A Multivariable Prediction Model for Mild Cognitive Impairment and Dementia: Algorithm Development and Validation DOI Creative Commons
Sarah Soyeon Oh, Bada Kang, Dahye Hong

et al.

JMIR Medical Informatics, Journal Year: 2024, Volume and Issue: 12, P. e59396 - e59396

Published: Nov. 22, 2024

Background Mild cognitive impairment (MCI) poses significant challenges in early diagnosis and timely intervention. Underdiagnosis, coupled with the economic social burden of dementia, necessitates more precise detection methods. Machine learning (ML) algorithms show promise managing complex data for MCI dementia prediction. Objective This study assessed predictive accuracy ML models identifying onset using Korean Longitudinal Study Aging (KLoSA) dataset. Methods used from KLoSA, a comprehensive biennial survey that tracks demographic, health, socioeconomic aspects middle-aged older adults 2018 to 2020. Among 6171 initial households, 4975 eligible adult participants aged 60 years or were selected after excluding individuals based on age missing data. The identification relied self-reported diagnoses, sociodemographic health-related variables serving as key covariates. dataset was categorized into training test sets predict by multiple models, including logistic regression, light gradient-boosting machine, XGBoost (extreme gradient boosting), CatBoost, random forest, boosting, AdaBoost, support vector classifier, k-nearest neighbors, evaluate performance. performance area under receiver operating characteristic curve (AUC). Class imbalances addressed via weights. Shapley additive explanation values determine contribution each feature prediction rate. Results participants, best model predicting median AUC 0.6729 (IQR 0.3883-0.8152), followed neighbors 0.5576 0.4555-0.6761) classifier 0.5067 0.3755-0.6389). For prediction, XGBoost, achieving 0.8185 0.8085-0.8285), closely machine 0.8069 0.7969-0.8169) AdaBoost 0.8007 0.7907-0.8107). highlighted pain everyday life, being widowed, living alone, exercising, partner strongest predictors MCI. most features other contributing factors, education at high school level, middle monthly engagement. Conclusions algorithms, especially exhibited potential KLoSA However, no has demonstrated robust dementia. Sociodemographic factors are crucial initiating conditions, emphasizing need multifaceted These findings underscore limitations community-dwelling adults.

Language: Английский

Artificial intelligence for diagnostic and prognostic neuroimaging in dementia: A systematic review DOI Creative Commons
Robin Borchert, Tiago Azevedo, AmanPreet Badhwar

et al.

Alzheimer s & Dementia, Journal Year: 2023, Volume and Issue: 19(12), P. 5885 - 5904

Published: Aug. 10, 2023

Abstract Introduction Artificial intelligence (AI) and neuroimaging offer new opportunities for diagnosis prognosis of dementia. Methods We systematically reviewed studies reporting AI in and/or cognitive neurodegenerative diseases. Results A total 255 were identified. Most relied on the Alzheimer's Disease Neuroimaging Initiative dataset. Algorithmic classifiers most commonly used method (48%) discriminative models performed best differentiating disease from controls. The accuracy algorithms varied with patient cohort, imaging modalities, stratifiers used. Few validation an independent cohort. Discussion literature has several methodological limitations including lack sufficient algorithm development descriptions standard definitions. make recommendations to improve model addressing key clinical questions, providing description methods validating findings datasets. Collaborative approaches between experts medicine will help achieve promising potential tools practice. Highlights There been a rapid expansion use machine learning (71%) (ADNI) dataset no other individual more than five times recent rise complex (e.g., neural networks) that better classification AD vs healthy controls address considerations, also field broadly standardize outcome measures, gaps literature, monitor sources bias

Language: Английский

Citations

43
Artificial Intelligence and Neuroscience: Transformative Synergies in Brain Research and Clinical Applications DOI Open Access

Răzvan Onciul,

Cătălina-Ioana Tătaru,

Adrian Dumitru

et al.

Journal of Clinical Medicine, Journal Year: 2025, Volume and Issue: 14(2), P. 550 - 550

Published: Jan. 16, 2025

The convergence of Artificial Intelligence (AI) and neuroscience is redefining our understanding the brain, unlocking new possibilities in research, diagnosis, therapy. This review explores how AI’s cutting-edge algorithms—ranging from deep learning to neuromorphic computing—are revolutionizing by enabling analysis complex neural datasets, neuroimaging electrophysiology genomic profiling. These advancements are transforming early detection neurological disorders, enhancing brain–computer interfaces, driving personalized medicine, paving way for more precise adaptive treatments. Beyond applications, itself has inspired AI innovations, with architectures brain-like processes shaping advances algorithms explainable models. bidirectional exchange fueled breakthroughs such as dynamic connectivity mapping, real-time decoding, closed-loop systems that adaptively respond states. However, challenges persist, including issues data integration, ethical considerations, “black-box” nature many systems, underscoring need transparent, equitable, interdisciplinary approaches. By synthesizing latest identifying future opportunities, this charts a path forward integration neuroscience. From harnessing multimodal cognitive augmentation, fusion these fields not just brain science, it reimagining human potential. partnership promises where mysteries unlocked, offering unprecedented healthcare, technology, beyond.

Language: Английский

Citations

5
Artificial intelligence for dementia research methods optimization DOI Creative Commons
Magda Bucholc, Charlotte James, Ahmad Al Khleifat

et al.

Alzheimer s & Dementia, Journal Year: 2023, Volume and Issue: 19(12), P. 5934 - 5951

Published: Aug. 28, 2023

Abstract Artificial intelligence (AI) and machine learning (ML) approaches are increasingly being used in dementia research. However, several methodological challenges exist that may limit the insights we can obtain from high‐dimensional data our ability to translate these findings into improved patient outcomes. To improve reproducibility replicability, researchers should make their well‐documented code modeling pipelines openly available. Data also be shared where appropriate. enhance acceptability of models AI‐enabled systems users, prioritize interpretable methods provide how decisions generated. Models developed using multiple, diverse datasets robustness, generalizability, reduce potentially harmful bias. clarity reproducibility, adhere reporting guidelines co‐produced with multiple stakeholders. If overcome, AI ML hold enormous promise for changing landscape research care. Highlights Machine diagnosis, prevention, management dementia. Inadequate procedures affects reproduction/replication results. built on unrepresentative do not generalize new datasets. Obligatory metrics certain model structures use cases have been defined. Interpretability trust predictions barriers clinical translation.

Language: Английский

Citations

15
Artificial intelligence for neurodegenerative experimental models DOI Creative Commons
Sarah J. Marzi, Brian M. Schilder, Alexi Nott

et al.

Alzheimer s & Dementia, Journal Year: 2023, Volume and Issue: 19(12), P. 5970 - 5987

Published: Sept. 28, 2023

Experimental models are essential tools in neurodegenerative disease research. However, the translation of insights and drugs discovered model systems has proven immensely challenging, marred by high failure rates human clinical trials.

Language: Английский

Citations

11
Artificial Intelligence for Predicting Progression and Personalizing Healthcare to Alzheimer's Disease Patients DOI

I F T I Khar Ali,

Vijaya Kittu Manda

Advances in healthcare information systems and administration book series, Journal Year: 2025, Volume and Issue: unknown, P. 155 - 190

Published: Jan. 10, 2025

This chapter explains the use of Deep Learning Models from Artificial Intelligence (AI) that take Structural and Functional Magnetic Resonance Imaging (S/FMRI) data to classify Alzheimer's disease (AD) progression stages. Early accurate diagnosis AD is necessary for timely intervention, treatment planning, providing personalized healthcare. Current limitations in diagnostic methods necessitate using AI such as Convolutional Neural Networks (CNN) Recurrent (RNN) extract features MRI develop models predicting Mild Cognitive Impairment (MCI), AD, Dementia. Initial results a case study applied methodology demonstrated improved classification accuracy over traditional accurately classifying stages developing patient care. With more refinement technologies progress, these computational approaches can drastically positively change Healthcare professionals benefit this by understanding how be implemented deal with neurodegenerative diseases.

Language: Английский

Citations

0
Development and validation of a convenient dementia risk prediction tool for diabetic population: A large and longitudinal machine learning cohort study DOI
Pei Yang,

Xuan Xiao,

Yihui Li

et al.

Journal of Affective Disorders, Journal Year: 2025, Volume and Issue: 380, P. 298 - 307

Published: March 29, 2025

Language: Английский

Citations

0
Aptamer-functionalized graphene quantum dots combined with artificial intelligence detect bacteria for urinary tract infections DOI Creative Commons
Kun Li, Shiqiang Fang, Tangwei Wu

et al.

Frontiers in Cellular and Infection Microbiology, Journal Year: 2025, Volume and Issue: 15

Published: April 16, 2025

Urinary tract infection is one of the most prevalent forms bacterial infection, and prompt efficient identification pathogenic bacteria plays a pivotal role in management urinary infections. In this study, we propose novel approach utilizing aptamer-functionalized graphene quantum dots integrated with an artificial intelligence detection system (AG-AI system) for rapid highly sensitive Escherichia coli (E. coli). Firstly, were modified aptamer that can specifically recognize bind to E. coli. Therefore, fluorescence intensity was positively correlated concentration Finally, images processed by obtain result concentration. The AG-AI system, wide linearity (103-109 CFU/mL) low limit (3.38×102 CFU/mL), effectively differentiate between other bacteria. And good agreement MALDI-TOF MS. accurate effective way detect

Language: Английский

Citations

0
Current Applications of Artificial Intelligence in Psychiatry DOI
Nicholas A Kerna, Adina Boulos, Melany Abreu

et al.

Scientia. Technology, science and society., Journal Year: 2025, Volume and Issue: 2(4), P. 125 - 143

Published: April 1, 2025

The integration of artificial intelligence (AI) into psychiatric practice has accelerated rapidly, driven by advances in computational methods and the availability diverse data sources. present paper examines contemporary AI applications across diagnostic support, predictive analytics, therapeutic interventions, digital phenotyping, telepsychiatry integration, ethical, legal, social considerations. Foundations machine learning, deep natural language processing are delineated alongside relevant modalities, including structured clinical records, unstructured notes, multimodal signals. roles symptom detection, neuroimaging pattern recognition, biomarker discovery, differential diagnosis evaluated. Predictive models for suicide risk, relapse, treatment response reviewed, with attention to personalization algorithms. Therapeutic tools, such as conversational agents, virtual reality, gamified mobile applications, discussed. Passive monitoring techniques, workflows, clinician dashboards described. Ethical challenges, privacy, algorithmic bias, regulatory frameworks, considered. Implementation barriers adoption factors analyzed. Emerging trends, federated fusion, explainable AI, low-resource settings, explored. Implications patient outcomes, health systems, policy synthesized, concluding recommendations future research practice.

Language: Английский

Citations

0
Prediction of 28-day all-cause mortality in heart failure patients with Clostridioides difficile infection using machine learning models: evidence from the MIMIC-IV database DOI

Caiping Shi,

Qiong Jie,

Hongsong Zhang

et al.

Cardiology, Journal Year: 2024, Volume and Issue: unknown, P. 1 - 1

Published: Aug. 17, 2024

Introduction: Heart failure (HF) may induce bowel hypoperfusion, leading to hypoxia of the villa wall and occurrence Clostridioides difficile infection (CDI). However, risk factors for development CDI in HF patients have yet be fully illustrated, especially because a lack evidence from real-world data. Methods: Clinical data survival situations with admitted ICU were extracted Medical Information Mart Intensive Care (MIMIC)-IV database. For developing model that can predict 28-day all-cause mortality CDI, Recursive Feature Elimination Cross-Validation (RFE-CV) method was used feature selection. And nine machine learning (ML) algorithms, including logistic regression (LR), decision tree, Bayesian, adaptive boosting, random forest (RF), gradient boosting XGBoost, light machine, categorical applied construction. After training hyperparameter optimization models through grid search 5-fold cross-validation, performance evaluated by area under curve (AUC), accuracy, sensitivity, specificity, precision, negative predictive value, F1 score. Furthermore, SHapley Additive exPlanations (SHAP) interpret optimal model. Results: A total 526 included study, whom 99 cases (18.8%) experienced death within 28 days. Eighteen 57 variables selected construction algorithm Among ML considered, RF emerged as achieving F1-score, AUC values 0.821, 0.596, 0.864, respectively. The net benefit surpassed other at 16%–22% threshold probabilities based on analysis. According importance features model, red blood cell distribution width, urea nitrogen, Simplified Acute Physiology Score II, Sequential Organ Failure Assessment, white count highlighted five most influential variables. Conclusions: We developed associated ICU, which are more effective than conventional LR has best among all employed. It useful help clinicians identify high-risk CDI.

Language: Английский

Citations

2
Artificial intelligence–based rapid brain volumetry substantially improves differential diagnosis in dementia DOI Creative Commons
Jan Rudolph, Johannes Rueckel,

Jörg Döpfert

et al.

Alzheimer s & Dementia Diagnosis Assessment & Disease Monitoring, Journal Year: 2024, Volume and Issue: 16(4)

Published: Oct. 1, 2024

Language: Английский

Citations

2