AI Article Synopsis
- This study introduces the Multi-Scale Self-Attention Network (MUSAN) for better classification of Alzheimer’s disease (AD) by distinguishing between cognitively normal individuals, stable mild cognitive impairment (sMCI), and progressive mild cognitive impairment (pMCI).
- MUSAN processes structural MRI data to automatically extract features from various brain regions, using an occlusion sensitivity algorithm to identify which regions are most affected by AD, such as the hippocampus and amygdala.
- The method demonstrated high accuracy in classifying AD and sMCI, achieving significant sensitivity and specificity, and improves interpretability of deep learning models, thereby enhancing reliability in Alzheimer’s research.
Article Abstract
Background: Structural magnetic resonance imaging (sMRI) is vital for early Alzheimer's disease (AD) diagnosis, though confirming specific biomarkers remains challenging. Our proposed Multi-Scale Self-Attention Network (MUSAN) enhances classification of cognitively normal (CN) and AD individuals, distinguishing stable (sMCI) from progressive mild cognitive impairment (pMCI).
Objective: This study leverages AD structural atrophy properties to achieve precise AD classification, combining different scales of brain region features. The ultimate goal is an interpretable algorithm for this method.
Methods: The MUSAN takes whole-brain sMRI as input, enabling automatic extraction of brain region features and modeling of correlations between different scales of brain regions, and achieves personalized disease interpretation of brain regions. Furthermore, we also employed an occlusion sensitivity algorithm to localize and visualize brain regions sensitive to disease.
Results: Our method is applied to ADNI-1, ADNI-2, and ADNI-3, and achieves high performance on the classification of CN from AD with accuracy (0.93), specificity (0.82), sensitivity (0.96), and area under curve (AUC) (0.95), as well as notable performance on the distinguish of sMCI from pMCI with accuracy (0.85), specificity (0.84), sensitivity (0.74), and AUC (0.86). Our sensitivity masking algorithm identified key regions in distinguishing CN from AD: hippocampus, amygdala, and vermis. Moreover, cingulum, pallidum, and inferior frontal gyrus are crucial for sMCI and pMCI discrimination. These discoveries align with existing literature, confirming the dependability of our model in AD research.
Conclusion: Our method provides an effective AD diagnostic and conversion prediction method. The occlusion sensitivity algorithm enhances deep learning interpretability, bolstering AD research reliability.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.3233/JAD-230705 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Bayesian modeling framework for optimizing pre-hospital stroke triage decisions.
J Appl Stat
May 2024
Department of Biostatistics, College of Public Health, University of Iowa, Iowa City, IA, USA.
Ischemic stroke is responsible for significant morbidity and mortality in the United States and worldwide. Stroke treatment optimization requires emergency medical personnel to make rapid triage decisions concerning destination hospitals that may differ in their ability to provide highly time-sensitive pharmaceutical and surgical interventions. These decisions are particularly crucial in rural areas, where transport decisions can have a large impact on treatment times - often involving a trade-off between delay in pharmaceutical therapy or a delay in endovascular thrombectomy.
View Article and Find Full Text PDFA Method for Imaging the Ischemic Penumbra with MRI using IVIM.
AJNR Am J Neuroradiol
January 2025
From the Department of Radiology, Medical Physics (MML, TJC), Department of Interventional Radiology (NS, GAC), Department of Surgery and Large Animal Studies (MAN), and the Department of Statistics (MG), University of Chicago, Chicago, IL, USA; Department of Anesthesiology (SPR), University of Illinois, Chicago, IL, USA; Department of Radiology (MSS), University of Massachusetts Chan Medical School, Worcester, MA, USA; Department of Radiology, Biomedical Engineering and Imaging Institute (Current affiliation MML), Icahn School of Medicine at Mount Sinai, New York, NY, USA; Mount Carmel Health Systems (Current affiliation GAC), Columbus, OH, USA.
Background And Purpose: In acute ischemic stroke, the amount of "local" CBF distal to the occlusion, i.e. all blood flow within a region whether supplied antegrade or delayed and dispersed through the collateral network, may contain valuable information regarding infarct growth rate and treatment response.
View Article and Find Full Text PDFUtilizing Quantitative Analysis of CSF Volume from Clinical T1-Weighted MRI to Predict Thrombectomy Outcomes.
J Neuroimaging
January 2025
Department of Radiology, Wake Forest University School of Medicine, Winston-Salem, North Carolina, USA.
Background And Purpose: Endovascular thrombectomy (EVT) is the standard for acute ischemic stroke from large vessel occlusion, but post-EVT functional independence varies. Brain atrophy, linked to higher cerebrospinal fluid volume (CSFV), may affect outcomes. Baseline CSFV could predict EVT benefit by assessing brain health.
View Article and Find Full Text PDFReduced contrast agent volume using a heart-rate dependent and free-breathing scanning protocol in coronary computed tomography angiography (CTA) for patients with chronic obstructive pulmonary disease (COPD).
BMC Cardiovasc Disord
January 2025
Department of Radiology, Central Hospital Affiliated to Shandong First Medical University, Jinan City, Shandong Province, China.
Background: The personalized, free-breathing, heart rate-dependent computed tomography angiography (CTA) protocol can significantly reduce the utilization of contrast medium (CM). This proves especially beneficial for patients with chronic obstructive pulmonary disease (COPD) undergoing coronary artery CTA examinations.
Objective: The aim of this study was to evaluate the feasibility of a personalized CT scanning protocol that was tailored to patients' heart rate and free-breathing for coronary CTA of patients with COPD.
Comparative Evaluation of Two Novel Natural Desensitizing Agents on Dentinal Tubule Occlusion: A Scanning Electron Microscopy Study.
J Contemp Dent Pract
September 2024
Department of Conservative Dentistry and Endodontics, PMNM Dental College and Hospital, Bagalkot, Karnataka, India.
Aim: The present study is aimed to compare the effectiveness of dentinal tubule occlusion of two natural-based desensitizing toothpaste (Bentodent Desensitizer and Fang Farm Mint) and Novamin-based toothpaste (Shy-NM) under scanning electron microscope (SEM).
Materials And Methods: A total of 60 root dentin discs (3 mm-thick) were obtained from recently extracted permanent premolar teeth and were randomly divided into four groups based on the desensitizing toothpaste used, each group with 15 samples: Group I: Control, group II: Shy-NM, group III: Bentodent desensitizer, group IV: Fang Farm Mint. Samples were brushed for 2 min twice daily with a soft toothbrush with respective pea size amount of toothpaste for 14 days manually.
Want AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!