Elkabouli, Marah [1 ]
Talbot, Zoe [1 ]
Melkus, Gerd [1 ,2 ]
Aviv, Richard, I [1 ,2 ,12 ]
机构
[1] Ottawa Hosp Res Inst, Ottawa, ON, Canada
[2] Univ Ottawa, Dept Radiol Radiat Oncol & Med Phys, Ottawa, ON, Canada
[3] Univ Waterloo, Dept Syst Design Engn, Waterloo, ON, Canada
[4] Univ Ottawa, Dept Med Neurol, Ottawa, ON, Canada
[5] Foothills Med Ctr, Dept Med Neurol, Calgary, AB, Canada
[6] Univ Toronto, Sunnybrook Hlth Sci Ctr, Dept Med Neurol, Toronto, ON, Canada
[7] ASST Spedali Civili Brescia, Dept Neurol Sci & Vis, Neurol Unit, Brescia, Italy
[8] Univ Ferrara, Neurol Clin, Ferrara, Italy
[9] Univ Florence, Dept Expt & Clin Biomed Sci, Neuroradiol Unit, Florence, Italy
[10] Mayo Clin, Dept Radiol, Jacksonville, FL USA
[11] Mayo Clin, Dept Neurosurg, Jacksonville, FL USA
[12] Univ Ottawa, Ottawa Hosp, Dept Radiol, Gen Campus,CPCR Bldg,Room L2121, Ottawa, ON K1H 1M2, Canada
来源
CANADIAN ASSOCIATION OF RADIOLOGISTS JOURNAL-JOURNAL DE L ASSOCIATION CANADIENNE DES RADIOLOGISTES | 2023年 / 74卷 / 04期
基金
加拿大创新基金会;
关键词
hematoma expansion; intracerebral hemorrhage; radiomics; machine learning; non-contrast CT; NONCONTRAST COMPUTED-TOMOGRAPHY; GROWTH; SCORES;
D O I
10.1177/08465371231168383
中图分类号
R8 [特种医学]; R445 [影像诊断学];
学科分类号
1002 ; 100207 ; 1009 ;
摘要
PurposeRapid identification of hematoma expansion (HE) risk at baseline is a priority in intracerebral hemorrhage (ICH) patients and may impact clinical decision making. Predictive scores using clinical features and Non-Contract Computed Tomography (NCCT)-based features exist, however, the extent to which each feature set contributes to identification is limited. This paper aims to investigate the relative value of clinical, radiological, and radiomics features in HE prediction.MethodsOriginal data was retrospectively obtained from three major prospective clinical trials ["Spot Sign" Selection of Intracerebral Hemorrhage to Guide Hemostatic Therapy (SPOTLIGHT)NCT01359202; The Spot Sign for Predicting and Treating ICH Growth Study (STOP-IT)NCT00810888] Patients baseline and follow-up scans following ICH were included. Clinical, NCCT radiological, and radiomics features were extracted, and multivariate modeling was conducted on each feature set.Results317 patients from 38 sites met inclusion criteria. Warfarin use (p=0.001) and GCS score (p=0.046) were significant clinical predictors of HE. The best performing model for HE prediction included clinical, radiological, and radiomic features with an area under the curve (AUC) of 87.7%. NCCT radiological features improved upon clinical benchmark model AUC by 6.5% and a clinical & radiomic combination model by 6.4%. Addition of radiomics features improved goodness of fit of both clinical (p=0.012) and clinical & NCCT radiological (p=0.007) models, with marginal improvements on AUC. Inclusion of NCCT radiological signs was best for ruling out HE whereas the radiomic features were best for ruling in HE.ConclusionNCCT-based radiological and radiomics features can improve HE prediction when added to clinical features. Visual Abstract Objectif : L'identification rapide du risque d'expansion d'un hematome avant le debut des traitements est une priorite chez les patients ayant une hemorragie intracerebrale (HI) et elle peut avoir des consequences sur la prise de decisions cliniques. Il existe des scores predictifs utilisant les caracteristiques cliniques et les caracteristiques basees sur la tomodensitometrie sans contraste (TDM-sc). Toutefois, la portee de la contribution a l'identification de chaque ensemble de caracteristiques est limitee. Cet article vise a etudier la valeur relative des caracteristiques cliniques, radiologiques et radiomiques pour la prediction de l'expansion des hematomes. Methodes : Les donnees originales ont ete obtenues retrospectivement a partir de deux etudes cliniques prospectives majeures Spotlight et Spot-ITrr; : << Spot Sign >> Selection of Intracerebral Hemorrhage to Guide Hemostatic Therapy (SPOTLIGHT) (NCT01359202); The Spot Sign for Predicting and Treating ICH Growth Study (STOP-IT) (NCT00810888). Des examens d'imagerie de patients ayant souffert d'une HI effectues avant les traitements et au cours du suivi ont ete inclus. Les caracteristiques cliniques, radiologiques (TDM-sc) et radiomiques ont ete extraites et une modelisation multifactorielle a ete effectuee sur chaque ensemble de caracteristiques. Resultats : Une population de 317 patients provenant de 38 centres satisfaisait les criteres d'inclusion. L'utilisation de la warfarine (P = 0.001) et le score GCS (P = 0.046) ont ete des facteurs cliniques predictifs de l'expansion de l'hematome. Le modele le plus performant pour la prediction de l'expansion de l'hematome incluait les caracteristiques cliniques, radiologiques et radiomiques et presentait une aire sous la courbe (ASC) de 87.7 %. Les caracteristiques radiologiques (TDM-sc) ont ameliore de 6.5 % l'ASC du modele de reference clinique et de 6.4 % celle d'un modele combinant clinique et radiomique. L'ajout des caracteristiques radiomiques a ameliore la qualite d'adaptation des modeles cliniques et radiologiques (TDM-sc) avec une amelioration marginale de l'ASC. L'inclusion des signes radiologiques (TDM-sc) a ete le meilleur moyen d'ecarter un diagnostic d'hematome, tandis que les caracteristiques cliniques ont ete les meilleures pour confirmer son existence. Conclusions : Les caracteristiques radiologiques basees sur une TDM sans contraste et les caracteristiques radiomiques peuvent ameliorer la prediction de l'expansion d'un hematome intracerebral quand elles sont ajoutees aux caracteristiques cliniques.
引用
收藏
页码:713 / 722
页数:10
相关论文
共 50 条
  • [31] Neoplastic and Non-neoplastic Acute Intracerebral Hemorrhage in CT Brain Scans: Machine Learning-Based Prediction Using Radiomic Image Features
    Nawabi, Jawed
    Kniep, Helge
    Kabiri, Reza
    Broocks, Gabriel
    Faizy, Tobias D.
    Thaler, Christian
    Schoen, Gerhard
    Fiehler, Jens
    Hanning, Uta
    FRONTIERS IN NEUROLOGY, 2020, 11
  • [32] Feasibility of a combined swirl and blending sign on non-contrast computed tomography for predicting early hematoma expansion after spontaneous intracerebral hemorrhage
    Kim, Jang-Hun
    Choi, Jong-Il
    JOURNAL OF NEUROSURGICAL SCIENCES, 2022, 66 (06) : 582 - 588
  • [33] Independent Validation of the Hematoma Expansion Prediction Score: A Non-contrast Score Equivalent in Accuracy to the Spot Sign
    Vignan Yogendrakumar
    Tim Ramsay
    Dean A. Fergusson
    Andrew M. Demchuk
    Richard I. Aviv
    David Rodriguez-Luna
    Carlos A. Molina
    Yolanda Silva Blas
    Imanuel Dzialowski
    Adam Kobayashi
    Jean-Martin Boulanger
    Cheemun Lum
    Gord Gubitz
    Padma Srivastava
    Jayanta Roy
    Carlos S. Kase
    Rohit Bhatia
    Michael D. Hill
    Magdy Selim
    Dar Dowlatshahi
    Neurocritical Care, 2019, 31 : 1 - 8
  • [34] The Attenuation Value Within the Non-hypodense Region on Non-contrast Computed Tomography of Spontaneous Cerebral Hemorrhage: A Long-Neglected Predictor of Hematoma Expansion
    Chen, Yong
    Cao, Dan
    Guo, Zheng-Qian
    Ma, Xiao-Ling
    Ou, Yi-Bo
    He, Yue
    Chen, Xu
    Chen, Jian
    FRONTIERS IN NEUROLOGY, 2022, 13
  • [35] Automatic and Efficient Prediction of Hematoma Expansion in Patients with Hypertensive Intracerebral Hemorrhage Using Deep Learning Based on CT Images
    Ma, Chao
    Wang, Liyang
    Gao, Chuntian
    Liu, Dongkang
    Yang, Kaiyuan
    Meng, Zhe
    Liang, Shikai
    Zhang, Yupeng
    Wang, Guihuai
    JOURNAL OF PERSONALIZED MEDICINE, 2022, 12 (05):
  • [36] Neoplastic and Non-neoplastic Acute Intracerebral Hemorrhage in CT Brain Scans: Machine Learning-Based Prediction Using Radiomic Image Features (vol 11, 285, 2020)
    Nawabi, Jawed
    Kniep, Helge
    Kabiri, Reza
    Broocks, Gabriel
    Faizy, Tobias D.
    Thaler, Christian
    Schon, Gerhard
    Fiehler, Jens
    Hanning, Uta
    FRONTIERS IN NEUROLOGY, 2021, 12
  • [37] Hematoma expansion prediction in intracerebral hemorrhage patients by using synthesized CT images in an end-to-end deep learning framework
    Yalcin, Cansu
    Abramova, Valeriia
    Terceno, Mikel
    Oliver, Arnau
    Silva, Yolanda
    Llado, Xavier
    COMPUTERIZED MEDICAL IMAGING AND GRAPHICS, 2024, 117
  • [38] Potential of high dimensional radiomic features to assess blood components in intraaortic vessels in non-contrast CT scans
    Mahmoudi, Scherwin
    Martin, Simon S.
    Ackermann, Joerg
    Zhdanovich, Yauheniya
    Koch, Ina
    Vogl, Thomas J.
    Albrecht, Moritz H.
    Lenga, Lukas
    Bernatz, Simon
    BMC MEDICAL IMAGING, 2021, 21 (01)
  • [39] Potential of high dimensional radiomic features to assess blood components in intraaortic vessels in non-contrast CT scans
    Scherwin Mahmoudi
    Simon S. Martin
    Jörg Ackermann
    Yauheniya Zhdanovich
    Ina Koch
    Thomas J. Vogl
    Moritz H. Albrecht
    Lukas Lenga
    Simon Bernatz
    BMC Medical Imaging, 21
  • [40] Outcome Prediction of Spontaneous Supratentorial Intracerebral Hemorrhage after Surgical Treatment Based on Non-Contrast Computed Tomography: A Multicenter Study
    Zhang, Kangwei
    Zhou, Xiang
    Xi, Qian
    Wang, Xinyun
    Yang, Baoqing
    Meng, Jinxi
    Liu, Ming
    Dong, Ningxin
    Wu, Xiaofen
    Song, Tao
    Wei, Lai
    Wang, Peijun
    JOURNAL OF CLINICAL MEDICINE, 2023, 12 (04)