Estimation of Parkinson's disease severity using speech features and extreme gradient boosting

Diagnosis of Parkinson's disease based on voice signals using SHAP and hard voting ensemble method

Parkinson's disease (PD) is the second most common progressive neurological condition after Alzheimer's. The significant number of individuals afflicted with this illness makes it essential to develop a method to diagnose the conditions in their early phases. PD is typically identified from motor symptoms or via other Neuroimaging techniques. Expensive, time-consuming, and unavailable to the general public, these methods are not very accurate. Another issue to be addressed is the black-box nature of machine learning methods that needs interpretation. These issues encourage us to develop a novel technique using Shapley additive explanations (SHAP) and Hard Voting Ensemble Method based on voice signals to diagnose PD more accurately. Another purpose of this study is to interpret the output of the model and determine the most important features in diagnosing PD. The present article uses Pearson Correlation Coefficients to understand the relationship between input features and the output. Input features with high correlation are selected and then classified by the Extreme Gradient Boosting, Light Gradient Boosting Machine, Gradient Boosting, and Bagging. Moreover, the weights in Hard Voting Ensemble Method are determined based on the performance of the mentioned classifiers. At the final stage, it uses SHAP to determine the most important features in PD diagnosis. The effectiveness of the proposed method is validated using 'Parkinson Dataset with Replicated Acoustic Features' from the UCI machine learning repository. It has achieved an accuracy of 85.42%. The findings demonstrate that the proposed method outperformed state-of-the-art approaches and can assist physicians in diagnosing Parkinson's cases.

Predictive Data Analytics in Telecare and Telehealth: Systematic Scoping Review

BACKGROUND

Telecare and telehealth are important care-at-home services used to support individuals to live more independently at home. Historically, these technologies have reactively responded to issues. However, there has been a recent drive to make better use of the data from these services to facilitate more proactive and predictive care.

OBJECTIVE

This review seeks to explore the ways in which predictive data analytics techniques have been applied in telecare and telehealth in at-home settings.

METHODS

The PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews) checklist was adhered to alongside Arksey and O'Malley's methodological framework. English language papers published in MEDLINE, Embase, and Social Science Premium Collection between 2012 and 2022 were considered and results were screened against inclusion or exclusion criteria.

RESULTS

In total, 86 papers were included in this review. The types of analytics featuring in this review can be categorized as anomaly detection (n=21), diagnosis (n=32), prediction (n=22), and activity recognition (n=11). The most common health conditions represented were Parkinson disease (n=12) and cardiovascular conditions (n=11). The main findings include: a lack of use of routinely collected data; a dominance of diagnostic tools; and barriers and opportunities that exist, such as including patient-reported outcomes, for future predictive analytics in telecare and telehealth.

CONCLUSIONS

All papers in this review were small-scale pilots and, as such, future research should seek to apply these predictive techniques into larger trials. Additionally, further integration of routinely collected care data and patient-reported outcomes into predictive models in telecare and telehealth offer significant opportunities to improve the analytics being performed and should be explored further. Data sets used must be of suitable size and diversity, ensuring that models are generalizable to a wider population and can be appropriately trained, validated, and tested.

Patient-specific game-based transfer method for Parkinson's disease severity prediction

Dysphonia is one of the early symptoms of Parkinson's disease (PD). Most existing methods use feature selection methods to find the optimal subset of voice features for all PD patients. Few have considered the heterogeneity between patients, which implies the need to provide specific prediction models for different patients. However, building the specific model faces the challenge of small sample size, which makes it lack generalization ability. Instance transfer is an effective way to solve this problem. Therefore, this paper proposes a patient-specific game-based transfer (PSGT) method for PD severity prediction. First, a selection mechanism is used to select PD patients with similar disease trends to the target patient from the source domain, which reduces the risk of negative transfer. Then, the contribution of the transferred subjects and their instances to the disease estimation of the target subject is fairly evaluated by the Shapley value, which improves the interpretability of the method. Next, the proportion of valid instances in the transferred subjects is determined, and the instances with higher contribution are transferred to further reduce the difference between the transferred instance subset and the target subject. Finally, the selected subset of instances is added to the training set of the target subject, and the extended data is fed into the random forest to improve the performance of the method. Parkinson's telemonitoring dataset is used to evaluate the feasibility and effectiveness. The mean values of mean absolute error, root mean square error, and volatility obtained by predicting motor-UPDRS and total-UPDRS for target patients are 1.59, 1.95, 1.56 and 1.98, 2.54, 1.94, respectively. Experiment results show that the PSGT has better performance in both prediction error and stability over compared methods.

Machine Learning-Based Classification of Parkinson's Disease Patients Using Speech Biomarkers

BACKGROUND

Parkinson's disease (PD) is the most prevalent neurodegenerative movement disorder and a growing health concern in demographically aging societies. The prevalence of PD among individuals over the age of 60 and 80 years has been reported to range between 1% and 4%. A timely diagnosis of PD is desirable, even though it poses challenges to medical systems.

OBJECTIVE

This study aimed to classify PD and healthy controls based on the analysis of voice records at different frequencies using machine learning (ML) algorithms.

METHODS

The voices of 252 individuals aged 33 to 87 years were recorded. Based on the voice record data, ML algorithms can distinguish PD patients and healthy controls. One binary decision variable was associated with 756 instances and 754 attributes. Voice records data were analyzed through supervised ML algorithms and pipelines. A 10-fold cross-validation method was used to validate models.

RESULTS

In the classification of PD patients, ML models were performed with 84.21 accuracy, 93 precision, 89 Sensitivity, 89 F1-scores, and 87 AUC. The pipeline performance improved to accuracy: 85.09, precision: 92, Sensitivity:91, F1-score: 89, and AUC: 90. The Pipeline methods improved the performance of classifying PD from voice record.

CONCLUSIONS

Our study demonstrated that ML classifiers and pipelines can classify PD patients based on speech biomarkers. It was found that pipelines were more effective at selecting the most relevant features from high-dimensional data and at accurately classifying PD patients and healthy controls. This approach can therefore be used for early diagnosis of initial forms of PD.

Robust and language-independent acoustic features in Parkinson's disease

Introduction

The analysis of vocal samples from patients with Parkinson's disease (PDP) can be relevant in supporting early diagnosis and disease monitoring. Intriguingly, speech analysis embeds several complexities influenced by speaker characteristics (e.g., gender and language) and recording conditions (e.g., professional microphones or smartphones, supervised, or non-supervised data collection). Moreover, the set of vocal tasks performed, such as sustained phonation, reading text, or monologue, strongly affects the speech dimension investigated, the feature extracted, and, as a consequence, the performance of the overall algorithm.

Methods

We employed six datasets, including a cohort of 176 Healthy Control (HC) participants and 178 PDP from different nationalities (i.e., Italian, Spanish, Czech), recorded in variable scenarios through various devices (i.e., professional microphones and smartphones), and performing several speech exercises (i.e., vowel phonation, sentence repetition). Aiming to identify the effectiveness of different vocal tasks and the trustworthiness of features independent of external co-factors such as language, gender, and data collection modality, we performed several intra- and inter-corpora statistical analyses. In addition, we compared the performance of different feature selection and classification models to evaluate the most robust and performing pipeline.

Results

According to our results, the combined use of sustained phonation and sentence repetition should be preferred over a single exercise. As for the set of features, the Mel Frequency Cepstral Coefficients demonstrated to be among the most effective parameters in discriminating between HC and PDP, also in the presence of heterogeneous languages and acquisition techniques.

Conclusion

Even though preliminary, the results of this work can be exploited to define a speech protocol that can effectively capture vocal alterations while minimizing the effort required to the patient. Moreover, the statistical analysis identified a set of features minimally dependent on gender, language, and recording modalities. This discloses the feasibility of extensive cross-corpora tests to develop robust and reliable tools for disease monitoring and staging and PDP follow-up.

Computerized analysis of speech and voice for Parkinson's disease: A systematic review

BACKGROUND AND OBJECTIVE

Speech impairment is an early symptom of Parkinson's disease (PD). This study has summarized the literature related to speech and voice in detecting PD and assessing its severity.

METHODS

A systematic review of the literature from 2010 to 2021 to investigate analysis methods and signal features. The keywords "Automatic analysis" in conjunction with "PD speech" or "PD voice" were used, and the PubMed and ScienceDirect databases were searched. A total of 838 papers were found on the first run, of which 189 were selected. One hundred and forty-seven were found to be suitable for the review. The different datasets, recording protocols, signal analysis methods and features that were reported are listed. Values of the features that separate PD patients from healthy controls were tabulated. Finally, the barriers that limit the wide use of computerized speech analysis are discussed.

RESULTS

Speech and voice may be valuable markers for PD. However, large differences between the datasets make it difficult to compare different studies. In addition, speech analytic methods that are not informed by physiological understanding may alienate clinicians.

CONCLUSIONS

The potential usefulness of speech and voice for the detection and assessment of PD is confirmed by evidence from the classification and correlation results.

Artificial Intelligence and Its Application to Minimal Hepatic Encephalopathy Diagnosis

Hepatic encephalopathy (HE) is a brain dysfunction caused by liver insufficiency and/or portosystemic shunting. HE manifests as a spectrum of neurological or psychiatric abnormalities. Diagnosis of overt HE (OHE) is based on the typical clinical manifestation, but covert HE (CHE) has only very subtle clinical signs and minimal HE (MHE) is detected only by specialized time-consuming psychometric tests, for which there is still no universally accepted gold standard. Significant progress has been made in artificial intelligence and its application to medicine. In this review, we introduce how artificial intelligence has been used to diagnose minimal hepatic encephalopathy thus far, and we discuss its further potential in analyzing speech and handwriting data, which are probably the most accessible data for evaluating the cognitive state of the patient.

Prediction of readmission in patients with acute exacerbation of chronic obstructive pulmonary disease within one year after treatment and discharge

Background

To investigate the risk factors and construct a logistic model and an extreme gradient boosting (XGBoost) model to compare the predictive performances for readmission in acute exacerbation of chronic obstructive pulmonary disease (AECOPD) patients within one year.

Methods

In total, 636 patients with AECOPD were recruited and divided into readmission group (n = 449) and non-readmission group (n = 187). Backward stepwise regression method was used to analyze the risk factors for readmission. Data were divided into training set and testing set at a ratio of 7:3. Variables with statistical significance were included in the logistic model and variables with P < 0.1 were included in the XGBoost model, and receiver operator characteristic (ROC) curves were plotted.

Results

Patients with acute exacerbations within the previous 1 year [odds ratio (OR) = 4.086, 95% confidence interval (CI) 2.723–6.133, P < 0.001), long-acting β agonist (LABA) application (OR = 4.550, 95% CI 1.587–13.042, P = 0.005), inhaled corticosteroids (ICS) application (OR = 0.227, 95% CI 0.076–0.672, P = 0.007), glutamic-pyruvic transaminase (ALT) level (OR = 0.985, 95% CI 0.971–0.999, P = 0.042), and total CAT score (OR = 1.091, 95% CI 1.048–1.136, P < 0.001) were associated with the risk of readmission. The AUC value of the logistic model was 0.743 (95% CI 0.692–0.795) in the training set and 0.699 (95% CI 0.617–0.780) in the testing set. The AUC value of XGBoost model was 0.814 (95% CI 0.812–0.815) in the training set and 0.722 (95% CI 0.720–0.725) in the testing set.

Conclusions

The XGBoost model showed a better predictive value in predicting the risk of readmission within one year in the AECOPD patients than the logistic regression model. The findings of our study might help identify patients with a high risk of readmission within one year and provide timely treatment to prevent the reoccurrence of AECOPD.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12890-021-01692-3.

Comparison of Prediction Models for Acute Kidney Injury Among Patients with Hepatobiliary Malignancies Based on XGBoost and LASSO-Logistic Algorithms

Background

Based on the admission data, we applied the XGBoost algorithm to create a prediction model to estimate the AKI risk in patients with hepatobiliary malignancies and then compare its prediction capacity with the logistic model.

Methods

We reviewed clinical data of 7968 and 589 liver/gallbladder cancer patients admitted to Zhongshan Hospital during 2014 and 2015. They were randomly divided into the training set and test set. Data were collected from the electronic medical record system. XGBoost and LASSO-logistic were used to develop prediction models, respectively. The performance measures included the classification matrix, the area under the receiver operating characteristic curve (AUC), lift chart and learning curve.

Results

Of 6846 participants in the training set, 792 (11.6%) cases developed AKI. In XGBoost model, the top 3 most important variables for AKI were serum creatinine (SCr), glomerular filtration rate (eGFR) and antitumor treatment in liver cancer patients. Similarly, SCr and eGFR also ranked second and third most important variables in the gallbladder cancer-related AKI model just after phosphorus. In the classification matrix, XGBoost model possessed a comparably better agreement between the actual observations and the predictions than LASSO-logistic model. The Youden’s index of XGBoost model was 47.5% and 59.3%, respectively, which was significantly higher than that of LASSO-logistic model (41.6% and 32.7%). The AUCs of XGBoost model were 0.822 in liver cancer and 0.850 in gallbladder cancer. By comparison, the AUC values of Logistic models were significantly lower as 0.793 and 0.740 (p=0.024 and 0.018). With the accumulation of training samples, XGBoost model maintained greater robustness in the learning curve.

Conclusion

XGBoost model based on admission data has higher accuracy and stronger robustness in predicting AKI. It will benefit AKI risk classification management in clinical practice and take an advanced intervention among patients with hepatobiliary malignancies.