Detection of Alzheimer's Disease in Neuroimages using Vision Transformers: A Systematic Review and Meta-Analysis (Preprint) DOI Creative Commons
Vivens Mubonanyikuzo,

Hongjie Yan,

Temitope Emmanuel Komolafe

и другие.

Journal of Medical Internet Research, Год журнала: 2024, Номер 27, С. e62647 - e62647

Опубликована: Ноя. 30, 2024

Background Alzheimer disease (AD) is a progressive condition characterized by cognitive decline and memory loss. Vision transformers (ViTs) are emerging as promising deep learning models in medical imaging, with potential applications the detection diagnosis of AD. Objective This review systematically examines recent studies on application ViTs detecting AD, evaluating diagnostic accuracy impact network architecture model performance. Methods We conducted systematic search across major databases, including China National Knowledge Infrastructure, CENTRAL (Cochrane Central Register Controlled Trials), ScienceDirect, PubMed, Web Science, Scopus, covering publications from January 1, 2020, to March 2024. A manual was also performed include relevant gray literature. The included papers used ViT for AD versus healthy controls based neuroimaging data, magnetic resonance imaging positron emission tomography. Pooled estimates, sensitivity, specificity, likelihood ratios, odds were derived using random-effects models. Subgroup analyses comparing performance different architectures performed. Results meta-analysis, encompassing 11 95% CIs P values, demonstrated pooled accuracy: sensitivity 0.925 (95% CI 0.892-0.959; P<.01), specificity 0.957 0.932-0.981; positive ratio 21.84 12.26-38.91; negative 0.08 0.05-0.14; P<.01). area under curve notably high at 0.924. findings highlight effective tools early accurate diagnosis, offering insights future neuroimaging-based approaches. Conclusions provides valuable evidence utility distinguishing patients controls, thereby contributing advancements methodologies. Trial Registration PROSPERO CRD42024584347; https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=584347

Язык: Английский

Review on Pharmaceutical Industries Production of Medicines for Lung Cancer Diseases Prevention and Side Effects DataSets View and Analysis with Orange Software Visualization Techniques DOI Open Access

V. Srinivasan,

S. Soumya

International Journal of Management Technology and Social Sciences, Год журнала: 2025, Номер unknown, С. 17 - 44

Опубликована: Янв. 30, 2025

Purpose: The purpose of this research is to explore how Orange, a powerful information extraction and predictive modeling software, can be applied in the pharmaceutical industry assess visualize effectiveness cancer prevention medicines. By focusing on companies like Genentech Inc. (USA), AstraZeneca Pharmaceutical PLC (UK), Boehringer Ingelheim (Germany), Chugai Co. Ltd. (Japan), study seeks evaluate which cancer-preventing drugs from these provide best efficacy while minimizing side effects for patients. goal assist healthcare professionals (doctors pharmacists) making informed decisions about most suitable medications prevention, ensuring patient safety optimal treatment outcomes. Design/Methodology/Approach: This utilizes Orange software’s machine learning data visualization tools, specifically scatterplot graphs, analyze complex datasets related drugs. using scatterplots concurrently examine multiple parameters, such as Company Name, Drug Class, Medicine Prevention Cancer Diseases, Side Effects Percentage aims identify patterns correlations that help drug safety. approach involves analyzing relationship between characteristics effects, providing actionable insights into different treatments might interact with health conditions. Findings/Results: findings suggest Orange’s visualizations valuable various medicines across companies. enabling simultaneous analysis software helps are effective preventing effects. provides clearer understanding characteristics, outcomes, supporting data-driven decision-making development practices. Originality/Value: originality lies application mining capabilities relationships within datasets. use efficacy, an innovative offers richer, more nuanced contributes optimizing choice strategies, ultimately improving therapeutic Paper Type: analytical paper applies techniques focuses tools extract interpret data, professionals.

Язык: Английский

Процитировано

0

Digital Biomarkers for Parkinson's Disease: A Bibliometric Analysis and a Scoping Review of Deep Learning for Freezing of Gait (Preprint) DOI Creative Commons
Wenhao Qi,

S. Shen,

Chaoqun Dong

и другие.

Journal of Medical Internet Research, Год журнала: 2025, Номер 27, С. e71560 - e71560

Опубликована: Март 19, 2025

With the rapid development of digital biomarkers in Parkinson disease (PD) research, it has become increasingly important to explore current research trends and key areas focus. This study aimed comprehensively evaluate status, hot spots, future global PD biomarker provide a systematic review deep learning models for freezing gait (FOG) biomarkers. used bibliometric analysis based on Web Science Core Collection database conduct comprehensive multidimensional landscape After identifying also followed PRISMA-ScR (Preferred Reporting Items Systematic Reviews Meta-Analyses Extension Scoping Reviews) guidelines scoping FOG from 5 databases: Science, PubMed, IEEE Xplore, Embase, Google Scholar. A total 750 studies were included analysis, 40 review. The revealed growing number related publications, with 3700 researchers contributing. Neurology had highest average annual participation rate (12.46/19, 66%). United States contributed most (192/1171, 16.4%), 210 participating institutions, which was among all countries. In FOG, accuracy 0.92, sensitivity 0.88, specificity 0.90, area under curve 0.91. addition, 31 (78%) indicated that best primarily convolutional neural networks or networks-based architectures. Research is currently at stable stage development, widespread interest countries, researchers. However, challenges remain, including insufficient interdisciplinary interinstitutional collaboration, as well lack corporate funding projects. Current focus motor-related studies, particularly monitoring. still external validation standardized performance reporting. Future will likely progress toward deeper applications artificial intelligence, enhanced different data types, exploration broader range symptoms. Open Foundation (OSF Registries) OSF.IO/RG8Y3; https://doi.org/10.17605/OSF.IO/RG8Y3.

Язык: Английский

Процитировано

0

Digital Biomarkers for Parkinson Disease: Bibliometric Analysis and a Scoping Review of Deep Learning for Freezing of Gait (Preprint) DOI
Wenhao Qi,

S. Shen,

Chaoqun Dong

и другие.

Опубликована: Янв. 21, 2025

BACKGROUND With the rapid development of digital biomarkers in Parkinson disease (PD) research, it has become increasingly important to explore current research trends and key areas focus. OBJECTIVE This study aimed comprehensively evaluate status, hot spots, future global PD biomarker provide a systematic review deep learning models for freezing gait (FOG) biomarkers. METHODS used bibliometric analysis based on Web Science Core Collection database conduct comprehensive multidimensional landscape After identifying also followed PRISMA-ScR (Preferred Reporting Items Systematic Reviews Meta-Analyses Extension Scoping Reviews) guidelines scoping FOG from 5 databases: Science, PubMed, IEEE Xplore, Embase, Google Scholar. RESULTS A total 750 studies were included analysis, 40 review. The revealed growing number related publications, with 3700 researchers contributing. Neurology had highest average annual participation rate (12.46/19, 66%). United States contributed most (192/1171, 16.4%), 210 participating institutions, which was among all countries. In FOG, accuracy 0.92, sensitivity 0.88, specificity 0.90, area under curve 0.91. addition, 31 (78%) indicated that best primarily convolutional neural networks or networks–based architectures. CONCLUSIONS Research is currently at stable stage development, widespread interest countries, researchers. However, challenges remain, including insufficient interdisciplinary interinstitutional collaboration, as well lack corporate funding projects. Current focus motor-related studies, particularly monitoring. still external validation standardized performance reporting. Future will likely progress toward deeper applications artificial intelligence, enhanced different data types, exploration broader range symptoms. CLINICALTRIAL Open Foundation (OSF Registries) OSF.IO/RG8Y3; https://doi.org/10.17605/OSF.IO/RG8Y3

Язык: Английский

Процитировано

0

Mapping knowledge landscapes and emerging trends of artificial intelligence in the early screening of cognitive impairment diseases DOI

Xin Liu,

Sixie Li,

Shuying Shen

и другие.

Опубликована: Март 28, 2025

Язык: Английский

Процитировано

0

Detection of Alzheimer's Disease in Neuroimages using Vision Transformers: A Systematic Review and Meta-Analysis (Preprint) DOI Creative Commons
Vivens Mubonanyikuzo,

Hongjie Yan,

Temitope Emmanuel Komolafe

и другие.

Journal of Medical Internet Research, Год журнала: 2024, Номер 27, С. e62647 - e62647

Опубликована: Ноя. 30, 2024

Background Alzheimer disease (AD) is a progressive condition characterized by cognitive decline and memory loss. Vision transformers (ViTs) are emerging as promising deep learning models in medical imaging, with potential applications the detection diagnosis of AD. Objective This review systematically examines recent studies on application ViTs detecting AD, evaluating diagnostic accuracy impact network architecture model performance. Methods We conducted systematic search across major databases, including China National Knowledge Infrastructure, CENTRAL (Cochrane Central Register Controlled Trials), ScienceDirect, PubMed, Web Science, Scopus, covering publications from January 1, 2020, to March 2024. A manual was also performed include relevant gray literature. The included papers used ViT for AD versus healthy controls based neuroimaging data, magnetic resonance imaging positron emission tomography. Pooled estimates, sensitivity, specificity, likelihood ratios, odds were derived using random-effects models. Subgroup analyses comparing performance different architectures performed. Results meta-analysis, encompassing 11 95% CIs P values, demonstrated pooled accuracy: sensitivity 0.925 (95% CI 0.892-0.959; P<.01), specificity 0.957 0.932-0.981; positive ratio 21.84 12.26-38.91; negative 0.08 0.05-0.14; P<.01). area under curve notably high at 0.924. findings highlight effective tools early accurate diagnosis, offering insights future neuroimaging-based approaches. Conclusions provides valuable evidence utility distinguishing patients controls, thereby contributing advancements methodologies. Trial Registration PROSPERO CRD42024584347; https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=584347

Язык: Английский

Процитировано

0