Enhanced CBAMWDNet: A deep learning approach for accurate dementia multiclassification using MRI scans

The Relationship Between Dementia Subtypes and Nutritional Parameters in Older Adults

The aim of this study was to assess the prescription practices and judgment of efficacy of physicians of drugs used for the cognitive and non-cognitive symptoms of dementia. Physicians from 88 Italian Alzheimer Evaluation Units were surveyed by means of a structured questionnaire assessing the proportion of patients with four different types of dementia prescribed with drugs for cognitive and non-cognitive symptoms, and physicians' perceived efficacy of cholinesterase inhibitors. The Units prescribed cholinesterase inhibitors for 73 patients per year on average. Cholinesterase inhibitors are prescribed to 90% of patients with Alzheimer's disease (AD), 80% with with Lewy body dementia (LBD), and 35-45% with vascular dementia (VD) and frontotemporal lobar degeneration (FTLD). Selective serotonin uptake inhibitors (SSRIs) are prescribed for 28-45% of patients with all dementias except LBD (16%). Atypical neuroleptics were prescribed for 23-31% of patients, with no difference across types of dementia. Other drugs, such as ginkgo and nootropics, were prescribed less frequently, except in VD (20%). The perceived efficacy on cognitive and non-cognitive symptoms, assessed on a 0-to-10 ordinal scale, was highest in AD (4.3-6.1), intermediate in LBD (3.5-5.3) and VD (3.3- 4.7), and lowest in FTLD (2.0-2.7). The data indicate that, in specialized Italian centers, cholinesterase inhibitors and atypical neuroleptics are largely used in patients with AD and LBD, but the former are prescribed off-label to a remarkable proportion of patients with VD and FTLD. The efficacy of cholinesterase inhibitors is perceived to be highest in AD and poorest in FTLD. Perceived efficacy is affected more by whom is treated than by what is used.

Functional near-infrared spectroscopy (fNIRS) is commonly used to study human brain function by measuring the hemodynamic signals originating from cortical activation and provides a new noninvasive detection method for identifying dementia. To investigate the fNIRS imaging technique and its clinical application in differential diagnosis of subtype dementias including frontotemporal lobe dementia, Lewy body dementia, Parkinson's disease dementia (PDD) and Alzheimer's disease (AD). Four patients with different types of dementia were examined with fNIRS during two tasks and a resting state. We adopted the verbal fluency task, working memory task and resting state task. Each patient was compared on the same task. We conducted and analyzed the fNIRS data using a general linear model and Pearson's correlation analysis. Compared with other types of dementias, fNIRS showed the left frontotemporal and prefrontal lobes to be poorly activated during the verbal fluency task in frontotemporal dementia. In Lewy body dementia, severe asymmetry of prefrontal lobes appeared during both verbal fluency and working memory tasks, and the patient had low functional connectivity during a resting state. In PDD, the patient's prefrontal cortex showed lower excitability than the temporal lobe during the verbal fluency task, while the prefrontal cortex showed higher excitability during the working memory task. The patient with AD showed poor prefrontal and temporal activation during the working memory task, and more activation of frontopolar instead of the dorsolateral prefrontal cortex. Different hemodynamic characteristics of four types of dementia (as seen by fNIRS imaging) provides evidence that fNIRS can serve as a potential tool for the diagnosis between dementia subtypes.

The Benton Judgment of Line Orientation (JLO) test is one of the most frequently used tests for assessing visuospatial function. This study aimed to determine the diagnostic and differential performance of JLO for different types of dementia. A total of 258 participants, including 68 patients with Alzheimer's disease (AD), 86 with subcortical ischemic vascular dementia (SIVD), 30 with frontotemporal lobar degeneration (FTLD), 22 with Lewy body dementia (LBD), and 52 cognitively unimpaired (CU) controls, were enrolled from a memory clinic. The total scores and error types in the JLO test were compared between groups. Receiver operating characteristic curve analyses were used to estimate the diagnostic and differential abilities of the JLO test for patients with different types of dementia. We found that the JLO score was significantly lower in patients with AD, SIVD, FTLD, or LBD than in CU controls (12.90 ± 8.72 versus 17.06 ± 6.14 versus 15.47 ± 8.39 versus 9.23 ± 8.96 versus 21.69 ± 3.72, respectively; all p < 0.05). In particular, for patients with LBD, the JLO score was significantly lower than that for patients in the other groups (all p < 0.05) and showed excellent performance in distinguishing LBD patients from CU controls, with an AUC of 0.888 (sensitivity 72.73% and specificity 94.23%) at a cutoff value of 16. Intraquadrant oblique error was the most common type of error in dementia patients. The JLO test is an effective tool for evaluating visuospatial function in patients with dementia, particularly for identifying LBD patients.

Mutation and association analyses of dementia-causal genes in Han Chinese patients with early-onset and familial Alzheimer's disease

Introduction: Diagnosing the etiology underlying dementia symptoms can be difficult because of the broad nature of shared cognitive impairments across dementia subtypes. Therefore, we sought to differentiate the cognitive profiles of Alzheimer’s disease (AD) from vascular dementia (VaD) and Lewy body dementia (LBD). Methods: PubMed, ScienceDirect, Web of Science, Google Scholar, and PsychINFO were searched for studies comparing the cognitive profile of AD to those of VaD and LBD along the domains of memory, language, and executive function. Results: Short-term and episodic memory were more severely impaired in AD than VaD and LBD. Semantic memory was more impaired in AD than LBD, but it was similarly impaired in AD and VaD. Semantic fluency was worse in AD than in VaD, and phonemic fluency was worse in AD compared to VaD and LBD. Naming was more impaired in AD compared to VaD and LBD. Executive function impairments were similar or less severe in AD relative to VaD and LBD. Discussion: Findings may be explained through neuropathological correlates of each disease. Tau proteins targeting the medial temporal lobes and synaptic loss in prefrontal cortices in AD may explain greater memory deficits in AD relative to VaD and LBD. In those with AD, the temporal lobes undergo greater atrophy than in those with VaD and LBD, possibly contributing to the greater semantic fluency impairments in AD. Greater white matter loss in frontal lobes in VaD may be a reason for a worse phonemic fluency in VaD relative to AD. Executive function impairments may be attributable to more deep white matter hyperintensities in those with VaD and more dopaminergic dysfunction of the basal ganglia in those with LBD relative to those with AD. Conclusion: Understanding the cognitive profiles that differentiate AD from VaD and LBD would aid in more efficient and accurate diagnoses of dementia etiologies. Diagnoses could be further improved by using cognitive assessment in addition to neural and physiological measures. This knowledge may help identify individuals at risk of developing dementia, helping clinicians intervene early and prevent progression to severe stages.

Individuals with mild cognitive impairment (MCI) are at increased risk for dementia of Alzheimer's type (DAT), vascular dementia (VaD), Lewy Body (LBD) and Fronto-temporal dementias (FTD). Risk factors and conversion rates of MCI to dementia have not been thoroughly investigated in developing countries. Chinese and English versions of Mini-Mental State Examination were administered serially among well-matched subjects from two clinics located in Xi'an, China and Houston, USA. Subtle cognitive impairments were weighed according to MCI criteria as defined previously. Subjects with MCI were followed for an additional 3 years after their identification. Diagnoses of VaD and DAT were made according to established criteria. During screening period, 73 American and 65 Chinese individuals were identified with MCI. After 3 years of MCI follow-up, of the 73 American MCI subjects, 35 (47.9%) developed DAT and 15 (20.5%) developed VaD. Of the 65 Chinese MCI subjects, 12 (18.5%) developed DAT and 19 (29.2%) developed VaD. According to Kaplan-Meier analysis, Chinese MCI subjects, despite their lower educational level, are 1.7 times less likely to progress to DAT and 2.3 times more likely to progress to VaD than American subjects within 3 years of MCI being identified (p<0.01). Data suggest that progression rates of MCI vary considerably among subjects from two countries. American MCI subjects are more prone to DAT, while Chinese subjects are more prone to VaD. Differences in genetic factors, cultures, educational levels, and preventive treatments of vascular risk factors are proposed as responsible for this uneven geographic distribution for different types of dementia.

Anemia is a major public health concern and has a significant impact on women and children worldwide. This condition is caused by a lack of sufficient Red Blood Cells (RBCs). Traditionally, anemia diagnosis has relied on invasive techniques that require blood samples, leading to pain and discomfort. This study explores deep learning techniques for automating and enhancing anemia detection accuracy using the Blood Cell Count and Detection (BCCD) dataset. We employ the VGG16 convolutional neural network architecture augmented with the Convolutional Block Attention Module (CBAM) to boost feature representation and performance. The BCCD dataset, consisting of 364 images with 4888 labeled blood cells, was used for training and evaluation. Our enhanced VGG16 model achieved accuracy for RBCs, WBCs, and platelets with values of 0.84, 0.93, and 0.92, respectively, effectively identifying various blood cell types and detecting anemia. Precision-recall analysis and confusion matrix metrics confirmed the robustness of the model, with high precision and recall rates and minimal false positives and negatives. These results suggest that advanced deep learning models with attention mechanisms, such as CBAM, can significantly improve medical diagnostics by providing reliable, efficient, and accurate early disease detection tools. Future development will be directed towards fine-tuning and validating the proposed model on other medical imaging datasets for clinical applications.

Several of the presenters in this session focused on behavioral and psychological symptoms of dementia (BPSD) as related to different types of dementia, including frontal lobe dementia, Lewy body dementia, and vascular dementia. A considerable amount of this discussion focused on mixed dementias (patients who might have more than one type of dementia) and the possibility of new syndromes of dementia being identified through autopsy studies and other neuropathologic studies. Dr. Hendrie described an autopsy study that reported that vascular dementia is rare, and that most cases of vascular dementia are in fact Alzheimer's disease (AD) with vascular changes. On autopsy, these patients have the plaques and tangles characteristic of AD, but also have vascular changes. Other researchers have reported that changes related to Lewy body dementia are common in patients with AD. Dr. Hendrie suggested that researchers could learn a great deal if pathologists reported the relative number of plaques and tangles, evidence of Lewy body dementia, or evidence of vascular dementia in each brain, rather than trying to categorize their findings as a single diagnosis.

Aims. To examine the relationship between different types of dementia and hippocampal volume. Methods. Hippocampal volume was measured using FL3D sequence magnetic resonance imaging in 26 Alzheimer's, vascular dementia, mixed dementia, and normal pressure hydrocephalus patients and 15 healthy controls and also hippocampal ratio, analyzed. Minimental scale was used to stratify patients on cognitive function impairments. Results. Hippocampal volume and ratio was reduced by 25% in Alzheimer's disease, 21% in mixed dementia, 11% in vascular dementia and 5% in normal pressure hydrocephalus in comparison to control. Also an asymmetrical decrease in volume of left hippocampus was noted. The severity of dementia increased in accordance to decreasing hippocampal volume. Conclusion. Measurement in hippocampal volume may facilitate in differentiating different types of dementia and in disease progression. There was a correlation between hippocampal volume and severity of cognitive impairment.

Combating dementia is a public health priority, and exercise training is one promising strategy for dementia prevention. However, its efficacy in promoting cognitive outcomes in different types of dementia remains unknown. We conducted a systematic review (N = 27) and meta-analysis (N = 24) of randomized controlled trials with cognitive function as a primary or secondary outcome. We aimed to assess the effect of exercise interventions on the cognitive function of older adults (>60 years) diagnosed with different types of dementia. We synthesized data from 2,441 older adults with dementia. Eleven trials included older adults with multiple types of dementia, eight with Alzheimer's disease, six with unspecified types of dementia, and two with vascular cognitive impairment. We performed random-effects models using robust variance estimation (RVE) and tested potential moderators using the approximate Hotelling-Zhang test (HTZ). Results suggest a small effect of exercise on cognitive function for all-cause dementia (g = 0.18; 95% CI: 0.04, 0.33; p = 0.016); however, the effects did not differ by type of dementia. Moderation analyses showed that trials that did not specify participants' severity of dementia, applied individual-level randomization, and had higher intervention adherence demonstrated larger exercise effects on cognitive function for all-cause dementia. We conclude that exercise promotes small improvements in the cognitive function of older adults with all-cause dementia. More research including different types of dementia is needed if we hope to determine the precise effects of exercise for each type of dementia.

Combating dementia is a public health priority, and exercise training is one promising strategy for dementia prevention. However, its efficacy in promoting cognitive outcomes in different types of dementia remains unknown. We conducted a systematic review (N = 27) and meta-analysis (N = 24) of randomized controlled trials with cognitive function as a primary or secondary outcome. We aimed to assess the effect of exercise interventions on the cognitive function of older adults (>60 years) diagnosed with different types of dementia. We synthesized data from 2,441 older adults with dementia. Eleven trials included older adults with multiple types of dementia, eight with Alzheimer's disease, six with unspecified types of dementia, and two with vascular cognitive impairment. We performed random-effects models using robust variance estimation (RVE) and tested potential moderators using the approximate Hotelling-Zhang test (HTZ). Results suggest a small effect of exercise on cognitive function for all-cause dementia (g = 0.18; 95% CI: 0.04, 0.33; p = 0.016); however, the effects did not differ by type of dementia. Moderation analyses showed that trials that did not specify participants' severity of dementia, applied individual-level randomization, and had higher intervention adherence demonstrated larger exercise effects on cognitive function for all-cause dementia. We conclude that exercise promotes small improvements in the cognitive function of older adults with all-cause dementia. More research including different types of dementia is needed if we hope to determine the precise effects of exercise for each type of dementia.

The cognitive impairment known as dementia affects millions of individuals throughout the globe. The use of machine learning (ML) and deep learning (DL) algorithms has shown great promise as a means of early identification and treatment of dementia. Dementias such as Alzheimer’s Dementia, frontotemporal dementia, Lewy body dementia, and vascular dementia are all discussed in this article, along with a literature review on using ML algorithms in their diagnosis. Different ML algorithms, such as support vector machines, artificial neural networks, decision trees, and random forests, are compared and contrasted, along with their benefits and drawbacks. As discussed in this article, accurate ML models may be achieved by carefully considering feature selection and data preparation. We also discuss how ML algorithms can predict disease progression and patient responses to therapy. However, overreliance on ML and DL technologies should be avoided without further proof. It’s important to note that these technologies are meant to assist in diagnosis but should not be used as the sole criteria for a final diagnosis. The research implies that ML algorithms may help increase the precision with which dementia is diagnosed, especially in its early stages. The efficacy of ML and DL algorithms in clinical contexts must be verified, and ethical issues around the use of personal data must be addressed, but this requires more study.

A simple linear measurement of the minimum width of the medial temporal lobe (MTL) on angled CT scans has been suggested as an accurate ante-mortem marker for Alzheimer’s disease (AD). To determine the clinical utility and specificity of this finding, we performed angled CT scans with 5-mm slices in 116 subjects referred to a geographically based Old Age Psychiatry service in Newcastle. Diagnoses were of NINCDS/ADRDA AD (n = 69, 36 probable and 33 possible). NINDS/AIREN vascular dementia (VaD, n = 25), consensus criteria for dementia with Lewy bodies (DLB, n = 9) and DSM-IV criteria for major depression (n = 13). Subjects were well matched for age. Minimum MTL width was significantly greater in depressed subjects (13.7 mm) compared to those with dementia, though no differences were seen within the dementia groups (AD 10.8, VaD 10.4, and DLB 10.9 mm). An MTL width below 11.5 mm had a sensitivity of 54% (56/103) and a specificity of 77% (10/13) for distinguishing dementia from depression. We conclude that a single cross-sectional measurement of MTL width on CT does not help differentiate between different types of dementia, though it may provide some supportive evidence when distinguishing depression from dementia.

Mortality rates in Alzheimer's disease and non-Alzheimer's dementias: a systematic review and meta-analysis