Random Forest Model in the Diagnosis of Dementia Patients with Normal Mini-Mental State Examination Scores

Factors associated with a high level of suicide risk among patients with late-life depression: a cross-sectional study from a tertiary psychiatric hospital in Guangzhou China

BACKGROUND

As global aging accelerates, depression among the elderly becomes more common. Research had revealed that patients with late-life depression (LLD) face a higher risk of suicide compared to their counterparts in other age groups, with the pathways to suicide being multifaceted. Thus, investigating the various factors linked to the elevated risk of suicide in patients with LLD is critical.

OBJECTIVE

To investigate the factors associated with a high level of suicide risk among patients with LLD.

METHODS

A total of 108 patients with LLD were recruited for this study. From October 2022 to November 2023, a cross-sectional study was conducted on patients with LLD from the Affiliated Brain Hospital of Guangzhou Medical University. Suicide risk was evaluated using the Chinese version of the Nurses' Global Assessment of Suicide Risk Scale (NGASR). Potential influencing factors were included and analyzed through multivariate linear regression to identify the factors associated with a high level of suicide risk among patients with LLD.

RESULTS

The mean NGASR score among patients with LLD was 7.30 ± 4.34 (range: 0 ~ 19). Multiple linear regression analyses revealed that depression-anxiety of the Brief Psychiatric Rating Scale (BPRS) (β = 0.31, 95% CI = 0.13, 0.45, p<0.001), activation of the BPRS (β=-0.29, 95% CI=-1.22, -0.35, p<0.001), normal cognitive function of the Mini-Mental State Examination (MMSE) (β = 0.21, 95% CI = 0.50, 3.48, p<0.05), involuntary admission (β = 0.20, 95% CI = 0.44, 3.43, p<0.05), and objective support of the Social Support Rating Scale (SSRS) (β = 0.21, 95% CI = 0.08, 0.66, p<0.05) were statistically associated with a high level of suicide risk in patients with LLD.

CONCLUSION

This study found that LLD patients with severe depression-anxiety, low activation, normal cognitive function, involuntary admission, and strong objective support exhibited a high level of suicide risk. These patients should receive intensified monitoring and comprehensive measures should be implemented to prevent the occurrence of suicidal behaviors during hospitalization.

Comparison of machine learning algorithms for predicting cognitive impairment using neuropsychological tests

OBJECTIVES

Neuropsychological tests (NPTs) are standard tools for assessing cognitive function. These tools can evaluate the cognitive status of a subject, which can be time-consuming and expensive for interpretation. Therefore, this paper aimed to optimize the systematic NPTs by machine learning and develop new classification models for differentiating healthy controls (HC), mild cognitive impairment, and Alzheimer's disease dementia (ADD) among groups of subjects.

PATIENTS AND METHODS

A total dataset of 14,926 subjects was obtained from the formal 46 NPTs based on the Seoul Neuropsychological Screening Battery (SNSB). The statistical values of the dataset included an age of 70.18 ± 7.13 with an education level of 8.18 ± 5.50 and a diagnosis group of three; HC, MCI, and ADD. The dataset was preprocessed and classified in two- and three-way machine-learning classification from scikit-learn (www.scikit-learn.org) to differentiate between HC versus MCI, HC versus ADD, HC versus Cognitive Impairment (CI) (MCI + ADD), and HC versus MCI versus ADD. We compared the performance of seven machine learning algorithms, including Naïve Bayes (NB), random forest (RF), decision tree (DT), k-nearest neighbors (KNN), support vector machine (SVM), AdaBoost, and linear discriminant analysis (LDA). The accuracy, sensitivity, specificity, positive predicted value (PPV), negative predictive value (NPV), area under the curve (AUC), confusion matrixes, and receiver operating characteristic (ROC) were obtained from each model based on the test dataset.

RESULTS

The trained models based on 29 best-selected NPT features were evaluated, the model with the RF algorithm yielded the best accuracy, sensitivity, specificity, PPV, NPV, and AUC in all four models: HC versus MCI was 98%, 98%, 97%, 98%, 97%, and 99%; HC versus ADD was 98%, 99%, 96%, 97%, 98%, and 99%; HC versus CI was 97%, 99%, 92%, 97%, 97%, and 99% and HC versus MCI versus ADD was 97%, 96%, 98%, 97%, 98%, and 99%, respectively, in predicting of cognitive impairment among subjects.

CONCLUSION

According to the results, the RF algorithm was the best classification model for both two- and three-way classification among the seven algorithms trained on an imbalanced NPTs SNSB dataset. The trained models proved useful for diagnosing MCI and ADD in patients with normal NPTs. These models can optimize cognitive evaluation, enhance diagnostic accuracy, and reduce missed diagnoses.

Designing an effective semantic fluency test for early MCI diagnosis with machine learning

Semantic fluency tests are one of the key tests used in batteries for the early detection of Mild Cognitive Impairment (MCI) as the impairment in speech and semantic memory are among the first symptoms, attracting the attention of a large number of studies. Several new semantic categories and variables capable of providing complementary information of clinical interest have been proposed to increase their effectiveness. However, this also extends the time required to complete all tests and get the overall diagnosis. Therefore, there is a need to reduce the number of tests in the batteries and thus the time spent on them while maintaining or increasing their effectiveness. This study used machine learning methods to determine the smallest and most efficient combination of semantic categories and variables to achieve this goal. We utilized a database containing 423 assessments from 141 subjects, with each subject having undergone three assessments spaced approximately one year apart. Subjects were categorized into three diagnostic groups: Healthy (if diagnosed as healthy in all three assessments), stable MCI (consistently diagnosed as MCI), and heterogeneous MCI (when exhibiting alternations between healthy and MCI diagnoses across assessments). We obtained that the most efficient combination to distinguish between these categories of semantic fluency tests included the animals and clothes semantic categories with the variables corrects, switching, clustering, and total clusters. This combination is ideal for scenarios that require a balance between time efficiency and diagnosis capability, such as population-based screenings.

Applications of Artificial Intelligence in the Neuropsychological Assessment of Dementia: A Systematic Review

In the context of advancing healthcare, the diagnosis and treatment of cognitive disorders, particularly Mild Cognitive Impairment (MCI) and Alzheimer's Disease (AD), pose significant challenges. This review explores Artificial Intelligence (AI) and Machine Learning (ML) in neuropsychological assessment for the early detection and personalized treatment of MCI and AD. The review includes 37 articles that demonstrate that AI could be an useful instrument for optimizing diagnostic procedures, predicting cognitive decline, and outperforming traditional tests. Three main categories of applications are identified: (1) combining neuropsychological assessment with clinical data, (2) optimizing existing test batteries using ML techniques, and (3) employing virtual reality and games to overcome the limitations of traditional tests. Despite advancements, the review highlights a gap in developing tools that simplify the clinician's workflow and underscores the need for explainable AI in healthcare decision making. Future studies should bridge the gap between technical performance measures and practical clinical utility to yield accurate results and facilitate clinicians' roles. The successful integration of AI/ML in predicting dementia onset could reduce global healthcare costs and benefit aging societies.

Toward Quantitative Neurology: Sensors to Assess Motor Deficits in Dementia

Background

Alzheimer's disease (AD) is the most common neurodegenerative disorder which primarily involves memory and cognitive functions. It is increasingly recognized that motor involvement is also a common and significant aspect of AD, contributing to functional decline and profoundly impacting quality of life. Motor impairment, either at early or later stages of cognitive disorders, can be considered as a proxy measure of cognitive impairment, and technological devices can provide objective measures for both diagnosis and prognosis purposes. However, compared to other neurodegenerative disorders, the use of technological tools in neurocognitive disorders, including AD, is still in its infancy.

Objective

This report aims to evaluate the role of technological devices in assessing motor involvement across the AD spectrum and in other dementing conditions, providing an overview of the existing devices that show promise in this area and exploring their clinical applications.

Methods

The evaluation involves a review of the existing literature in the PubMed, Web of Science, Scopus, and Cochrane databases on the effectiveness of these technologies. 21 studies were identified and categorized as: wearable inertial sensors/IMU, console/kinect, gait analysis, tapping device, tablet/mobile, and computer.

Results

We found several parameters, such as speed and stride length, that appear promising for detecting abnormal motor function in MCI or dementia. In addition, some studies have found correlations between these motor aspects and cognitive state.

Conclusions

Clinical application of technological tools to assess motor function in people with cognitive impairments of a neurodegenerative nature, such as AD, may improve early detection and stratification of patients.

Mycophenolate mofetil reduces the risk of relapse in anti-leucine-rich glioma-inactivated protein 1 encephalitis: a prospective observational cohort study

BACKGROUND

Mycophenolate mofetil (MMF) is frequently used in the treatment of neurological autoimmune disorders. However, its effect on the relapse risk in anti-leucine-rich glioma-inactivated protein 1 (anti-LGI1) encephalitis is not well studied.

METHODS

In this prospective observational cohort study, anti-LGI1 encephalitis patients were grouped according to MMF treatment status (MMF and non-MMF groups). The primary outcome was relapse after disease onset.

RESULTS

A total of 83 patients were included, with a median onset age of 60 years. Fifty-four patients were men (65.1%). The MMF group comprised 28 patients and the non-MMF group comprised 55. Median follow-up from symptom onset was 26 months. Relapse occurred in 43 patients (51.8%). Median modified Rankin scale (mRS) score at enrollment was significantly higher in the MMF group than the non-MMF group (3 vs. 2; p = 0.001). Median mRS score at last follow-up was comparable between groups (1 vs. zero; p = 0.184). Both MMF treatment (HR 0.463; 95% CI, 0.231-0.929; p = 0.030) and cognitive impairment at enrollment (HR 3.391; 95% CI, 1.041-11.044; p = 0.043) were independent predictors of relapse. Starting immunotherapy before development of cognitive impairment trended towards reducing relapse risk. Outcome at last follow-up was good (mRS score 0-2) in all patients except for one in the non-MMF group. Adverse events associated with MMF treatment were mild and transient.

CONCLUSION

Although the outcome of anti-LGI1 encephalitis patients is generally favorable, relapse is common, especially in those with cognitive impairment. MMF treatment is well-tolerated and can significantly reduce the risk of relapse.

The Relationship between Trait Impulsivity and Everyday Executive Functions among Patients with Type 2 Diabetes Mellitus: The Mediating Effect of Negative Emotions

Background

In recent years, the incidence of type 2 diabetes mellitus (T2DM) has dramatically increased, imposing a heavy financial burden on society and individuals. The most cost-effective way to control diabetes is diabetes self-management, which depends on patients' executive functions (EFs). However, the level of EFs among patients with T2DM varies greatly. In addition to diabetes-related factors contributing to a decline in EFs, trait impulsivity as a relatively stable personality trait may explicate individual differences in EFs. The objective of this study was to verify the mediating effect of negative emotions on the relationship between trait impulsivity and EFs among patients with T2DM in China.

Methods

A total of 305 patients with T2DM were enrolled consecutively from the endocrinology departments of three tertiary hospitals in China using convenience sampling. The participants completed the Sociodemographic Questionnaire, Mini-Mental State Examination (MMSE), Barratt Impulsiveness Scale-Brief (BIS-Brief), Depression Anxiety and Stress Scales with 21 items (DASS-21), and Behavior Rating Inventory of Executive Function-Adult (BRIEF-A) version. A structural equation modeling was used to verify the mediating effect of negative emotions on the relationship between trait impulsivity and EFs.

Results

A total of 32.46% of the participants experienced at least one aspect of daily EF decline. The mediating effect of trait impulsivity on the Behavioral Regulation Index (BRI) of EFs through negative emotions was significant, accounting for 29.57% of the total effect. The mediating effect of trait impulsivity on the Metacognitive Index (MI) of EFs through negative emotions was significant, accounting for 31.67% of the total effect.

Conclusions

Trait impulsivity can positively predict EF decline, which can be alleviated by improving the negative emotions of patients with T2DM. Future research exploring interventions to improve the EFs of patients with T2DM should therefore consider their trait impulsivity and negative emotions.

Association of APOE ε4/ε4 with fluid biomarkers in patients from the PUMCH dementia cohort

Background

Apolipoprotein-E (APOE) ε4 is a major genetic risk factor for Alzheimer’s disease (AD). Current studies, which were mainly based on the clinical diagnosis rather than biomarkers, come to inconsistent conclusions regarding the associations of APOE ε4 homozygotes (APOE ε4/ε4) and cerebrospinal fluid (CSF) biomarkers of AD. In addition, few studies have explored the associations of APOE ε4/ε4 with plasma biomarkers. Therefore, we aimed to investigate the associations of APOE ε4/ε4 with fluid biomarkers in dementia and biomarker-diagnosed AD.

Methods

A total of 297 patients were enrolled. They were classified into Alzheimer’s continuum, AD, and non-AD, according to CSF biomarkers and/or β amyloid PET results. AD was a subgroup of the AD continuum. Plasma Amyloid β (Aβ) 40, Aβ42, glial fibrillary acidic protein (GFAP), neurofilament light chain (NFL), and phosphorylated tau (P-tau)181 were quantified in 144 of the total population using an ultra-sensitive Simoa technology. We analyzed the associations of APOE ε4/ε4 on CSF and plasma biomarkers in dementia and biomarker diagnosed AD.

Results

Based on the biomarker diagnostic criteria, 169 participants were diagnosed with Alzheimer’s continuum and 128 individuals with non-AD, and among the former, 120 patients with AD. The APOE ε4/ε4 frequencies were 11.8% (20/169), 14.2% (17/120), and 0.8% (1/128) in Alzheimer’s continuum, AD and non-AD, respectively. Only CSF Aβ42 was shown to be decreased in APOE ε4/ε4 carriers than in non-carriers for patients with AD (p = 0.024). Furthermore, we did not find any associations of APOE ε4 with plasma biomarkers of AD and non-AD. Interestingly, we found that in non-AD patients, APOE ε4 carriers had lower CSF Aβ42 (p = 0.018) and higher T-tau/Aβ42 ratios (p &lt; 0.001) and P-tau181/Aβ42 ratios (p = 0.002) than non-carriers.

Conclusion

Our data confirmed that of the three groups (AD continuum, AD, and non-AD), those with AD had the highest frequency of APOE ɛ4/ɛ4 genotypes. The APOE ɛ4/ɛ4 was associated with CSF levels of Aβ42 but not tau for AD and non-AD, suggesting that APOE ɛ4/ɛ4 affected the Aβ metabolism of both. No associations between APOE ε4/ɛ4 and plasma biomarkers of AD and non-AD were found.

Learning Cognitive-Test-Based Interpretable Rules for Prediction and Early Diagnosis of Dementia Using Neural Networks

Background: Accurate, cheap, and easy to promote methods for dementia prediction and early diagnosis are urgently needed in low- and middle-income countries. Integrating various cognitive tests using machine learning provides promising solutions. However, most effective machine learning models are black-box models that are hard to understand for doctors and could hide potential biases and risks. Objective: To apply cognitive-test-based machine learning models in practical dementia prediction and diagnosis by ensuring both interpretability and accuracy. Methods: We design a framework adopting Rule-based Representation Learner (RRL) to build interpretable diagnostic rules based on the cognitive tests selected by doctors. According to the visualization and test results, doctors can easily select the final rules after analysis and trade-off. Our framework is verified on the Alzheimer’s Disease Neuroimaging Initiative (ADNI) dataset (n = 606) and Peking Union Medical College Hospital (PUMCH) dataset (n = 375). Results: The predictive or diagnostic rules learned by RRL offer a better trade-off between accuracy and model interpretability than other representative machine learning models. For mild cognitive impairment (MCI) conversion prediction, the cognitive-test-based rules achieve an average area under the curve (AUC) of 0.904 on ADNI. For dementia diagnosis on subjects with a normal Mini-Mental State Exam (MMSE) score, the learned rules achieve an AUC of 0.863 on PUMCH. The visualization analyses also verify the good interpretability of the learned rules. Conclusion: With the help of doctors and RRL, we can obtain predictive and diagnostic rules for dementia with high accuracy and good interpretability even if only cognitive tests are used.

Development of a Machine Learning Model to Discriminate Mild Cognitive Impairment Subjects from Normal Controls in Community Screening

Background: Mild cognitive impairment (MCI) is a transitional stage between normal aging and probable Alzheimer's disease. It is of great value to screen for MCI in the community. A novel machine learning (ML) model is composed of electroencephalography (EEG), eye tracking (ET), and neuropsychological assessments. This study has been proposed to identify MCI subjects from normal controls (NC). Methods: Two cohorts were used in this study. Cohort 1 as the training and validation group, includes184 MCI patients and 152 NC subjects. Cohort 2 as an independent test group, includes 44 MCI and 48 NC individuals. EEG, ET, Neuropsychological Tests Battery (NTB), and clinical variables with age, gender, educational level, MoCA-B, and ACE-R were selected for all subjects. Receiver operating characteristic (ROC) curves were adopted to evaluate the capabilities of this tool to classify MCI from NC. The clinical model, the EEG and ET model, and the neuropsychological model were compared. Results: We found that the classification accuracy of the proposed model achieved 84.5 ± 4.43% and 88.8 ± 3.59% in Cohort 1 and Cohort 2, respectively. The area under curve (AUC) of the proposed tool achieved 0.941 (0.893-0.982) in Cohort 1 and 0.966 (0.921-0.988) in Cohort 2, respectively. Conclusions: The proposed model incorporation of EEG, ET, and neuropsychological assessments yielded excellent classification performances, suggesting its potential for future application in cognitive decline prediction.

Association Between High Serum Tetrahydrofolate and Low Cognitive Functions in the United States: A Cross-Sectional Study

Background: The relationship between serum folate status and cognitive functions is still controversial. Objective: To evaluate the association between serum tetrahydrofolate and cognitive functions. Methods: A total of 3,132 participants (60–80 years old) from the 2011–2014 NHANES were included in this cross-sectional study. The primary outcome measure was cognitive function assessment, determined by the Consortium to Establish a Registry for Alzheimer’s Disease Word Learning Test (CERAD-WL), CERAD-Delayed Recall Test (CERAD-DR), Animal Fluency Test (AF), Digit Symbol Substitution Test (DSST), and global cognitive score. Generalized linear model (GLM), multivariate logistic regression models, weighted generalized additive models (GAM), and subgroup analyses were performed to evaluate the association between serum tetrahydrofolate and low cognitive functions. Results: In GLM, and the crude model, model 1, model 2 of multivariate logistic regression models, increased serum tetrahydrofolate was associated with reduced cognitive functions via AF, DSST, CERAD-WL, CERAD-DR, and global cognitive score (p <  0.05). In GAM, the inflection points were 1.1, 2.8, and 2.8 nmol/L tetrahydrofolate, determined by a two-piece wise linear regression model of AF, DSST, and global cognitive score, respectively. Also, in GAM, there were no non-linear relationship between serum tetrahydrofolate and low cognitive functions, as determined by CERAD-WL or CERAD-DR. The results of subgroup analyses found that serum tetrahydrofolate levels and reduced cognitive functions as determined by AF had significant interactions for age and body mass index. The association between high serum tetrahydrofolate level and reduced cognitive functions as determined using DSST, CERAD-WL, CERAD-DR, or global cognitive score had no interaction with the associations between cognition and gender, or age, or so on. Conclusion: High serum tetrahydrofolate level is associated with significantly reduced cognitive function.