Diagnosis of Acute Central Dizziness With Simple Clinical Information Using Machine Learning

Risk factors for imaging abnormalities in patients with dizziness complaints: an algorithm for ordering brain imaging

AIM

The diagnostic detection of abnormal findings with head imaging is low for dizziness. This study aimed to investigate the risk factors associated with abnormal computed tomography (CT) or magnetic resonance imaging (MRI) findings for patients with dizziness.

MATERIALS AND METHODS

Medical records of patients who had CT or MRI examinations for dizziness complaints between January 1, 2019, and December 31, 2020, were retrospectively reviewed. Imaging outcomes were grouped as normal or abnormal findings. Risk factors, including demographics, dizziness pattern, symptoms, comorbidities, and medical history were assessed. A Chi-square automatic interaction detection decision tree model was used to classify abnormal imaging findings based on risk factors identified through multivariable analyses.

RESULTS

A total of 2,342 scans were examined. Detection of abnormal findings was 4.8% (n = 96), including acute cerebral infarction (n = 33), acute cranial hemorrhage (n = 15), cancer/tumor-like lesions (n = 27), and inner ear abnormalities (n = 21). The risk factor most indicative of abnormal findings were loss of consciousness and neurologic deficit (Odds Ratio 55.57, p < 0.001). The likelihood of abnormality indicating acute brain lesions was 44.4% for patients with loss of consciousness and neurologic deficits. Loss of consciousness and neurologic deficits, hearing loss, nausea/vomiting, and comorbid malignancy distinguished abnormal findings from negative imaging findings (AUC 0.729; 95%CI 0.672-0.785; p < 0.001). Patients with unspecific dizziness complaints were less likely to have abnormal imaging findings.

CONCLUSION

These findings highlighted the significance of specific risk factors in recognizing individuals with dizziness complaints who may have abnormal imaging findings indicative of serious diseases. Further studies are warranted to verify the findings.

Neurological update: neuro-otology 2023

Much has changed since our last review of recent advances in neuro-otology 7 years ago. Unfortunately there are still not many practising neuro-otologists, so that most patients with vestibular problems need, in the first instance, to be evaluated and treated by neurologists whose special expertise is not neuro-otology. The areas we consider here are mostly those that almost any neurologist should be able to start managing: acute spontaneous vertigo in the Emergency Room-is it vestibular neuritis or posterior circulation stroke; recurrent spontaneous vertigo in the office-is it vestibular migraine or Meniere's disease and the most common vestibular problem of all-benign positional vertigo. Finally we consider the future: long-term vestibular monitoring and the impact of machine learning on vestibular diagnosis.

An explainable artificial intelligence framework for risk prediction of COPD in smokers

BACKGROUND

Since the inconspicuous nature of early signs associated with Chronic Obstructive Pulmonary Disease (COPD), individuals often remain unidentified, leading to suboptimal opportunities for timely prevention and treatment. The purpose of this study was to create an explainable artificial intelligence framework combining data preprocessing methods, machine learning methods, and model interpretability methods to identify people at high risk of COPD in the smoking population and to provide a reasonable interpretation of model predictions.

METHODS

The data comprised questionnaire information, physical examination data and results of pulmonary function tests before and after bronchodilatation. First, the factorial analysis for mixed data (FAMD), Boruta and NRSBoundary-SMOTE resampling methods were used to solve the missing data, high dimensionality and category imbalance problems. Then, seven classification models (CatBoost, NGBoost, XGBoost, LightGBM, random forest, SVM and logistic regression) were applied to model the risk level, and the best machine learning (ML) model's decisions were explained using the Shapley additive explanations (SHAP) method and partial dependence plot (PDP).

RESULTS

In the smoking population, age and 14 other variables were significant factors for predicting COPD. The CatBoost, random forest, and logistic regression models performed reasonably well in unbalanced datasets. CatBoost with NRSBoundary-SMOTE had the best classification performance in balanced datasets when composite indicators (the AUC, F1-score, and G-mean) were used as model comparison criteria. Age, COPD Assessment Test (CAT) score, gross annual income, body mass index (BMI), systolic blood pressure (SBP), diastolic blood pressure (DBP), anhelation, respiratory disease, central obesity, use of polluting fuel for household heating, region, use of polluting fuel for household cooking, and wheezing were important factors for predicting COPD in the smoking population.

CONCLUSION

This study combined feature screening methods, unbalanced data processing methods, and advanced machine learning methods to enable early identification of COPD risk groups in the smoking population. COPD risk factors in the smoking population were identified using SHAP and PDP, with the goal of providing theoretical support for targeted screening strategies and smoking population self-management strategies.

Machine learning is an effective method to predict the 90-day prognosis of patients with transient ischemic attack and minor stroke

OBJECTIVE

We aimed to investigate factors related to the 90-day poor prognosis (mRS≥3) in patients with transient ischemic attack (TIA) or minor stroke, construct 90-day poor prognosis prediction models for patients with TIA or minor stroke, and compare the predictive performance of machine learning models and Logistic model.

METHOD

We selected TIA and minor stroke patients from a prospective registry study (CNSR-III). Demographic characteristics,smoking history, drinking history(≥20g/day), physiological data, medical history,secondary prevention treatment, in-hospital evaluation and education,laboratory data, neurological severity, mRS score and TOAST classification of patients were assessed. Univariate and multivariate logistic regression analyses were performed in the training set to identify predictors associated with poor outcome (mRS≥3). The predictors were used to establish machine learning models and the traditional Logistic model, which were randomly divided into the training set and test set according to the ratio of 70:30. The training set was used to construct the prediction model, and the test set was used to evaluate the effect of the model. The evaluation indicators of the model included the area under the curve (AUC) of the discrimination index and the Brier score (or calibration plot) of the calibration index.

RESULT

A total of 10967 patients with TIA and minor stroke were enrolled in this study, with an average age of 61.77 ± 11.18 years, and women accounted for 30.68%. Factors associated with the poor prognosis in TIA and minor stroke patients included sex, age, stroke history, heart rate, D-dimer, creatinine, TOAST classification, admission mRS, discharge mRS, and discharge NIHSS score. All models, both those constructed by Logistic regression and those by machine learning, performed well in predicting the 90-day poor prognosis (AUC >0.800). The best performing AUC in the test set was the Catboost model (AUC=0.839), followed by the XGBoost, GBDT, random forest and Adaboost model (AUCs equal to 0.838, 0, 835, 0.832, 0.823, respectively). The performance of Catboost and XGBoost in predicting poor prognosis at 90-day was better than the Logistic model, and the difference was statistically significant(P<0.05). All models, both those constructed by Logistic regression and those by machine learning had good calibration.

CONCLUSION

Machine learning algorithms were not inferior to the Logistic regression model in predicting the poor prognosis of patients with TIA and minor stroke at 90-day. Among them, the Catboost model had the best predictive performance. All models provided good discrimination.

Risk Factors and Risk Model Construction of Stroke in Patients with Vertigo in Emergency Department

Objective

We aimed to explore the risk factors of stroke in patients with vertigo in the emergency department and establish a risk prediction model for stroke patients.

Methods

A total of 301 patients experiencing vertigo in our hospital from January 2020 to January 2021 were retrospectively included. Patients were divided into the stroke group (n = 56) and the nonstroke group (n = 245). The clinical characteristics of patients in both groups were collected and compared, followed by binary logistic regression that was employed to determine the risk factors that affect stroke diagnosis. The receiver operating characteristic (ROC) curve was used to clarify the effectiveness of the constructed model.

Results

Patients in the stroke group were older and had higher systolic and diastolic blood pressure on admission than the nonstroke group. Meanwhile, they demonstrated a higher proportion of diabetes and atrial fibrillation and focal muscle weakness, dysarthria, dysphagia, or ataxia in neurological examinations compared to the nonstroke group (all P < 0.05). The proportion of patients in the nonstroke group who had a history of vertigo or inner ear disease was significantly higher than that in the stroke group (P < 0.05). The patient's age ≥ 60 years old (OR = 3.57), diabetes (OR = 4.57), atrial fibrillation (OR = 4.26), previous history of vertigo or inner ear disease (OR = 0.16), focal muscle weakness (OR = 4.34), and dysphagia or ataxia (OR = 4.08) were associated with a higher risk of stroke. The area under the curve for stroke was 0.87, and the sensitivity and specificity were 98.2% and 57.6%, respectively, as the sum of the assigned scores was greater than 3.

Conclusions

Age ≥ 60 years old, diabetes, atrial fibrillation, previous history of vertigo or inner ear disease, focal muscle weakness, dysphagia, or ataxia were associated with a higher risk of stroke. The risk model constructed based on our findings may help to assess the risk of stroke in patients with vertigo in the emergency department.

A Dynamic Nomogram to Predict the Risk of Stroke in Emergency Department Patients With Acute Dizziness

Objective

To develop a risk prediction tool for acute ischemic stroke (AIS) for patients presenting to the emergency department (ED) with acute dizziness/vertigo or imbalance.

Method

A prospective, multicenter cohort study was designed, and adult patients presenting with dizziness/vertigo or imbalance within 14 days were consecutively enrolled from the EDs of 4 tertiary hospitals between August 10, 2020, and June 10, 2021. Stroke was diagnosed by CT or MRI performed within 14 days of symptom onset. Participants were followed-up for 30 days. The least absolute shrinkage and selection operator (LASSO) logistic regression analysis was conducted to extract predictive factors that best identified patients at high risk of stroke to establish a prediction model. Model discrimination and calibration were assessed and its prediction performance was compared with the age, blood pressure, clinical features, duration, and diabetes (ABCD2) score, nystagmus scheme, and finger to nose test.

Results

In this study, 790 out of 2,360 patients were enrolled {median age, 60.0 years [interquartile range (IQR), 51–68 years]; 354 (44.8%) men}, with complete follow-up data available. AIS was identified in 80 patients. An online web service tool (https://neuroby.shinyapps.io/dynnomapp/) was developed for stroke risk prediction, including the variables of sex, trigger, isolated symptom, nausea, history of brief dizziness, high blood pressure, finger to nose test, and tandem gait test. The model exhibited excellent discrimination with an area under the receiver operating characteristic (ROC) curve (AUC) of 0.889 (95% CI: 0.855–0.923), compared with the ABCD2 score, nystagmus scheme, and finger to nose test [0.712 (95% CI, 0.652–0.771), 0.602 (95% CI, 0.556–0.648), and 61.7 (95% CI, 0.568–0.666) respectively].

Conclusion

Our new prediction model exhibited good performance and could be useful for stroke identification in patients presenting with dizziness, vertigo, or imbalance. Further externally validation study is needed to increase the strength of our findings.

Machine Learning Techniques for Differential Diagnosis of Vertigo and Dizziness: A Review

Vertigo is a sensation of movement that results from disorders of the inner ear balance organs and their central connections, with aetiologies that are often benign and sometimes serious. An individual who develops vertigo can be effectively treated only after a correct diagnosis of the underlying vestibular disorder is reached. Recent advances in artificial intelligence promise novel strategies for the diagnosis and treatment of patients with this common symptom. Human analysts may experience difficulties manually extracting patterns from large clinical datasets. Machine learning techniques can be used to visualize, understand, and classify clinical data to create a computerized, faster, and more accurate evaluation of vertiginous disorders. Practitioners can also use them as a teaching tool to gain knowledge and valuable insights from medical data. This paper provides a review of the literatures from 1999 to 2021 using various feature extraction and machine learning techniques to diagnose vertigo disorders. This paper aims to provide a better understanding of the work done thus far and to provide future directions for research into the use of machine learning in vertigo diagnosis.