Skip to main content
Download PDF

Application of machine learning model to predict lacunar cerebral infarction in elderly patients with femoral neck fracture before surgery

BMC Geriatrics volume 22, Article number: 912 (2022) Cite this article

Abstract

Background

Femoral neck fracture and lacunar cerebral infarction (LCI) are the most common diseases in the elderly. When LCI patients undergo a series of traumas such as surgery, their postoperative recovery results are often poor. Moreover, few studies have explored the relationship between LCI and femoral neck fracture in the elderly. Therefore, this study will develop a ML (machine learning)-based model to predict LCI before surgery in elderly patients with a femoral neck fracture.

Methods

Professional medical staff retrospectively collected the data of 161 patients with unilateral femoral neck fracture who underwent surgery in the Second Affiliated Hospital of Wenzhou Medical University database from January 1, 2015, to January 1, 2020. Patients were divided into two groups based on LCI (diagnosis based on cranial CT image): the LCI group and the non-LCI group. Preoperative clinical characteristics and preoperative laboratory data were collected for all patients. Features were selected by univariate and multivariate logistic regression analysis, with age, white blood cell (WBC), prealbumin, aspartate aminotransferase (AST), total protein, globulin, serum creatinine (Scr), blood urea nitrogen (Bun)/Scr, lactate dehydrogenase (LDH), serum sodium and fibrinogen as the features of the ML model. Five machine learning algorithms, Logistic regression (LR), Gradient Boosting Machine (GBM), Extreme Gradient Boosting (XGBoost), Random Forest (RF), and Decision tree (DT), were used in combination with preoperative clinical characteristics and laboratory data to establish a predictive model of LCI in patients with a femoral neck fracture. Furthermore, indices like the area under the receiver operating characteristic (AUROC), sensitivity, specificity, and accuracy were calculated to test the models’ performance.

Results

The AUROC of 5 ML models ranged from 0.76 to 0.95. It turned out that the RF model demonstrated the highest performance in predicting LCI for femoral neck fracture patients before surgery, whose AUROC was 0.95, sensitivity 1.00, specificity 0.81, and accuracy 0.90 in validation sets. Furthermore, the top 4 high-ranking variables in the RF model were prealbumin, fibrinogen, globulin and Scr, in descending order of importance.

Conclusion

In this study, 5 ML models were developed and validated for patients with femoral neck fracture to predict preoperative LCI. RF model provides an excellent predictive value with an AUROC of 0.95. Clinicians can better conduct multidisciplinary perioperative management for patients with femoral neck fractures through this model and accelerate the postoperative recovery of patients.

Peer Review reports

Introduction

With the rapid development of medical technology in today’s world, the life expectancy of modern people continues to improve, and age-related diseases have become one of the main burdens of global medical expenditure. Femoral neck fracture and lacunar cerebral infarction (LCI) are the most common diseases in the elderly [1, 2]. Considering that the use of cannulated screw for reduction of femoral neck fracture requires a long time of bed immobilization and a high possibility of femoral head necrosis, clinicians generally recommend a hip replacement for elderly patients with femoral neck fracture [1, 3]. Hip replacement is a complex and challenging operation that requires strict perioperative management of patients, especially patients with cardiovascular and cerebrovascular conditions, to prevent serious complications such as pulmonary embolism [4].

Compared with non-LCI patients, LCI patients often experience additional problems after trauma, such as surgery [5, 6]. When patients with hip fractures are complicated by cerebral infarction, recovery of limb function is poor and postoperative mortality is higher [7, 8]. In addition, when fracture patients are complicated with cerebral infarction, multidisciplinary management often accelerates postoperative recovery and improves their quality of life [9, 10]. However, for elderly patients with a femoral neck fracture, head CT is not a routine preoperative examination to detect LCI in advance due to a series of economic factors. Therefore, there is an urgent need for a predictive model to predict whether elderly patients with femoral neck fractures are associated with LCI before surgery. Therefore, timely multidisciplinary management of patients with LCI can be carried out to promote the recovery of patients.

As a new algorithm model, the machine learning (ML) algorithm has been widely used in various fields, especially in the medical field [11]. The traditional ML model, Logistic regression (LR), Gradient boosting machine (GBM), Extreme gradient boosting (XGBoost), Random forest (RF) and Decision tree (DT) showed good predictive efficacy in different diseases [12, 13]. These prediction models provide better help for clinicians in making clinical decisions.

Therefore, this study will aim to develop a ML-based model to predict LCI before surgery in elderly patients with a femoral neck fracture. Clinicians will use ML model results to manage better perioperative management of elderly patients with a femoral neck fracture.

Methods

Study design

Variables including demographic characteristics, comorbidities, and laboratory data were collected from elderly patients with femoral neck fractures who underwent head CT. According to CT results, patients were divided into the LCI group and the non-LCI group. Univariate and multivariate logistic regression analyses were performed to select the preoperative variables of the two groups. The selected variables were placed into five machine learning models to predict LCI in patients with a femoral neck fracture. The detailed process is shown in Fig. 1.

Fig. 1
figure 1

Flow chart showing analyses and model making process for the study. LR, Logistic regression; GBM, Gradient boosting machine; RF, Random forest; DT, Decision tree; XGBoost, Extreme gradient boosting

Full size image

Setting

With the approval of the institutional review Committee, professional medical staff retrospectively collected and analyzed the data of patients with unilateral femoral neck fracture who underwent surgery in the database of the Second Affiliated Hospital of Wenzhou Medical University from January 1, 2015, to January 1, 2020.

Participants

The inclusion criteria were:1) Unilateral femoral neck fracture, 2) No other fractures, 3) Age > 60 years, 4) LCI was diagnosed based on cranial CT images. The exclusion criteria were: 1) No CT scan of the head, 2) Previous history of brain disease, 3) With other fractures, 4) Pathological fracture, 5) Missing clinical characteristics or laboratory data.

Variables

The study included 161 elderly patients (age > 60 years) with unilateral femoral neck fracture. Patients were divided into two groups based on lacunar cerebral infarct (diagnosis based on cranial CT image): the LCI (lacunar cerebral infarct) group and the non-LCI group. Preoperative clinical characteristics and preoperative laboratory data were collected for all patients. These preoperative variables include age, BMI (body mass index), injury mechanism, gender, types of admission, current drinking, current smoking, injured limb, coronary heart disease, hypertension, diabetes, fatty liver, kidney stone, gallbladder stone, pulmonary nodule, pneumonia, thrombus of lower limb, VTE (venous thrombus embolism), Barthel index [14], education level, prealbumin, total protein, albumin, globulin, A/G (albumin/globulin), AST (aspartate aminotransferase), ALT (alanine aminotransferase), ASL/ALT, ALP (alkaline phosphatase), GGT (gamma-glutamyltransferase), total bilirubin, direct bilirubin, indirect bilirubin, FBG (fasting blood glucose) cystatin-C, BUN (blood urea nitrogen), Scr (serum creatinine), BUN/Scr, CPK (creatine phosphate kinase), homocysteine, LDH (lactate dehydrogenase), blood uric acid, serum sodium, serum kalium, serum calcium, serum chlorine, WBC (white blood cell), neutrophils, lymphocyte, monocyte, eosinophilic granulocyte, basophilic granulocyte, RBC (red blood cell), hemoglobin, hematocrit, MCV (mean corpuscular volume), MCH (mean corpuscular hemoglobin), MCHC (mean corpusular hemoglobin concerntration), platelet count, plateletcrit, PDW (platelet distribution width), MPV (mean platelet volume), PT (prothrombin time), INR (international normalized ratio), APTT (activated partial thromboplastin time), TT (thrombin time), fibrinogen and d-dimer.

Statistics

Data distribution was tested using the Shapiro-Wilk test. As appropriate, patient characteristics were described using median (interquartile range [IQR]) and mean ± standard deviation, frequency, and percentage. A nonparametric test (Mann-Whitney U test or Kruskal-Wallis test) was applied for data with non-normal distribution or heterogeneity of variances. Categorical variables were expressed as percentages and analyzed using the Pearson Chi-squared test. Univariate logistic regression analysis determined the independent risk factors for lacunar cerebral infarct. For the independent risk factors of LCI, the multivariable logistic regression included risk factors significantly associated with LCI in the univariate analysis (P < 0.1).

This study incorporated risk factors selected by univariate and multivariable logistics regression analysis into the machine learning model-like features. To make the ML model more accurate, we conducted zero-mean normalization of the data. Moreover, we randomly split our dataset into two groups: the training sets (70%) for ML model development and the validation sets (30%) for performance evaluation. Besides, we developed five types of ML algorithms to model our data: Logistic regression (LR), Gradient boosting machine (GBM), Extreme gradient boosting (XGBoost), Random forest (RF) and Decision tree (DT). In order to make the ML model more reliable and stable, we carried out a 10-fold cross-validation on the data set during training. Then the ML algorithms were further trained to predict the risk of LCI before surgery, and we evaluated the predictive ability of each ML classifier in validation sets where the area under the receiver operating characteristic (AUROC) value and the corresponding sensitivity, specificity, as well as overall accuracy of ML algorithms were all calculated. All statistics were calculated using SPSS software (version 26.0; SPSS Inc., Chicago, IL, USA) and Python 3.7.6 (Python Software Foundation, http://python.org).

Results

Baseline characteristics of the study population

After excluding a series of variables affecting the results of this study, such as the absence of head CT and the history of brain disease, 161 patients were included in this study, including 100 patients in the LCI group and 61 patients in the non-LCI group. There were statistically significant differences between the two groups in age, hypertension, Barthel index, VTE, prealbumin, total protein, albumin, globulin, BUN/Scr, LDH, cholinesterase, serum sodium, serum chlorine, WBC, Neutrophils, PT, TT, fibrinogen, and d-dimer (all P values < 0.05). Moreover, there were no statistically significant differences between the two groups in BMI, education level, gender, current drinking, current smoking, injury mechanism, injured limb, thrombus of the lower limb, types of admission, coronary heart disease, diabetes, pulmonary nodule, fatty liver, pneumonia, kidney stones, gallbladder stone, A/G, AST, ALT, AST/ALT, ALP, GGT, total bilirubin, direct bilirubin, indirect bilirubin, FBG, BUN, Scr, cystatin-C, CPK, homocysteine, blood uric acid, serum kalium, serum calcium, lymphocyte, monocyte, eosinophilic granulocyte, basophilic granulocyte, RBC, hemoglobin, hematocrit, MCV, MCH, MCHC, platelet count, plateletcrit, PDW, MPV, INR, and APTT (Details are shown in Table 1).

Table 1 Comparison of preoperative clinical characteristics and preoperative laboratory data between two groups
Full size table

Logistic regression analysis for independent risk factors of LCI in femoral neck fracture patients

The univariate logistics regressions analysis was applied to the baseline variables, laboratory tests, and comorbidities. Age, BMI, injury mechanism, gender, types of admission, current drinking, current smoking, injured limb, hypertension, diabetes, fatty liver, kidney stone, gallbladder stone, coronary heart disease, pulmonary nodule, pneumonia, thrombosis of the lower limb, Barthel index, VTE, education level, prealbumin, total protein, albumin, globulin, A/G, AST, ALT, ASL/ALT, ALP, GGT, total bilirubin, direct bilirubin, indirect bilirubin, FBG, BUN, Scr, BUN/Scr, cystatin-C, CPK, homocysteine, LDH, cholinesterase, blood uric acid, serum sodium, serum kalium, serum calcium, serum chlorine, WBC, neutrophils, lymphocyte, monocyte, eosinophilic granulocyte, basophilic granulocyte, RBC, hemoglobin, hematocrit, MCV, MCH, MCHC, platelet count, plateletcrit, PDW, MPV, PT, INR, APTT, TT, fibrinogen, and d-dimer were analyzed during the univariate analysis. The significant parameters (p < 0.1): age, current drinking hypertension, kidney stone, Barthel index, prealbumin, total protein, albumin, globulin, AST, FBG, Scr, BUN/Scr, LDH, cholinesterase, serum sodium, serum chlorine, WBC, neutrophils, PT, INR, TT, fibrinogen, and d-dimer were included in multiple logistic regression analysis (Table 2). The results showed that age, WBC, prealbumin, AST, Total protein, globulin, Scr, BUN/Scr, LDH, serum sodium and fibrinogen were independent predictors of LCI in femoral neck fracture patients (Table 3).

Table 2 Univariate logistics regressions analysis of risk factors to femoral neck fracture patients with lacunar cerebral infarct
Full size table
Table 3 Multivariate logistics regressions analysis of risk factors to femoral neck fracture patients with lacunar cerebral infarct
Full size table

Performance of machine learning algorithms

Comparisons of the prediction performance among the 5 ML models in validation sets are detailed in Table 4 and Fig. 2. It turned out that the RF model demonstrated the highest performance in predicting LCI for femoral neck fracture patients before surgery, whose AUROC was 0.95, sensitivity 1.00, specificity 0.81, and accuracy 0.90 in validation sets.

Table 4 Predictive performance comparison of the five types of machine learning algorithms in the validation sets
Full size table
Fig. 2
figure 2

ROC curve analysis of machine learning algorithms for prediction of femoral neck fracture patients with LCI in the validation set. XGB, Extreme gradient boosting; ROC, receiver operating characteristic; AUC, area under the curve; LCI, lacunar cerebral infarction

Full size image

Relative importance of variables in machine learning algorithms

The relative importance of variables in each LCI-predicting ML algorithm is shown in Fig. 3. There are general evidence trends: although the importance of variables in these ML algorithms varies slightly, factors including prealbumin, globulin, fibrinogen and Scr are more critical than other factors such as AST and WBC. The importance of high-ranking variables in the RF model is arranged in descending order: prealbumin, fibrinogen, globulin and Scr.

Fig. 3
figure 3

Relative importance ranking of each input variable for prediction of LCI in the machine learning algorithms. LR, Logistic regression; GBM, Gradient boosting machine; RF, Random forest; DT, Decision tree; XGBoost, Extreme gradient boosting; WBC, white blood cell; AST, aspartate aminotransferase; Scr, serum creatinine; Bun blood urea nitrogen; LDH, lactate dehydrogenase

Full size image

Discussion

Previous studies [15,16,17] only discussed LCI-related risk factors such as renal function. However, these studies only identified risk factors closely associated with LCI and did not integrate these risk factors into a model to predict LCI. Compared with traditional models, machine learning has strong predictive power for classification problems and has been widely used in the medical field [11]. Therefore, to our knowledge, this study is the first to use a ML algorithm to predict whether an elderly patient with a femoral neck fracture has LCI before surgery.

In this study, our team used five popular machine learning models, LR, XGBoost, DT, GBM, and RT, to predict lacunar cerebral infarction in elderly patients with femoral neck fracture before surgery. To our satisfaction, these 5 ML models all showed excellent predictive ability, among which RF had the most robust predictive ability with an AUROC of 0.95. Therefore, clinicians can use the RF model to predict whether elderly patients with femoral neck complicated with LCI before surgery.

These 5 ML models confirmed that prealbumin, globulin, fibrinogen and Scr were important preoperative predictors of LCI in elderly patients with a femoral neck fracture. As a good reflection of the nutritional status of the human body, prealbumin has excellent predictive value for a variety of diseases, such as stroke [18, 19]. The results of this study are similar to those of previous studies [20], suggesting that prealbumin can be an effective indicator of LCI in elderly patients with a femoral neck fracture. Similarly, as an immune indicator, globulin has good predictive value in a range of diseases such as LCI [21, 22]. Zecca B et al. [23] found that globulin levels were strongly associated with ischemic stroke, which partially supports the findings of this study that globulin can predict LCI in elderly patients with a femoral neck fracture. Fibrinogen is closely related to thrombosis, and high fibrinogen levels are often regarded as biomarkers of vascular diseases such as myocardial infarction or stroke [24]. Some studies [25,26,27] have shown that high levels of fibrinogen can predict the occurrence and clinical prognosis of LCI, which further supports the results of this study, suggesting that fibrinogen can effectively predict the combination of LCI in elderly patients with a femoral neck fracture. Scr is one of the most commonly used indicators of kidney function. In addition, some studies have clarified the relationship between renal function and LCI [28,29,30]. Akoudad S et al. [31] demonstrated that renal function was closely related to the lesions of brain microvessels, especially the albumin-to-creatinine ratio. Similarly, Scr was a helpful predictor of LCI in elderly patients with a femoral neck fracture.

Compared with previous studies, this study has the following advantages. First, few studies have explored the risk factors associated with LCI and constructed predictive models. This study is the first to use an ML model to predict LCI in elderly patients with femoral neck fractures before surgery. Secondly, 5 ML models were used in this study to predict LCI, and the prediction effect was excellent. This supports the accuracy and reliability of the results of this study to a certain extent. Finally, the RF model had the most potent predictive power among the 5 ML models, with an AUROC of 0.95. Therefore, clinicians can use the RF model to predict whether elderly patients with femoral neck fracture have LCI before surgery, providing a basis for preoperative head CT examination and better perioperative management of patients.

However, this study has the following limitations. First, head CT is not a routine examination for elderly patients with femoral neck fracture in our institution, resulting in a relatively small number of subjects in this study. Secondly, this study is a retrospective cohort study, which will lead to a certain degree of bias. The ML model in this study consists of regression methods, so the model mainly shows which variable is the most predictive. In addition, the importance value of featured variables in the RF model in this study is relatively small, which may affect the effectiveness of RF to a certain extent. Therefore, a large prospective cohort study is urgently needed in the future to verify the results of this study. Finally, the ML model in this study was established based on patients in our institution, which may cause the lack of universality of the ML model. Therefore, data from external institutions are needed to verify the ML model of this study in the future.

Conclusions

In this study, 5 ML models were developed and validated for patients with femoral neck fracture to predict preoperative LCI. RF model provides a good predictive value with an AUROC of 0.95. Clinicians can better conduct multidisciplinary perioperative management for patients with femoral neck fracture through this model and accelerate postoperative recovery of patients.

Availability of data and materials

The datasets analyzed in the study are available from the corresponding author on reasonable request.

Abbreviations

BMI:

body mass index

VTE:

venous thrombus embolism

WBC:

white blood cell

RBC:

red blood cell

MPV:

mean platelet volume

AST:

aspartate aminotransferase

ALT:

alanine aminotransferase

A/G:

albumin/globulin

CPK:

creatine phosphate kinase

BUN:

blood urea nitrogen

Scr:

serum creatinine

MCHC:

mean corpusular hemoglobin concerntration

ALP:

alkaline phosphatase

GGT:

gamma-glutamyltransferase

FBG:

fasting blood glucose

LDH:

lactate dehydrogenase

MCV:

mean corpuscular volume

PDW:

platelet distribution width

MCH:

mean corpuscular hemoglobin

PT:

prothrombin time

INR:

international normalized ratio

APTT:

activated partial thromboplastin time

TT:

thrombin time

LR:

Logistic regression

GBM:

Gradient boosting machine

RF:

Random forest

DT:

Decision tree

XGBoost:

Extreme gradient boosting

LCI:

lacunar cerebral infarction

ML:

machine learning

References

  1. Alexiou KI, Roushias A, Varitimidis SE, Malizos KN. Quality of life and psychological consequences in elderly patients after a hip fracture: a review. Clin Interv Aging. 2018;13:143–50. https://doi.org/10.2147/cia.S150067.

    Article  PubMed  PubMed Central  Google Scholar 

  2. Cannistraro RJ, Badi M, Eidelman BH, Dickson DW, Middlebrooks EH, Meschia JF. CNS small vessel disease: a clinical review. Neurology. 2019;92:1146–56. https://doi.org/10.1212/wnl.0000000000007654.

    Article  PubMed  PubMed Central  Google Scholar 

  3. Veldman HD, Heyligers IC, Grimm B, Boymans TA. Cemented versus cementless hemiarthroplasty for a displaced fracture of the femoral neck: a systematic review and meta-analysis of current generation hip stems. Bone Joint J. 2017;99-b:421–31. https://doi.org/10.1302/0301-620x.99b4.Bjj-2016-0758.R1.

    Article  CAS  PubMed  Google Scholar 

  4. Ferguson RJ, Palmer AJ, Taylor A, Porter ML, Malchau H, Glyn-Jones S. Hip replacement. Lancet. 2018;392:1662–71. https://doi.org/10.1016/s0140-6736(18)31777-x.

    Article  PubMed  Google Scholar 

  5. Miki K, Nakano M, Aizawa K, Hasebe Y, Kimura Y, Morosawa S, et al. Risk factors and localization of silent cerebral infarction in patients with atrial fibrillation. Heart Rhythm. 2019;16:1305–13. https://doi.org/10.1016/j.hrthm.2019.03.013.

    Article  PubMed  Google Scholar 

  6. Pantoni L, Fierini F, Poggesi A. Thrombolysis in acute stroke patients with cerebral small vessel disease. Cerebrovasc Dis. 2014;37:5–13. https://doi.org/10.1159/000356796.

    Article  CAS  PubMed  Google Scholar 

  7. Kristoffersen DT, Helgeland J, Waage HP, Thalamus J, Clemens D, Lindman AS, et al. Survival curves to support quality improvement in hospitals with excess 30-day mortality after acute myocardial infarction, cerebral stroke and hip fracture: a before-after study. BMJ Open. 2015;5:e006741. https://doi.org/10.1136/bmjopen-2014-006741.

    Article  PubMed  PubMed Central  Google Scholar 

  8. Yuan J, Zhu G, Zhao Y, Huang J. Effect of hip fracture on prognosis of acute cerebral infarction. Clinics. 2021;76:e3059. https://doi.org/10.6061/clinics/2021/e3059.

    Article  PubMed  PubMed Central  Google Scholar 

  9. de Sire A, Invernizzi M, Baricich A, Lippi L, Ammendolia A, Grassi FA, et al. Optimization of transdisciplinary management of elderly with femur proximal extremity fracture: a patient-tailored plan from orthopaedics to rehabilitation. World J Orthop. 2021;12:456–66. https://doi.org/10.5312/wjo.v12.i7.456.

    Article  PubMed  PubMed Central  Google Scholar 

  10. van der Burg DA, Diepstraten M, Wouterse B. Long-term care use after a stroke or femoral fracture and the role of family caregivers. BMC Geriatr. 2020;20:150. https://doi.org/10.1186/s12877-020-01526-7.

    Article  PubMed  PubMed Central  Google Scholar 

  11. Deo RC. Machine learning in medicine. Circulation. 2015;132:1920–30. https://doi.org/10.1161/circulationaha.115.001593.

    Article  PubMed  PubMed Central  Google Scholar 

  12. Ngiam KY, Khor IW. Big data and machine learning algorithms for health-care delivery. Lancet Oncol. 2019;20:e262–e73. https://doi.org/10.1016/s1470-2045(19)30149-4.

    Article  PubMed  Google Scholar 

  13. Feeny AK, Chung MK, Madabhushi A, Attia ZI, Cikes M, Firouznia M, et al. Artificial intelligence and machine learning in arrhythmias and cardiac electrophysiology. Circ Arrhythm Electrophysiol. 2020;13:e007952. https://doi.org/10.1161/circep.119.007952.

    Article  PubMed  PubMed Central  Google Scholar 

  14. Sainsbury A, Seebass G, Bansal A, Young JB. Reliability of the Barthel index when used with older people. Age Ageing. 2005;34:228–32. https://doi.org/10.1093/ageing/afi063.

    Article  PubMed  Google Scholar 

  15. Cai Y, Chen B, Zeng X, Xie M, Wei X, Cai J. The triglyceride glucose index is a risk factor for enlarged perivascular space. Front Neurol. 2022;13:782286. https://doi.org/10.3389/fneur.2022.782286.

    Article  PubMed  PubMed Central  Google Scholar 

  16. Kim SH, Shin DW, Yun JM, Lee JE, Lim JS, Cho BL, et al. Kidney dysfunction and silent brain infarction in generally healthy adults. J Neurol Sci. 2017;379:89–93. https://doi.org/10.1016/j.jns.2017.05.043.

    Article  PubMed  Google Scholar 

  17. Wu RH, Li Q, Tan Y, Liu XY, Huang J. Depression in silent lacunar infarction: a cross-sectional study of its association with location of silent lacunar infarction and vascular risk factors. Neurol Sci. 2014;35:1553–9. https://doi.org/10.1007/s10072-014-1794-5.

    Article  PubMed  PubMed Central  Google Scholar 

  18. Li C, Yang C, Zhu J, Huang H, Zheng J, Hu X, et al. Predictive value of globulin to Prealbumin ratio for 3-month functional outcomes in acute ischemic stroke patients. Dis Markers. 2022;2022:1120192. https://doi.org/10.1155/2022/1120192.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  19. Loftus TJ, Brown MP, Slish JH, Rosenthal MD. Serum levels of Prealbumin and albumin for preoperative risk stratification. Nutr Clin Pract. 2019;34:340–8. https://doi.org/10.1002/ncp.10271.

    Article  PubMed  Google Scholar 

  20. Zhang SQ, Peng B, Stary CM, Jian ZH, Xiong XX, Chen QX. Serum prealbumin as an effective prognostic indicator for determining clinical status and prognosis in patients with hemorrhagic stroke. Neural Regen Res. 2017;12:1097–102. https://doi.org/10.4103/1673-5374.211188.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  21. Lamb YN, Syed YY, Dhillon S. Immune globulin subcutaneous (human) 20% (Hizentra(®)): a review in chronic inflammatory demyelinating polyneuropathy. CNS Drugs. 2019;33:831–8. https://doi.org/10.1007/s40263-019-00655-x.

    Article  CAS  PubMed  Google Scholar 

  22. Shah AB, Beamer N, Coull BM. Enhanced in vivo platelet activation in subtypes of ischemic stroke. Stroke. 1985;16:643–7. https://doi.org/10.1161/01.str.16.4.643.

    Article  CAS  PubMed  Google Scholar 

  23. Zecca B, Mandelli C, Maino A, Casiraghi C, Bolla G, Consonni D, et al. A bioclinical pattern for the early diagnosis of cardioembolic stroke. Emerg Med Int. 2014;2014:242171. https://doi.org/10.1155/2014/242171.

    Article  PubMed  PubMed Central  Google Scholar 

  24. Grottke O, Mallaiah S, Karkouti K, Saner F, Haas T. Fibrinogen supplementation and its indications. Semin Thromb Hemost. 2020;46:38–49. https://doi.org/10.1055/s-0039-1696946.

    Article  CAS  PubMed  Google Scholar 

  25. Fan H, Yang S, Li Y, Yin J, Qin W, Yang L, et al. Assessment of homocysteine as a diagnostic and early prognostic biomarker for patients with acute lacunar infarction. Eur Neurol. 2018;79:54–62. https://doi.org/10.1159/000484893.

    Article  CAS  PubMed  Google Scholar 

  26. Li L, He S, Liu H, Pan M, Dai F. Potential risk factors of persistent postural-perceptual dizziness: a pilot study. J Neurol. 2021. https://doi.org/10.1007/s00415-021-10899-7.

  27. Zheng L, Wang Z, Liu J, Yang X, Zhang S, Hao Z, et al. Association between admission blood fibrinogen-to-albumin ratio and clinical outcomes after acute lacunar stroke. Biomark Med. 2021;15:87–96. https://doi.org/10.2217/bmm-2019-0537.

    Article  CAS  PubMed  Google Scholar 

  28. Peralta CA, McClure LA, Scherzer R, Odden MC, White CL, Shlipak M, et al. Effect of intensive versus usual blood pressure control on kidney function among individuals with prior lacunar stroke: a post hoc analysis of the secondary prevention of small subcortical strokes (SPS3) randomized trial. Circulation. 2016;133:584–91. https://doi.org/10.1161/circulationaha.115.019657.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  29. Saji N, Sato T, Sakuta K, Aoki J, Kobayashi K, Matsumoto N, et al. Chronic kidney disease is an independent predictor of adverse clinical outcomes in patients with recent small subcortical infarcts. Cerebrovasc Dis Extra. 2014;4:174–81. https://doi.org/10.1159/000365565.

    Article  PubMed  PubMed Central  Google Scholar 

  30. Yang L, Yu L, Kong Y, Zhang X, Li Y, Yang S, et al. Relationship between white matter hyperintensities and chronic kidney disease in patients with acute lacunar stroke. Neurol Sci. 2020;41:3307–13. https://doi.org/10.1007/s10072-020-04397-3.

    Article  PubMed  Google Scholar 

  31. Akoudad S, Sedaghat S, Hofman A, Koudstaal PJ, van der Lugt A, Ikram MA, et al. Kidney function and cerebral small vessel disease in the general population. Int J Stroke. 2015;10:603–8. https://doi.org/10.1111/ijs.12465.

    Article  PubMed  Google Scholar 

Download references

Acknowledgments

The authors thank all the colleagues for their valuable assistance during the implementation of this study.

Funding

This work was supported by the Wenzhou Science and Technology Plan Project (Grant Y20210041). And all authors have read and agreed to the published version of the manuscript.

Author information

Author notes

  1. Cheng-bin Huang and Kai Tan contributed equally to this work.

Authors and Affiliations

  1. Department of Orthopaedic Surgery, The Second Affiliated Hospital and Yuying Childrens Hospital of Wenzhou Medical University, Wenzhou, 325000, China

    Cheng-bin Huang, Kai Tan, Zong-yi Wu & Lei Yang

  2. Key Laboratory of Orthopaedics of Zhejiang Province, Wenzhou, 325000, China

    Cheng-bin Huang, Kai Tan & Lei Yang

Authors
  1. Cheng-bin Huang
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Kai Tan
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Zong-yi Wu
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Lei Yang
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

C.B.H and K.T designed the study; collected, analyzed, and interpreted data. Z.Y.W interpreted data and obtained funding. L.Y designed and supervised the study and drafted the manuscript. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Lei Yang.

Ethics declarations

Ethics approval and consent to participate

All procedures performed in studies involving human participants were in accordance with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. The need of ethics approval was waived by the Medical Ethics Committee, Yuying Children’s Hospital, The Second Affiliated Hospital of Wenzhou Medical University. And the informed consent was waived by the Medical Ethics Committee, Yuying Children’s Hospital, The Second Affiliated Hospital of Wenzhou Medical University.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Huang, Cb., Tan, K., Wu, Zy. et al. Application of machine learning model to predict lacunar cerebral infarction in elderly patients with femoral neck fracture before surgery. BMC Geriatr 22, 912 (2022). https://doi.org/10.1186/s12877-022-03631-1

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s12877-022-03631-1

Keywords

BMC Geriatrics

ISSN: 1471-2318

Contact us