Random forest to differentiate dementia with Lewy bodies from Alzheimer's disease.
Ontology highlight
ABSTRACT: Introduction
The aim of this study was to build a random forest classifier to improve the diagnostic accuracy in differentiating dementia with Lewy bodies (DLB) from Alzheimer's disease (AD) and to quantify the relevance of multimodal diagnostic measures, with a focus on electroencephalography (EEG).Methods
A total of 66 DLB, 66 AD patients, and 66 controls were selected from the Amsterdam Dementia Cohort. Quantitative EEG (qEEG) measures were combined with clinical, neuropsychological, visual EEG, neuroimaging, and cerebrospinal fluid data. Variable importance scores were calculated per diagnostic variable.Results
For discrimination between DLB and AD, the diagnostic accuracy of the classifier was 87%. Beta power was identified as the single-most important discriminating variable. qEEG increased the accuracy of the other multimodal diagnostic data with almost 10%.Discussion
Quantitative EEG has a higher discriminating value than the combination of the other multimodal variables in the differentiation between DLB and AD.
SUBMITTER: Dauwan M
PROVIDER: S-EPMC5050257 | biostudies-literature | 2016
REPOSITORIES: biostudies-literature
Publications
Random forest to differentiate dementia with Lewy bodies from Alzheimer's disease.
Dauwan Meenakshi M van der Zande Jessica J JJ van Dellen Edwin E Sommer Iris E C IE Sommer Iris E C IE Scheltens Philip P Lemstra Afina W AW Stam Cornelis J CJ
Alzheimer's & dementia (Amsterdam, Netherlands) 20160819
<h4>Introduction</h4>The aim of this study was to build a random forest classifier to improve the diagnostic accuracy in differentiating dementia with Lewy bodies (DLB) from Alzheimer's disease (AD) and to quantify the relevance of multimodal diagnostic measures, with a focus on electroencephalography (EEG).<h4>Methods</h4>A total of 66 DLB, 66 AD patients, and 66 controls were selected from the Amsterdam Dementia Cohort. Quantitative EEG (qEEG) measures were combined with clinical, neuropsychol ...[more]
Similar Datasets
Project description:BackgroundEarly differential diagnosis of Alzheimer's disease (AD) and dementia with Lewy bodies (DLB) is important for treatment and disease management, but it remains challenging. Although computer-based drawing analysis may help differentiate AD and DLB, it has not been studied.ObjectiveWe aimed to identify the differences in features characterizing the drawing process between AD, DLB, and cognitively normal (CN) individuals, and to evaluate the validity of using these features to identify and differentiate AD and DLB.MethodsWe collected drawing data with a digitizing tablet and pen from 123 community-dwelling older adults in three clinical diagnostic groups of mild cognitive impairment or dementia due to AD (n = 47) or Lewy body disease (LBD; n = 27), and CN (n = 49), matched for their age, sex, and years of education. We then investigated drawing features in terms of the drawing speed, pressure, and pauses.ResultsReduced speed and reduced smoothness in speed and pressure were observed particularly in the LBD group, while increased pauses and total durations were observed in both the AD and LBD groups. Machine-learning models using these features achieved an area under the receiver operating characteristic curve (AUC) of 0.80 for AD versus CN, 0.88 for LBD versus CN, and 0.77 for AD versus LBD.ConclusionOur results indicate how different types of drawing features were particularly discriminative between the diagnostic groups, and how the combination of these features can facilitate the identification and differentiation of AD and DLB.
| S-EPMC9697058 | biostudies-literature
Project description:Many subjects with neuropathologically-confirmed dementia with Lewy bodies (DLB) are never diagnosed during life, instead being categorized as Alzheimer's disease dementia (ADD) or unspecified dementia. Unrecognized DLB therefore is a critical impediment to clinical studies and treatment trials of both ADD and DLB. There are studies that suggest that olfactory function tests may be able to distinguish DLB from ADD, but few of these had neuropathological confirmation of diagnosis. We compared University of Pennsylvania Smell Identification Test (UPSIT) results in 257 subjects that went on to autopsy and neuropathological examination. Consensus clinicopathological diagnostic criteria were used to define ADD and DLB, as well as Parkinson's disease with dementia (PDD), with (PDD+AD) or without (PDD-AD) concurrent AD; a group with ADD and Lewy body disease (LBD) not meeting criteria for DLB (ADLB) and a clinically normal control group were also included. The subjects with DLB, PDD+AD and PDD-AD all had lower (one-way ANOVA p < 0.0001, pairwise Bonferroni p < 0.05) first and mean UPSIT scores than the ADD, ADLB or control groups. For DLB subjects with first and mean UPSIT scores less than 20 and 17, respectively, Firth logistic regression analysis, adjusted for age, gender and mean MMSE score, conferred statistically significant odds ratios of 17.5 and 18.0 for the diagnosis, vs ADD. For other group comparisons (PDD+AD and PDD-AD vs ADD) and UPSIT cutoffs of 17, the same analyses resulted in odds ratios ranging from 16.3 to 31.6 (p < 0.0001). To our knowledge, this is the largest study to date comparing olfactory function in subjects with neuropathologically-confirmed LBD and ADD. Olfactory function testing may be a convenient and inexpensive strategy for enriching dementia studies or clinical trials with DLB subjects, or conversely, reducing the inclusion of DLB subjects in ADD studies or trials.
| S-EPMC7176090 | biostudies-literature
Project description:BackgroundDementia often results in postural control impairment, which could signify central nervous system dysfunction. However, no studies have compared postural control characteristics among various types of dementia. This study aimed to compare static postural control in patients with Alzheimer's disease (AD), dementia with Lewy bodies (DLB), and vascular dementia (VaD).MethodsCross-sectional relationship between the clinical diagnoses (AD, DLB, VaD, or normal cognition [NC]) of outpatients at a memory clinic and their upright postural control characteristics were examined. In the postural control test, participants were instructed to maintain a static upright standing on a stabilometer for 60 seconds under the eyes-open and eyes-closed conditions. Forty postural control parameters, including distance, position, and velocity in the anterior-posterior and medio-lateral directions, derived from the trajectory of the center of mass sway, were calculated. The characteristics of each type of dementia were compared to those of NC, and the differences among the 3 types of dementia were evaluated using linear regression models.ResultsThe study included 1 789 participants (1 206 with AD, 111 with DLB, 49 with VaD, and 423 with NC). Patients with AD exhibited distinct postural control characteristics, particularly in some distance and velocity parameters, only in the eyes-closed condition. Those with DLB exhibited features in the mean position in the anterior-posterior direction. In patients with VaD, significant differences were observed in most parameters, except the power spectrum.ConclusionsPatients with AD, DLB, and VaD display disease-specific postural control characteristics when compared to cognitively normal individuals.
| S-EPMC10949438 | biostudies-literature
Project description:BackgroundDementia with Lewy bodies (DLB) is a common cause of dementia in the elderly population after Alzheimer's disease (AD), and at early stages differential diagnosis between DLB and AD might be difficult due to their symptomatic overlap, e.g. cognitive and memory impairments. We aimed to investigate functional brain differences between both diseases in patients recently diagnosed.MethodsWe investigated regional functional synchronizations using regional homogeneity (ReHo) in patients clinically diagnosed with DLB (n = 19) and AD (n = 18), and for comparisons we also included healthy controls (HC, n = 16). Patient groups were matched by age, education, and by the level of cognitive impairment (MMSE p-value = 0.36). Additionally, correlations between ReHo values and clinical scores were investigated.ResultsThe DLB group showed lower ReHo in sensory-motor cortices and higher ReHo in left middle temporal gyrus when compared with HCs (p-value < 0.001 uncorrected). The AD group demonstrated lower ReHo in the cerebellum and higher ReHo in the left/right lingual gyri, precuneus cortex, and other occipital and parietal regions (p-value < 0.001 uncorrected).ConclusionsOur results agree with previous ReHo investigations in Parkinson's disease (PD), suggesting that functional alterations in motor-related regions might be a characteristic of the Lewy body disease spectrum. However, our results in AD contradict previously reported findings for this disease and ReHo, which we speculate are a reflection of compensatory brain responses at early disease stages. ReHo differences between patient groups were at regions related to the default mode and sensory-motor resting state networks which might reflect the aetiological divergences in the underlying disease processes between AD and DLB.
| S-EPMC4894061 | biostudies-literature
Project description:As the differential diagnosis of dementias based on established clinical criteria is often difficult, biomarkers for applicable diagnostic testing are currently under intensive investigation. Amyloid plaques deposited in the brain of patients suffering from Alzheimer's disease, dementia with Lewy bodies (DLB) and Parkinson's disease dementia (PDD) mainly consist of carboxy-terminally elongated forms of amyloid-beta (Abeta) peptides, such as Abeta1-42. Absolute Abeta1-42 levels in CSF have shown diagnostic value for the diagnosis of Alzheimer's disease, but the discrimination among Alzheimer's disease, DLB and PDD was poor. A recently established quantitative urea-based Abeta-sodium-dodecylsulphate-polyacrylamide-gel-electrophoresis with Western immunoblot (Abeta-SDS-PAGE/immunoblot) revealed a highly conserved Abeta peptide pattern of the carboxy-terminally truncated Abeta peptides 1-37, 1-38, 1-39 in addition to 1-40 and 1-42 in human CSF. We used the Abeta-SDS-PAGE/immunoblot to investigate the CSF of 23 patients with Alzheimer's disease, 21 with DLB, 21 with PDD and 23 non-demented disease controls (NDC) for disease-specific alterations of the Abeta peptide patterns in its absolute and relative quantities. The diagnostic groups were matched for age and severity of dementia. The present study is the first attempt to evaluate the meaning of Abeta peptide patterns in CSF for differential diagnosis of the three neurodegenerative diseases--Alzheimer's disease, DLB and PDD. The Abeta peptide patterns displayed disease-specific variations and the ratio of the differentially altered Abeta1-42 to the Abeta1-37 levels subsequently discriminated all diagnostic groups from each other at a highly significant level, except DLB from PDD. Additionally, a novel peptide with Abeta-like immunoreactivity was observed constantly in the CSF of all 88 investigated patients. The pronounced percentage increase of this peptide in DLB allowed a highly significant discrimination from PDD. Using a cut-off point of 0.954%, this marker yielded a diagnostic sensitivity and specificity of 81 and 71%, respectively. From several lines of indication, we consider this peptide to represent an oxidized alpha-helical form of Abeta1-40 (Abeta1-40*). The increased abundance of Abeta1-40* probably reflects a disease-specific alteration of the Abeta1-40 metabolism in DLB. We conclude that Abeta peptide patterns reflect disease-specific pathophysiological pathways of different dementia syndromes as distinct neurochemical phenotypes. Although Abeta peptide patterns failed to fulfil the requirements for a sole biomarker, their combined evaluation with other biomarkers is promising in neurochemical dementia diagnosis. It is noteworthy that DLB and PDD exhibit distinct clinical temporal courses, despite their similar neuropathological appearance. Their distinct molecular phenotypes support the view of different pathophysiological pathways for each of these neurodegenerative diseases.
| S-EPMC5403412 | biostudies-literature
Project description:We studied the dynamic functional connectivity profile of dementia with Lewy bodies (DLB) and Alzheimer's disease (AD) compared to controls, how it differs between the two dementia subtypes, and a possible relation between dynamic connectivity alterations and temporally transient clinical symptoms in DLB. Resting state fMRI data from 31 DLB, 29 AD, and 31 healthy control participants were analyzed using dual regression to determine between-network functional connectivity. Subsequently, we used a sliding window approach followed by k-means clustering and dynamic network analyses to study dynamic functional connectivity. Dynamic connectivity measures that showed significant group differences were tested for correlations with clinical symptom severity. Our results show that AD and DLB patients spent more time than controls in sparse connectivity configurations with absence of strong positive and negative connections and a relative isolation of motor networks from other networks. Additionally, DLB patients spent less time in a more strongly connected state and the variability of global brain network efficiency was reduced in DLB compared to controls. There were no significant correlations between dynamic connectivity measures and clinical symptom severity. An inability to switch out of states of low inter-network connectivity into more highly and specifically connected network configurations might be related to the presence of dementia in general as it was observed in both AD and DLB. In contrast, the loss of global efficiency variability in DLB might indicate the presence of an abnormally rigid brain network and the lack of economical dynamics, factors which could contribute to cognitive slowing and an inability to respond appropriately to situational demands.
| S-EPMC6462776 | biostudies-literature
Project description:Dementia with Lewy bodies (DLB) is one of the most common causes of degenerative dementia, after Alzheimer's disease (AD), and presents pathological and clinical overlap with both AD and Parkinson's disease (PD). Consequently, only one in three DLB cases is diagnosed correctly. Platelets, previously related to neurodegeneration, contain microRNAs (miRNAs) whose analysis may provide disease biomarkers. Here, we profiled the whole platelet miRNA transcriptome from DLB patients and healthy controls. Differentially expressed miRNAs were further validated in three consecutive studies from 2017 to 2019 enrolling 162 individuals, including DLB, AD, and PD patients, and healthy controls. Results comprised a seven-miRNA biosignature, showing the highest diagnostic potential for the differentiation between DLB and AD. Additionally, compared to controls, two miRNAs were down-regulated in DLB, four miRNAs were up-regulated in AD, and two miRNAs were down-regulated in PD. Predictive target analysis identified three disease-specific clusters of pathways as a result of platelet-miRNA deregulation. Our cross-sectional study assesses the identification of a novel, highly specific and sensitive platelet-associated miRNA-based biosignature, which distinguishes DLB from AD.
| S-EPMC8466211 | biostudies-literature
Project description:Patients who have dementia with Lewy bodies and Alzheimer's disease show early degeneration of the cholinergic nucleus basalis of Meynert. However, how white matter projections between the nucleus basalis of Meynert and the cortex are altered in neurodegenerative disease is unknown. Tractography of white matter pathways originating from the nucleus basalis of Meynert was performed using diffusion-weighted imaging in 46 patients with Alzheimer's disease dementia, 48 with dementia with Lewy bodies, 35 with mild cognitive impairment with Alzheimer's disease, 38 with mild cognitive impairment with Lewy bodies and 71 control participants. Mean diffusivity of the resulting pathways was compared between groups and related to cognition, attention, functional EEG changes and dementia conversion in the mild cognitive impairment groups. We successfully tracked a medial and a lateral pathway from the nucleus basalis of Meynert. Mean diffusivity of the lateral pathway was higher in both dementia and mild cognitive impairment groups than controls (all P < 0.03). In the patient groups, increased mean diffusivity of this pathway was related to more impaired global cognition (β = -0.22, P = 0.06) and worse performance on an attention task (β = 0.30, P = 0.03). In patients with mild cognitive impairment, loss of integrity of both nucleus basalis of Meynert pathways was associated with increased risk of dementia progression [hazard ratio (95% confidence interval), medial pathway: 2.51 (1.24-5.09); lateral pathway: 2.54 (1.24-5.19)]. Nucleus basalis of Meynert volume was reduced in all clinical groups compared to controls (all P < 0.001), but contributed less strongly to cognitive impairment and was not associated with attention or dementia conversion. EEG slowing in the patient groups as assessed by a decrease in dominant frequency was associated with smaller nucleus basalis of Meynert volumes (β = 0.22, P = 0.02) and increased mean diffusivity of the lateral pathway (β = -0.47, P = 0.003). We show that degeneration of the cholinergic nucleus basalis of Meynert in Alzheimer's disease and dementia with Lewy bodies is accompanied by an early reduction in integrity of white matter projections that originate from this structure. This is more strongly associated with cognition and attention than the volume of the nucleus basalis of Meynert itself and might be an early indicator of increased risk of dementia conversion in people with mild cognitive impairment.
| S-EPMC9166545 | biostudies-literature
Project description:Background and purposeThis study aimed to determine the neuropsychological differences between patients with early-stage Alzheimer's disease (AD) and dementia with Lewy bodies (DLB) with a Clinical Dementia Rating (CDR) score of ≤1.MethodsWe examined 168 patients with AD (126 with CDR score=0.5, 42 with CDR score=1) and 169 patients with DLB (104 with CDR score=0.5, 65 with CDR score=1) whose diagnoses were supported by 18F-flobetaben positron-emission tomography (PET) and 18F-N-(3-fluoropropyl)-2β-carbon ethoxy-3β-(4-iodophenyl) nortropane PET. Neuropsychological test scores were compared after controlling for age, sex, and education duration. Using a cutoff motor score on the Unified Parkinson's Disease Rating Scale of 20, patients with AD were further divided into AD with parkinsonism (ADP+, n=86) and AD without parkinsonism (ADP-, n=82).ResultsAt CDR scores of both 0.5 and 1, the DLB group had lower scores on the attention (digit-span forward at CDR score=0.5 and backward at CDR score=1), visuospatial, and executive (color reading Stroop test at CDR score=0.5 and phonemic fluency test, Stroop tests, and digit symbol coding at CDR score=1) tests than the AD group, but higher scores on the memory tests. The ADP- and ADP+ subgroups had comparable scores on most neuropsychological tests, but the ADP+ subgroup had lower scores on the color reading Stroop test.ConclusionsPatients with DLB had worse attention, visuospatial, and executive functions but better memory function than patients with AD. Parkinsonism was not uncommon in the patients with AD and could be related to attention and executive dysfunction.
| S-EPMC10622731 | biostudies-literature
Project description:AimsThe aim of this study was to identify brain regions with local, structural, and functional abnormalities in dementia with Lewy bodies (DLB) and uncover the differences between DLB and Alzheimer's disease (AD). The neural networks involved in the identified abnormal brain regions were further described.MethodsPubMed, Web of Science, OVID, Science Direct, and Cochrane Library databases were used to identify neuroimaging studies that included DLB versus healthy controls (HCs) or DLB versus AD. The coordinate-based meta-analysis and functional meta-analytic connectivity modeling were performed using the activation likelihood estimation algorithm.ResultsEleven structural studies and fourteen functional studies were included in this quantitative meta-analysis. DLB patients showed a dysfunction in the bilateral inferior parietal lobule and right lingual gyrus compared with HC patients. DLB patients showed a relative preservation of the medial temporal lobe and a tendency of lower metabolism in the right lingual gyrus compared with AD. The frontal-parietal, salience, and visual networks were all abnormally co-activated in DLB, but the default mode network remained normally co-activated compared with AD.ConclusionsThe convergence of local brain regions and co-activation neural networks might be potential specific imaging markers in the diagnosis of DLB. This might provide a pathway for the neural regulation in DLB patients, and it might contribute to the development of specific interventions for DLB and AD.
| S-EPMC8739049 | biostudies-literature