Research on classification method of multimodal magnetic resonance images of Alzheimer's disease based on generalized convolutional neural networks / 生物医学工程学杂志

Alzheimer's disease (AD) is a progressive and irreversible neurodegenerative disease. Neuroimaging based on magnetic resonance imaging (MRI) is one of the most intuitive and reliable methods to perform AD screening and diagnosis. Clinical head MRI detection generates multimodal image data, and to solve the problem of multimodal MRI processing and information fusion, this paper proposes a structural and functional MRI feature extraction and fusion method based on generalized convolutional neural networks (gCNN). The method includes a three-dimensional residual U-shaped network based on hybrid attention mechanism (3D HA-ResUNet) for feature representation and classification for structural MRI, and a U-shaped graph convolutional neural network (U-GCN) for node feature representation and classification of brain functional networks for functional MRI. Based on the fusion of the two types of image features, the optimal feature subset is selected based on discrete binary particle swarm optimization, and the prediction results are output by a machine learning classifier. The validation results of multimodal dataset from the AD Neuroimaging Initiative (ADNI) open-source database show that the proposed models have superior performance in their respective data domains. The gCNN framework combines the advantages of these two models and further improves the performance of the methods using single-modal MRI, improving the classification accuracy and sensitivity by 5.56% and 11.11%, respectively. In conclusion, the gCNN-based multimodal MRI classification method proposed in this paper can provide a technical basis for the auxiliary diagnosis of Alzheimer's disease.

Alzheimer's disease classification based on nonlinear high-order features and hypergraph convolutional neural network / 生物医学工程学杂志

Alzheimer's disease (AD) is an irreversible neurodegenerative disorder that damages patients' memory and cognitive abilities. Therefore, the diagnosis of AD holds significant importance. The interactions between regions of interest (ROIs) in the brain often involve multiple areas collaborating in a nonlinear manner. Leveraging these nonlinear higher-order interaction features to their fullest potential contributes to enhancing the accuracy of AD diagnosis. To address this, a framework combining nonlinear higher-order feature extraction and three-dimensional (3D) hypergraph neural networks is proposed for computer-assisted diagnosis of AD. First, a support vector machine regression model based on the radial basis function kernel was trained on ROI data to obtain a base estimator. Then, a recursive feature elimination algorithm based on the base estimator was applied to extract nonlinear higher-order features from functional magnetic resonance imaging (fMRI) data. These features were subsequently constructed into a hypergraph, leveraging the complex interactions captured in the data. Finally, a four-dimensional (4D) spatiotemporal hypergraph convolutional neural network model was constructed based on the fMRI data for classification. Experimental results on the Alzheimer's Disease Neuroimaging Initiative (ADNI) database demonstrated that the proposed framework outperformed the Hyper Graph Convolutional Network (HyperGCN) framework by 8% and traditional two-dimensional (2D) linear feature extraction methods by 12% in the AD/normal control (NC) classification task. In conclusion, this framework demonstrates an improvement in AD classification compared to mainstream deep learning methods, providing valuable evidence for computer-assisted diagnosis of AD.

Subjective sleep parameters in prodromal Alzheimer's disease: a case-control study

Objective: People with Alzheimer's disease (AD) dementia have impaired sleep. However, the characteristics of sleep in the early stages of AD are not well known, and studies with the aid of biomarkers are lacking. We assessed the subjective sleep characteristics of non-demented older adults and compared their amyloid profiles. Methods: We enrolled 30 participants aged ≥ 60 years, with no dementia or major clinical and psychiatric diseases. They underwent [11C]PiB-PET-CT, neuropsychological evaluations, and completed two standardized sleep assessments (Pittsburgh Sleep Quality Inventory and Epworth Sleep Scale). Results: Comparative analysis of subjective sleep parameters across the two groups showed longer times in bed (p = 0.024) and reduced sleep efficiency (p = 0.05) in individuals with positive amyloid. No differences in other subjective sleep parameters were observed. We also found that people with multiple-domain mild cognitive impairment (MCI) had shorter self-reported total sleep times (p = 0.034) and worse overall sleep quality (p = 0.027) compared to those with single-domain MCI. Conclusions: Older adults testing positive for amyloid had a longer time in bed and lower sleep efficiency, regardless of cognitive status. In parallel, individuals with multiple-domain MCI reported shorter sleep duration and lower overall sleep quality.

Research on the application of convolution neural network in the diagnosis of Alzheimer's disease / 生物医学工程学杂志

With the wide application of deep learning technology in disease diagnosis, especially the outstanding performance of convolutional neural network (CNN) in computer vision and image processing, more and more studies have proposed to use this algorithm to achieve the classification of Alzheimer's disease (AD), mild cognitive impairment (MCI) and normal cognition (CN). This article systematically reviews the application progress of several classic convolutional neural network models in brain image analysis and diagnosis at different stages of Alzheimer's disease, and discusses the existing problems and gives the possible development directions in order to provide some references.

Biomarcadores de neuroimagen para el diagnóstico temprano de la enfermedad de Alzheimer: un abordaje desde las redes neuronales / Neuroimaging biomarkers for early diagnosis of Alzheimers Disease: an approach from neural networks

Resumen: Las crecientes cifras mundiales de prevalencia e incidencia de la Enfermedad de Alzheimer exigen el desarrollo de métodos diagnósticos cada vez más precoces. La enfermedad suele diagnosticarse cuando la patología ya está avanzada y poco se puede hacer terapéuticamente, conllevando una gran pérdida de años de vida y altos costos sociales y familiares. Considerando esta patología como una disrupción a gran escala de las redes neuronales del cerebro humano, distintos estudios han propuesto biomarcadores basados en resonancia magnética y tomografía por emisión de positrones. El presente artículo revisa sistemáticamente dichos estudios considerando un abordaje desde la ciencia de redes neuronales.

Abstract: The increased global prevalence and incidence of Alzheimer's Disease demands the development of early diagnostic methods. This disease is usually diagnosed when the pathology is already advanced and therapeutically, there's not much to do, leading to a great loss of years of life, high socials and family costs. Considering this pathology as a large-scale disruption of neural networks of the human brain, different studies have proposed biomarkers based on functional magnetic resonance imaging and positron emission tomography. This article systematically reviews these studies considering an approach of neural networks science.

Topography of 11C-Pittsburgh compound B uptake in Alzheimer's disease: a voxel-based investigation of cortical and white matter regions

Objective: To compare results of positron emission tomography (PET) with carbon-11-labeled Pittsburgh compound B (11C-PIB) obtained with cerebellar or global brain uptake for voxel intensity normalization, describe the cortical sites with highest tracer uptake in subjects with mild Alzheimer's disease (AD), and explore possible group differences in 11C-PIB binding to white matter. Methods: 11C-PIB PET scans were acquired from subjects with AD (n=17) and healthy elderly controls (n=19). Voxel-based analysis was performed with statistical parametric mapping (SPM). Results: Cerebellar normalization showed higher 11C-PIB uptake in the AD group relative to controls throughout the cerebral cortex, involving the lateral temporal, orbitofrontal, and superior parietal cortices. With global uptake normalization, greatest cortical binding was detected in the orbitofrontal cortex; decreased 11C-PIB uptake in white matter was found in the posterior hippocampal region, corpus callosum, pons, and internal capsule. Conclusion: The present case-control voxelwise 11C-PIB PET comparison highlighted the regional distribution of amyloid deposition in the cerebral cortex of mildly demented AD patients. Tracer uptake was highest in the orbitofrontal cortex. Decreased 11C-PIB uptake in white-matter regions in this patient population may be a marker of white-matter damage in AD.

Volumetría hipocampal: análisis comparativo de los métodos de evaluación en enfermedad de Alzheimer / Hippocampal Volumetry: Comparative Analysis of Evaluation Methods in Alzheimer's Disease

Objetivo La atrofia hipocampal es uno de los biomarcadores radiológicos más sensibles de la enfermedad de Alzheimer (EA) y existen diferentes métodos para evaluarla: análisis subjetivo visual (ASV), análisis objetivo manual (AOM) y análisis objetivo automático (AOA). Nos proponemos comparar esos métodos, y evaluar si el AOA presenta una confiabilidad cercana al AOM (método de referencia) y superior al ASV. Materiales y Métodos Se seleccionaron retrospectivamente imágenes de resonancia magnética (RM) fast spoiled gradient-echo (FSPGR) de 28 sujetos (14 con deterioro cognitivo leve, 7 con EAy 7 controles). El ASV fue realizado por 10 radiólogos, clasificando la atrofia hipocampal en: nula, leve, moderada o severa. El AOM se basó en la segmentación manual de los hipocampos por dos operadores. El AOA fue realizada por medio del software FreeSurfer 5.3. Se calcularon coeficientes de correlación rho de Spearman para las variables discretas y coeficientes de correlación intraclase para las variables continuas. Resultados Los coeficientes de correlación entre los dos operadores que realizaron el AOM fueron de 0,88 (p < 0,0001) para los hipocampos izquierdos y de 0,86 (p < 0,0001) para los hipocampos derechos. El coeficiente de correlación entre todos los ASV (promediados) y AOM fue de-0,81 (IC 95%-0,96- -0,66). Los coeficientes de correlación entre el AOA y el AOM fue de 0,54 (p < 0,0001) para los hipocampos izquierdos y de 0,61 (p < 0,0001) para los hipocampos derechos. Conclusión Si bien el AOA tiene moderada correlación con el método de referencia, no es superior al ASV promedio y se deberían tomar recaudos antes de ser implementado en la práctica asistencial.

Objective Hippocampal atrophy is one of the most sensible radiological biomarkers of Alzheimer's disease. There are different methods to evaluate atrophy: visual subjective analysis (VSA), manual objective analysis (MOA) and automatic objective analysis (AOA). We will compare these methods and evaluate if AOA has a confidence similar to MOA (gold standard), and better than VSA. Materials and Methods We retrospectively selected 3D FSPGR MRI from 28 subjects of whom 14 had mild cognitive impairment, 7 Alzheimer's disease and 7 controls. VSA was performed by 10 radiologists who classified hippocampal atrophy in none, mild, moderate and severe. ForMOA, two operatorsmanually segmented both hippocampus from all subjects. AOA was performed by FreeSurfer 5.3. Spearman's rho correlation coefficient was calculated for discrete variables and intraclass correlation coefficient was calculated for continuous variables. Results Correlation coefficients between the operators that performed MOA was 0.88 (p < 0.0001) for left hippocampi and 0.86 (p < 0,0001) for right hippocampi. Correlation coefficients between mean VSA and MOA was-0,81 (95% CI-0,96- -0,66). Correlation coefficients between AOA and AOM was 0.54 (p < 0.0001) for left hippocampi and 0.61 (p < 0.0001) for right hippocampi. Conclusion Even though AOA has moderate correlation with the gold standard it is not superior to average VSA and should be implemented with care in clinical practice.

Support vector machine-based classification of neuroimages in Alzheimer's disease: direct comparison of FDG-PET, rCBF-SPECT and MRI data acquired from the same individuals

Objective: To conduct the first support vector machine (SVM)-based study comparing the diagnostic accuracy of T1-weighted magnetic resonance imaging (T1-MRI), F-fluorodeoxyglucose-positron emission tomography (FDG-PET) and regional cerebral blood flow single-photon emission computed tomography (rCBF-SPECT) in Alzheimer's disease (AD). Method: Brain T1-MRI, FDG-PET and rCBF-SPECT scans were acquired from a sample of mild AD patients (n=20) and healthy elderly controls (n=18). SVM-based diagnostic accuracy indices were calculated using whole-brain information and leave-one-out cross-validation. Results: The accuracy obtained using PET and SPECT data were similar. PET accuracy was 68∼71% and area under curve (AUC) 0.77∼0.81; SPECT accuracy was 68∼74% and AUC 0.75∼0.79, and both had better performance than analysis with T1-MRI data (accuracy of 58%, AUC 0.67). The addition of PET or SPECT to MRI produced higher accuracy indices (68∼74%; AUC: 0.74∼0.82) than T1-MRI alone, but these were not clearly superior to the isolated neurofunctional modalities. Conclusion: In line with previous evidence, FDG-PET and rCBF-SPECT more accurately identified patients with AD than T1-MRI, and the addition of either PET or SPECT to T1-MRI data yielded increased accuracy. The comparable SPECT and PET performances, directly demonstrated for the first time in the present study, support the view that rCBF-SPECT still has a role to play in AD diagnosis.

Argentina-Alzheimer's disease neuroimaging initiative (Arg-ADNI): neuropsychological evolution profile after one-year follow up / Argentina-Alzheimer's disease neuroimaging initiative (Arg-ADNI): evolução do perfil neuropsicológico depois de um ano de seguimento

ABSTRACT The Argentina-Alzheimer's disease neuroimaging initiative (Arg-ADNI) study is a longitudinal prospective cohort of 50 participants at a single institution in Buenos Aires, Argentina. Longitudinal assessments on a neuropsychological test battery were performed on 15 controls, 24 mild cognitive impairment (MCI) patients and 12 Alzheimer's disease (AD) dementia patients. In our study population, there was a high prevalence of positive AD biomarkers in the AD group, 92.3% (12/13); and a low prevalence in the normal controls, 20%; almost half (48%) of the patients diagnosed with MCI had positive amyloid detection. After a one year, the significant differences found at baseline on neuropsychological testing were similar at the follow-up assessment even though the AD group had significantly altered its functional performance (FAQ and CDR). The exception was semantic fluency, which showed greater impairment between the AD group and MCI and normal controls respectively. For these tests, the addition of AD biomarkers as a variable did not significantly alter the variations previously found for the established clinical group's model. Finally, the one-year conversion rate to dementia was 20% in the MCI cohort.

RESUMO El estudio de Argentina-Alzheimer's Disease Neuroimaging Initiative (Arg-ADNI) es una cohorte prospectiva de 50 pacientes seguidos en una misma institución. Fueron evaluados cognitivamente 15 controles normales (CN), 24 sujetos con deterioro cognitivo leve (DCL) y 12 con demencia tipo Alzheimer (DTA) leve. En los DTA, 92,3% tuvieron biomarcadores positivos para Alzheimer y 20% en los CN. Casi la mitad de los DCL presentaron biomarcadores positivos. Después de un año de seguimiento, la diferencias significativas halladas en la visita de inicio en las pruebas cognitivas fueron similares al año aunque los DTA tuvieron empeoramiento funcional medido en el FAQ y CDR. La excepción fue la fluencia semántica, la cual mostró mayor declinación entre DTA y los demás grupos. La incorporación de los biomarcadores como variable no alteró significativamente los hallazgos de grupo. La tasa de conversión a demencia anual fue del 20%.

Biomarcadores y marcadores de imagen de la enfermedad de Alzheimer / Biomarkers and image markers of Alzheimer's disease

Este artículo presenta un análisis desde el punto de vista bibliográfico de marcadores y biomarcadores de la enfermedad de Alzheimer (EA). Las metodologías usadas fueron los marcadores de imágenes (Resonancia Magnética y Tomografía por emisión de positrones) y biomarcadores de la proteína BA42, de la proteína Tau y de la Apoliproteína E (ALPE). De esta manera, son de importancia los niveles de BA42 disminuidos, la Tau incrementada, los polimorfismos de ALPE y las alteraciones constatadas en los marcadores de imagen, como factores de riesgo esenciales para el desarrollo de la EA. Se realiza una revisión de la literatura con respecto a los hallazgos clínicos de esta enfermedad.

This article presents a bibliographical analysis of markers and biomarkers of Alzheimer's disease (AD). The methodologies used were the imaging markers (Magnetic Resonance and Positron Emission Tomography) and biomarkers of the BA42 protein, Tau protein and Apoliprotein E (ALPE). Thus, decreased levels of BA42, increased Tau, ALPE polymorphisms, and alterations in imaging markers are important as risk factors for the development of AD. A review of the literature is made regarding the clinical findings of this disease.

Neuropsiquiatra: pet y spect / Neuropsychiatrist: pet and spect

Functional brain imaging with PET and SPECT have a definitive and well established role in the investigation of a variety of conditions such as dementia, epilepsy and drug addiction. With these methods it is possible to detect early rCBF (regional Cerebral Blood Flow) changes seen in dementia (even before clinical symptoms) and differentiate Alzheimer's disease from other dementias by means of the rCBF pattern change. 18-F-FDG PET imaging is a useful tool in partial epilepsy because both rCBF and brain metabolism are compromised at the epileptogenic focus. During the seizure, rCBF dramatically increases locally. Using SPECT it is possible to locate such foci with 97% accuracy. In drug addiction, particularly with cocaine, functional imaging has proven to be very sensitive to detect brain flow and metabolism derangement early in the course of this condition. These findings are important in many ways: prognostic value, they are used as a powerful reinforcement tool and to monitor functional recovery with rehabilitation. There are many other conditions in which functional brain imaging is of importance such as acute stroke treatment assessment, trauma rehabilitation and in psychiatric and abnormal movement diseases specially with the development of receptor imaging.

Existen numerosas indicaciones claramente establecidas para el uso del SPECT y PET en patología neuro-psiquiátrica, particularmente en el estudio de demencias, epilepsia y adicción a drogas. Estos métodos permiten detectar precozmente (aun antes de las manifestaciones clínicas) cambios en la perfusión y metabolismo cerebral en pacientes con demencias. Es posible además diferenciar la enfermedad de Alzheimer de otras causas de demencia, analizando el patrón de la alteración neuro- funcional. En epilepsia parcial, tanto el metabolismo como la perfusión están alterados en el foco epileptogénico, lo que puede ser detectado con F-18FDG PET. Durante la crisis epiléptica, el flujo sanguíneo puede aumentar dramáticamente en el foco epileptogénico, lo que puede ser detectado con SPECT con 97% de certeza. En pacientes drogadictos, especialmente a la cocaína, estos métodos han demostrado ser muy sensibles para la detección precoz de cambios en el flujo y metabolismo cerebral, lo que es clínicamente importante en varios aspectos: 1) Tiene valor pronóstico (neuro-funcional), 2) Se puede usar para aumentar la adherencia a la terapia y 3) Permite evaluar objetivamente la recuperación funcional. Existen muchas otras indicaciones presentes y futuras, por ejemplo: en la monitorización de la revascularización en accidentes vasculares cerebrales agudos, rehabilitación post TEC, estudio de patología psiquiátrica y movimientos anormales especialmente con el desarrollo de radioligandos.