98%
921
2 minutes
20
Background: Very late-onset psychosis (VLOP) is associated with higher rates of dementia but the proportion who develop dementia with Lewy bodies (DLB) is unknown. We aimed to identify individuals with VLOP who develop dementia and DLB and characterize the risk factors for progression.
Methods: Anonymized data were retrieved from electronic records for individuals with VLOP. Patients developing dementia after psychosis were identified, in addition to those with 2 core features of DLB at the time of dementia or DLB identified by a natural language processing application (NLP-DLB). Demographic factors, Health of the National Outcome Scale (HoNOS) and symptoms at index psychosis were explored as predictors of progression to dementia.
Results: In 1425 patients with VLOP over 4.29 years (mean) follow up, 197 (13.8%) received a subsequent diagnosis of dementia. Of these, 24.4% ( = 48) had 2 core features of DLB and 6% ( = 12) had NLP-DLB. In cox proportional hazard models, older age and cognitive impairment at the time of psychosis were associated with increased risk of incident dementia. Visual hallucinations and 2+ core features of DLB at index psychosis were associated with increased risk of dementia with 2+ symptoms of DLB but not all-cause dementia. Two or more core features of DLB at index psychosis were associated with 81% specificity and 67% sensitivity for incident NLP-DLB.
Conclusions: In patients with VLOP who develop dementia, core features of DLB are common. Visual hallucinations or two core features of DLB in VLOP should prompt clinicians to consider DLB and support further investigation.
Download full-text PDF |
Source |
---|---|
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC11496220 | PMC |
http://dx.doi.org/10.1017/S0033291724001922 | DOI Listing |
IEEE J Biomed Health Inform
September 2025
Epilepsy, a highly individualized neurological disorder, affects millions globally. Electroencephalography (EEG) remains the cornerstone for seizure diagnosis, yet manual interpretation is labor-intensive and often unreliable due to the complexity of multi-channel, high-dimensional data. Traditional machine learning models often struggle with overfitting and fail in fully capturing the highdimensional, temporal dynamics of EEG signals, restricting their clinical utility.
View Article and Find Full Text PDFIEEE Trans Pattern Anal Mach Intell
September 2025
Camouflaged Object Segmentation (COS) faces significant challenges due to the scarcity of annotated data, where meticulous pixel-level annotation is both labor-intensive and costly, primarily due to the intricate object-background boundaries. Addressing the core question, "Can COS be effectively achieved in a zero-shot manner without manual annotations for any camouflaged object?", we propose an affirmative solution. We analyze the learned attention patterns for camouflaged objects and introduce a robust zero-shot COS framework.
View Article and Find Full Text PDFProc Natl Acad Sci U S A
September 2025
Department of Biology, Massachusetts Institute of Technology, Cambridge, MA 02139.
The mutagenic translesion synthesis (TLS) pathway, which is critically dependent on REV1's ability to recruit inserter TLS polymerases and the POLζ extender polymerase, enables cancer cells to bypass DNA lesions while introducing mutations that likely contribute to the development of chemotherapy resistance and secondary malignancies. Targeting this pathway represents a promising therapeutic strategy. Here, we demonstrate that the expression of the C-terminal domain (CTD) of human REV1, a ca.
View Article and Find Full Text PDFCereb Cortex
August 2025
Section on Functional Imaging Methods & Functional MRI Core Facility, National Institute of Mental Health, 10 Center Drive, Rm 1D80, Bethesda, MD 20892, United States.
Statistical Parametric Mapping (SPM) has been profoundly influential to neuroimaging as it has fostered rigorous, statistically grounded structure for model-based inferences that have led to mechanistic insights about the human brain over the past 30 years. The statistical constructs shared with the world through SPM have been instrumental for deriving meaning from neuroimaging data; however, they require simplifying assumptions which can provide results that, while statistically sound, may not accurately reflect the mechanisms of brain function. A platform that fosters the exploration of the rich and varying neuronal and physiologic underpinnings of the measured signals and their associations to behavior and physiologic measures needs a different set of tools.
View Article and Find Full Text PDFAm J Forensic Med Pathol
September 2025
Department of Pathology, St Louis University School of Medicine, Office of the Medical Examiner - City of St. Louis, St. Louis, MO.
Myotonic dystrophy type 1, or dystrophia myotonica type 1 (DM1), is a multisystem disorder inherited in an autosomal dominant manner. It is caused by a CTG tri-nucleotide expansion in the 3'-untranslated region (3'-UTR) of the dystrophia myotonia protein kinase (DMPK) gene. Core clinical features include progressive skeletal muscle weakness, myotonia, and systemic complications, with premature mortality most often due to respiratory or cardiac dysfunction.
View Article and Find Full Text PDF