eduKateSG · Why Science?
Study cognitive ageing with evidence and care—connecting longitudinal cohorts, blood flow, the blood-brain barrier, imaging, multi-omics, digital tools and meaningful outcomes
A fact-checked guide to Dementia Research Centre (Singapore) and its research on Asian dementia, biomarkers, imaging, digital innovation and therapeutics.
LKCMedicine says Dementia Research Centre (Singapore) studies how and why Asians develop dementia, including perfusion and blood-brain-barrier changes examined with multi-omics and imaging.
The centre also describes work on scalable digital cognitive tools, novel and repurposed drugs, the longitudinal BIOCIS study and regional research collaboration.
Did you know? Dementia is an umbrella term for syndromes that can arise through different combinations of disease processes, so one marker or test rarely tells the whole story.
This guide offers no diagnosis, screening result or treatment recommendation. Qualified healthcare professionals and the official centre govern current research and care. Sources were checked 11 November 2026.
Chapter index
Chapters 1–9: cognitive ageing, dementia, longitudinal cohorts, cognition, perfusion, blood-brain barrier, imaging, biomarkers, multi-omics
Chapters 10–17: digital assessment, artificial intelligence, prediction, therapeutics, clinical utility, care outcomes, regional collaboration, ethics and evidence receipts
Cognition can change with age, health and disease through diverse pathways. Dementia research asks how cognition changes, why different disease processes emerge and which measurements can support useful prevention, detection or care. LKCMedicine says Dementia Research Centre (Singapore) studies Asian dementia through perfusion, blood-brain-barrier change, multi-omics, imaging, digital assessment and therapeutic research. Chapter 1 connects biology, behaviour and lived outcomes without collapsing them into one score.
Define domain, trajectory and functional outcome. A defensible study defines the cognitive domain, population, language and culture, health context, measurement conditions, follow-up interval and meaningful outcome. Researchers combine validated assessments with imaging, biomarkers, clinical observation and longitudinal data because each view has blind spots. Pre-specified analysis, calibration and external validation help distinguish a durable signal from chance or overfitting.
Normal variation and disease can overlap in one measurement. The boundary protects people. Memory lapses, a game score, an image or a biomarker cannot diagnose dementia in isolation; cognitive performance may be affected by education, language, sleep, mood, medicines, sensory ability and other health conditions. Screening and research are not clinical diagnosis. Qualified professionals and current health services govern personal assessment, treatment and support.
Study “Separate ageing and disease” with a fictional longitudinal cohort. Define baseline, repeated measures, possible confounders, missing data, meaningful change and the decision that a result could support. Draw a concept map. Do not score or label a real person, promise early detection, or suggest changing treatment. The classroom task is to examine evidence and communication boundaries.
Did you know? A test can detect change reliably at group level yet remain too uncertain for an individual decision. Chapter 1 shows why dementia research needs repeat measurement, population context, appropriate reference standards and evidence that a tool improves a real pathway, not only a high accuracy number reported from one dataset.
Relevant capabilities include neuroscience, geriatrics, psychology, neurology, imaging, vascular biology, genomics, data science, rehabilitation, nursing, social science and implementation research. Current programmes determine entry routes. School science prepares learners through systems thinking, careful measurement, probability, respectful interviewing and the habit of separating correlation, mechanism, prediction and clinical utility.
Memory, attention, language, executive function and visuospatial skills are related but distinct. Dementia research asks how cognition changes, why different disease processes emerge and which measurements can support useful prevention, detection or care. LKCMedicine says Dementia Research Centre (Singapore) studies Asian dementia through perfusion, blood-brain-barrier change, multi-omics, imaging, digital assessment and therapeutic research. Chapter 2 connects biology, behaviour and lived outcomes without collapsing them into one score.
Use validated tasks and accessible administration. A defensible study defines the cognitive domain, population, language and culture, health context, measurement conditions, follow-up interval and meaningful outcome. Researchers combine validated assessments with imaging, biomarkers, clinical observation and longitudinal data because each view has blind spots. Pre-specified analysis, calibration and external validation help distinguish a durable signal from chance or overfitting.
One total score can hide a specific pattern. The boundary protects people. Memory lapses, a game score, an image or a biomarker cannot diagnose dementia in isolation; cognitive performance may be affected by education, language, sleep, mood, medicines, sensory ability and other health conditions. Screening and research are not clinical diagnosis. Qualified professionals and current health services govern personal assessment, treatment and support.
Study “Measure cognitive domains” with a fictional longitudinal cohort. Define baseline, repeated measures, possible confounders, missing data, meaningful change and the decision that a result could support. Build a domain table. Do not score or label a real person, promise early detection, or suggest changing treatment. The classroom task is to examine evidence and communication boundaries.
Did you know? A test can detect change reliably at group level yet remain too uncertain for an individual decision. Chapter 2 shows why dementia research needs repeat measurement, population context, appropriate reference standards and evidence that a tool improves a real pathway, not only a high accuracy number reported from one dataset.
Relevant capabilities include neuroscience, geriatrics, psychology, neurology, imaging, vascular biology, genomics, data science, rehabilitation, nursing, social science and implementation research. Current programmes determine entry routes. School science prepares learners through systems thinking, careful measurement, probability, respectful interviewing and the habit of separating correlation, mechanism, prediction and clinical utility.
Repeated observations reveal trajectories before and after clinical change. Dementia research asks how cognition changes, why different disease processes emerge and which measurements can support useful prevention, detection or care. LKCMedicine says Dementia Research Centre (Singapore) studies Asian dementia through perfusion, blood-brain-barrier change, multi-omics, imaging, digital assessment and therapeutic research. Chapter 3 connects biology, behaviour and lived outcomes without collapsing them into one score.
Standardise follow-up and record attrition. A defensible study defines the cognitive domain, population, language and culture, health context, measurement conditions, follow-up interval and meaningful outcome. Researchers combine validated assessments with imaging, biomarkers, clinical observation and longitudinal data because each view has blind spots. Pre-specified analysis, calibration and external validation help distinguish a durable signal from chance or overfitting.
People who remain may differ from those lost to follow-up. The boundary protects people. Memory lapses, a game score, an image or a biomarker cannot diagnose dementia in isolation; cognitive performance may be affected by education, language, sleep, mood, medicines, sensory ability and other health conditions. Screening and research are not clinical diagnosis. Qualified professionals and current health services govern personal assessment, treatment and support.
Study “Design longitudinal cohorts” with a fictional longitudinal cohort. Define baseline, repeated measures, possible confounders, missing data, meaningful change and the decision that a result could support. Plan a cohort timeline. Do not score or label a real person, promise early detection, or suggest changing treatment. The classroom task is to examine evidence and communication boundaries.
Did you know? A test can detect change reliably at group level yet remain too uncertain for an individual decision. Chapter 3 shows why dementia research needs repeat measurement, population context, appropriate reference standards and evidence that a tool improves a real pathway, not only a high accuracy number reported from one dataset.
Relevant capabilities include neuroscience, geriatrics, psychology, neurology, imaging, vascular biology, genomics, data science, rehabilitation, nursing, social science and implementation research. Current programmes determine entry routes. School science prepares learners through systems thinking, careful measurement, probability, respectful interviewing and the habit of separating correlation, mechanism, prediction and clinical utility.
Education, language, culture and test familiarity influence measured performance. Dementia research asks how cognition changes, why different disease processes emerge and which measurements can support useful prevention, detection or care. LKCMedicine says Dementia Research Centre (Singapore) studies Asian dementia through perfusion, blood-brain-barrier change, multi-omics, imaging, digital assessment and therapeutic research. Chapter 4 connects biology, behaviour and lived outcomes without collapsing them into one score.
Validate tools in intended populations and languages. A defensible study defines the cognitive domain, population, language and culture, health context, measurement conditions, follow-up interval and meaningful outcome. Researchers combine validated assessments with imaging, biomarkers, clinical observation and longitudinal data because each view has blind spots. Pre-specified analysis, calibration and external validation help distinguish a durable signal from chance or overfitting.
Directly importing a cutoff can misclassify people. The boundary protects people. Memory lapses, a game score, an image or a biomarker cannot diagnose dementia in isolation; cognitive performance may be affected by education, language, sleep, mood, medicines, sensory ability and other health conditions. Screening and research are not clinical diagnosis. Qualified professionals and current health services govern personal assessment, treatment and support.
Study “Respect language and culture” with a fictional longitudinal cohort. Define baseline, repeated measures, possible confounders, missing data, meaningful change and the decision that a result could support. Audit a fictional assessment. Do not score or label a real person, promise early detection, or suggest changing treatment. The classroom task is to examine evidence and communication boundaries.
Did you know? A test can detect change reliably at group level yet remain too uncertain for an individual decision. Chapter 4 shows why dementia research needs repeat measurement, population context, appropriate reference standards and evidence that a tool improves a real pathway, not only a high accuracy number reported from one dataset.
Relevant capabilities include neuroscience, geriatrics, psychology, neurology, imaging, vascular biology, genomics, data science, rehabilitation, nursing, social science and implementation research. Current programmes determine entry routes. School science prepares learners through systems thinking, careful measurement, probability, respectful interviewing and the habit of separating correlation, mechanism, prediction and clinical utility.
| Evidence move | What to record | What it cannot prove alone |
|---|---|---|
| Cognitive change question | Population, variable, unit and time window | A definition does not establish a result |
| Comparison | Baseline, comparator and uncertainty | Association alone does not establish cause |
| Decision boundary | Owner, trigger, review and escalation | A model does not replace accountable judgement |
Perfusion measures how blood reaches brain tissue under defined conditions. Dementia research asks how cognition changes, why different disease processes emerge and which measurements can support useful prevention, detection or care. LKCMedicine says Dementia Research Centre (Singapore) studies Asian dementia through perfusion, blood-brain-barrier change, multi-omics, imaging, digital assessment and therapeutic research. Chapter 5 connects biology, behaviour and lived outcomes without collapsing them into one score.
Connect imaging protocol, vascular context and cognition. A defensible study defines the cognitive domain, population, language and culture, health context, measurement conditions, follow-up interval and meaningful outcome. Researchers combine validated assessments with imaging, biomarkers, clinical observation and longitudinal data because each view has blind spots. Pre-specified analysis, calibration and external validation help distinguish a durable signal from chance or overfitting.
Altered perfusion may be cause, consequence or correlate. The boundary protects people. Memory lapses, a game score, an image or a biomarker cannot diagnose dementia in isolation; cognitive performance may be affected by education, language, sleep, mood, medicines, sensory ability and other health conditions. Screening and research are not clinical diagnosis. Qualified professionals and current health services govern personal assessment, treatment and support.
Study “Study cerebral perfusion” with a fictional longitudinal cohort. Define baseline, repeated measures, possible confounders, missing data, meaningful change and the decision that a result could support. Draw a temporal model. Do not score or label a real person, promise early detection, or suggest changing treatment. The classroom task is to examine evidence and communication boundaries.
Did you know? A test can detect change reliably at group level yet remain too uncertain for an individual decision. Chapter 5 shows why dementia research needs repeat measurement, population context, appropriate reference standards and evidence that a tool improves a real pathway, not only a high accuracy number reported from one dataset.
Relevant capabilities include neuroscience, geriatrics, psychology, neurology, imaging, vascular biology, genomics, data science, rehabilitation, nursing, social science and implementation research. Current programmes determine entry routes. School science prepares learners through systems thinking, careful measurement, probability, respectful interviewing and the habit of separating correlation, mechanism, prediction and clinical utility.
The blood-brain barrier regulates movement between circulation and neural tissue. Dementia research asks how cognition changes, why different disease processes emerge and which measurements can support useful prevention, detection or care. LKCMedicine says Dementia Research Centre (Singapore) studies Asian dementia through perfusion, blood-brain-barrier change, multi-omics, imaging, digital assessment and therapeutic research. Chapter 6 connects biology, behaviour and lived outcomes without collapsing them into one score.
Combine imaging, molecular and cellular evidence. A defensible study defines the cognitive domain, population, language and culture, health context, measurement conditions, follow-up interval and meaningful outcome. Researchers combine validated assessments with imaging, biomarkers, clinical observation and longitudinal data because each view has blind spots. Pre-specified analysis, calibration and external validation help distinguish a durable signal from chance or overfitting.
A marker cannot show every barrier function. The boundary protects people. Memory lapses, a game score, an image or a biomarker cannot diagnose dementia in isolation; cognitive performance may be affected by education, language, sleep, mood, medicines, sensory ability and other health conditions. Screening and research are not clinical diagnosis. Qualified professionals and current health services govern personal assessment, treatment and support.
Study “Examine the blood-brain barrier” with a fictional longitudinal cohort. Define baseline, repeated measures, possible confounders, missing data, meaningful change and the decision that a result could support. Build an evidence matrix. Do not score or label a real person, promise early detection, or suggest changing treatment. The classroom task is to examine evidence and communication boundaries.
Did you know? A test can detect change reliably at group level yet remain too uncertain for an individual decision. Chapter 6 shows why dementia research needs repeat measurement, population context, appropriate reference standards and evidence that a tool improves a real pathway, not only a high accuracy number reported from one dataset.
Relevant capabilities include neuroscience, geriatrics, psychology, neurology, imaging, vascular biology, genomics, data science, rehabilitation, nursing, social science and implementation research. Current programmes determine entry routes. School science prepares learners through systems thinking, careful measurement, probability, respectful interviewing and the habit of separating correlation, mechanism, prediction and clinical utility.
Structural imaging can characterise anatomy and patterns of change. Dementia research asks how cognition changes, why different disease processes emerge and which measurements can support useful prevention, detection or care. LKCMedicine says Dementia Research Centre (Singapore) studies Asian dementia through perfusion, blood-brain-barrier change, multi-omics, imaging, digital assessment and therapeutic research. Chapter 7 connects biology, behaviour and lived outcomes without collapsing them into one score.
Standardise acquisition, segmentation and reference groups. A defensible study defines the cognitive domain, population, language and culture, health context, measurement conditions, follow-up interval and meaningful outcome. Researchers combine validated assessments with imaging, biomarkers, clinical observation and longitudinal data because each view has blind spots. Pre-specified analysis, calibration and external validation help distinguish a durable signal from chance or overfitting.
A structural difference is not a diagnosis by itself. The boundary protects people. Memory lapses, a game score, an image or a biomarker cannot diagnose dementia in isolation; cognitive performance may be affected by education, language, sleep, mood, medicines, sensory ability and other health conditions. Screening and research are not clinical diagnosis. Qualified professionals and current health services govern personal assessment, treatment and support.
Study “Use structural imaging” with a fictional longitudinal cohort. Define baseline, repeated measures, possible confounders, missing data, meaningful change and the decision that a result could support. Compare two imaging claims. Do not score or label a real person, promise early detection, or suggest changing treatment. The classroom task is to examine evidence and communication boundaries.
Did you know? A test can detect change reliably at group level yet remain too uncertain for an individual decision. Chapter 7 shows why dementia research needs repeat measurement, population context, appropriate reference standards and evidence that a tool improves a real pathway, not only a high accuracy number reported from one dataset.
Relevant capabilities include neuroscience, geriatrics, psychology, neurology, imaging, vascular biology, genomics, data science, rehabilitation, nursing, social science and implementation research. Current programmes determine entry routes. School science prepares learners through systems thinking, careful measurement, probability, respectful interviewing and the habit of separating correlation, mechanism, prediction and clinical utility.
CHAPTER 8 OF 17 · Measure converging evidence
8. Use functional and molecular imaging
Functional and molecular methods examine activity, connectivity or selected targets. Dementia research asks how cognition changes, why different disease processes emerge and which measurements can support useful prevention, detection or care. LKCMedicine says Dementia Research Centre (Singapore) studies Asian dementia through perfusion, blood-brain-barrier change, multi-omics, imaging, digital assessment and therapeutic research. Chapter 8 connects biology, behaviour and lived outcomes without collapsing them into one score.
Match signal, timing and biological question. A defensible study defines the cognitive domain, population, language and culture, health context, measurement conditions, follow-up interval and meaningful outcome. Researchers combine validated assessments with imaging, biomarkers, clinical observation and longitudinal data because each view has blind spots. Pre-specified analysis, calibration and external validation help distinguish a durable signal from chance or overfitting.
Tracer or activity patterns can be nonspecific. The boundary protects people. Memory lapses, a game score, an image or a biomarker cannot diagnose dementia in isolation; cognitive performance may be affected by education, language, sleep, mood, medicines, sensory ability and other health conditions. Screening and research are not clinical diagnosis. Qualified professionals and current health services govern personal assessment, treatment and support.
Study “Use functional and molecular imaging” with a fictional longitudinal cohort. Define baseline, repeated measures, possible confounders, missing data, meaningful change and the decision that a result could support. Create a modality guide. Do not score or label a real person, promise early detection, or suggest changing treatment. The classroom task is to examine evidence and communication boundaries.
Did you know? A test can detect change reliably at group level yet remain too uncertain for an individual decision. Chapter 8 shows why dementia research needs repeat measurement, population context, appropriate reference standards and evidence that a tool improves a real pathway, not only a high accuracy number reported from one dataset.
Relevant capabilities include neuroscience, geriatrics, psychology, neurology, imaging, vascular biology, genomics, data science, rehabilitation, nursing, social science and implementation research. Current programmes determine entry routes. School science prepares learners through systems thinking, careful measurement, probability, respectful interviewing and the habit of separating correlation, mechanism, prediction and clinical utility.
| Evidence move | What to record | What it cannot prove alone |
|---|---|---|
| Measurement question | Population, variable, unit and time window | A definition does not establish a result |
| Comparison | Baseline, comparator and uncertainty | Association alone does not establish cause |
| Decision boundary | Owner, trigger, review and escalation | A model does not replace accountable judgement |
Blood or cerebrospinal-fluid measures may reflect selected disease processes. Dementia research asks how cognition changes, why different disease processes emerge and which measurements can support useful prevention, detection or care. LKCMedicine says Dementia Research Centre (Singapore) studies Asian dementia through perfusion, blood-brain-barrier change, multi-omics, imaging, digital assessment and therapeutic research. Chapter 9 connects biology, behaviour and lived outcomes without collapsing them into one score.
Validate collection, assay, threshold and clinical context. A defensible study defines the cognitive domain, population, language and culture, health context, measurement conditions, follow-up interval and meaningful outcome. Researchers combine validated assessments with imaging, biomarkers, clinical observation and longitudinal data because each view has blind spots. Pre-specified analysis, calibration and external validation help distinguish a durable signal from chance or overfitting.
Analytical detection does not guarantee clinical usefulness. The boundary protects people. Memory lapses, a game score, an image or a biomarker cannot diagnose dementia in isolation; cognitive performance may be affected by education, language, sleep, mood, medicines, sensory ability and other health conditions. Screening and research are not clinical diagnosis. Qualified professionals and current health services govern personal assessment, treatment and support.
Study “Develop fluid biomarkers” with a fictional longitudinal cohort. Define baseline, repeated measures, possible confounders, missing data, meaningful change and the decision that a result could support. Build a biomarker ladder. Do not score or label a real person, promise early detection, or suggest changing treatment. The classroom task is to examine evidence and communication boundaries.
Did you know? A test can detect change reliably at group level yet remain too uncertain for an individual decision. Chapter 9 shows why dementia research needs repeat measurement, population context, appropriate reference standards and evidence that a tool improves a real pathway, not only a high accuracy number reported from one dataset.
Relevant capabilities include neuroscience, geriatrics, psychology, neurology, imaging, vascular biology, genomics, data science, rehabilitation, nursing, social science and implementation research. Current programmes determine entry routes. School science prepares learners through systems thinking, careful measurement, probability, respectful interviewing and the habit of separating correlation, mechanism, prediction and clinical utility.
Genomic, transcriptomic, proteomic and metabolomic data reveal different biological layers. Dementia research asks how cognition changes, why different disease processes emerge and which measurements can support useful prevention, detection or care. LKCMedicine says Dementia Research Centre (Singapore) studies Asian dementia through perfusion, blood-brain-barrier change, multi-omics, imaging, digital assessment and therapeutic research. Chapter 10 connects biology, behaviour and lived outcomes without collapsing them into one score.
Control batch, ancestry, tissue and multiple testing. A defensible study defines the cognitive domain, population, language and culture, health context, measurement conditions, follow-up interval and meaningful outcome. Researchers combine validated assessments with imaging, biomarkers, clinical observation and longitudinal data because each view has blind spots. Pre-specified analysis, calibration and external validation help distinguish a durable signal from chance or overfitting.
More features can increase false discovery and overfitting. The boundary protects people. Memory lapses, a game score, an image or a biomarker cannot diagnose dementia in isolation; cognitive performance may be affected by education, language, sleep, mood, medicines, sensory ability and other health conditions. Screening and research are not clinical diagnosis. Qualified professionals and current health services govern personal assessment, treatment and support.
Study “Integrate multi-omics” with a fictional longitudinal cohort. Define baseline, repeated measures, possible confounders, missing data, meaningful change and the decision that a result could support. Audit an omics pipeline. Do not score or label a real person, promise early detection, or suggest changing treatment. The classroom task is to examine evidence and communication boundaries.
Did you know? A test can detect change reliably at group level yet remain too uncertain for an individual decision. Chapter 10 shows why dementia research needs repeat measurement, population context, appropriate reference standards and evidence that a tool improves a real pathway, not only a high accuracy number reported from one dataset.
Relevant capabilities include neuroscience, geriatrics, psychology, neurology, imaging, vascular biology, genomics, data science, rehabilitation, nursing, social science and implementation research. Current programmes determine entry routes. School science prepares learners through systems thinking, careful measurement, probability, respectful interviewing and the habit of separating correlation, mechanism, prediction and clinical utility.
Digital tasks can standardise delivery and capture timing or interaction patterns. Dementia research asks how cognition changes, why different disease processes emerge and which measurements can support useful prevention, detection or care. LKCMedicine says Dementia Research Centre (Singapore) studies Asian dementia through perfusion, blood-brain-barrier change, multi-omics, imaging, digital assessment and therapeutic research. Chapter 11 connects biology, behaviour and lived outcomes without collapsing them into one score.
Test usability, accessibility, reliability and validity. A defensible study defines the cognitive domain, population, language and culture, health context, measurement conditions, follow-up interval and meaningful outcome. Researchers combine validated assessments with imaging, biomarkers, clinical observation and longitudinal data because each view has blind spots. Pre-specified analysis, calibration and external validation help distinguish a durable signal from chance or overfitting.
Gamification does not remove measurement bias. The boundary protects people. Memory lapses, a game score, an image or a biomarker cannot diagnose dementia in isolation; cognitive performance may be affected by education, language, sleep, mood, medicines, sensory ability and other health conditions. Screening and research are not clinical diagnosis. Qualified professionals and current health services govern personal assessment, treatment and support.
Study “Build digital cognitive tools” with a fictional longitudinal cohort. Define baseline, repeated measures, possible confounders, missing data, meaningful change and the decision that a result could support. Prototype a safe fictional task. Do not score or label a real person, promise early detection, or suggest changing treatment. The classroom task is to examine evidence and communication boundaries.
Did you know? A test can detect change reliably at group level yet remain too uncertain for an individual decision. Chapter 11 shows why dementia research needs repeat measurement, population context, appropriate reference standards and evidence that a tool improves a real pathway, not only a high accuracy number reported from one dataset.
Relevant capabilities include neuroscience, geriatrics, psychology, neurology, imaging, vascular biology, genomics, data science, rehabilitation, nursing, social science and implementation research. Current programmes determine entry routes. School science prepares learners through systems thinking, careful measurement, probability, respectful interviewing and the habit of separating correlation, mechanism, prediction and clinical utility.
AI can combine complex patterns for classification or prediction. Dementia research asks how cognition changes, why different disease processes emerge and which measurements can support useful prevention, detection or care. LKCMedicine says Dementia Research Centre (Singapore) studies Asian dementia through perfusion, blood-brain-barrier change, multi-omics, imaging, digital assessment and therapeutic research. Chapter 12 connects biology, behaviour and lived outcomes without collapsing them into one score.
Use external validation, calibration and subgroup analysis. A defensible study defines the cognitive domain, population, language and culture, health context, measurement conditions, follow-up interval and meaningful outcome. Researchers combine validated assessments with imaging, biomarkers, clinical observation and longitudinal data because each view has blind spots. Pre-specified analysis, calibration and external validation help distinguish a durable signal from chance or overfitting.
High accuracy in one dataset is not clinical utility. The boundary protects people. Memory lapses, a game score, an image or a biomarker cannot diagnose dementia in isolation; cognitive performance may be affected by education, language, sleep, mood, medicines, sensory ability and other health conditions. Screening and research are not clinical diagnosis. Qualified professionals and current health services govern personal assessment, treatment and support.
Study “Evaluate artificial intelligence” with a fictional longitudinal cohort. Define baseline, repeated measures, possible confounders, missing data, meaningful change and the decision that a result could support. Create a model card. Do not score or label a real person, promise early detection, or suggest changing treatment. The classroom task is to examine evidence and communication boundaries.
Did you know? A test can detect change reliably at group level yet remain too uncertain for an individual decision. Chapter 12 shows why dementia research needs repeat measurement, population context, appropriate reference standards and evidence that a tool improves a real pathway, not only a high accuracy number reported from one dataset.
Relevant capabilities include neuroscience, geriatrics, psychology, neurology, imaging, vascular biology, genomics, data science, rehabilitation, nursing, social science and implementation research. Current programmes determine entry routes. School science prepares learners through systems thinking, careful measurement, probability, respectful interviewing and the habit of separating correlation, mechanism, prediction and clinical utility.
| Evidence move | What to record | What it cannot prove alone |
|---|---|---|
| Tool evaluation question | Population, variable, unit and time window | A definition does not establish a result |
| Comparison | Baseline, comparator and uncertainty | Association alone does not establish cause |
| Decision boundary | Owner, trigger, review and escalation | A model does not replace accountable judgement |
Vascular, behavioural and social factors can shape cognitive-health trajectories. Dementia research asks how cognition changes, why different disease processes emerge and which measurements can support useful prevention, detection or care. LKCMedicine says Dementia Research Centre (Singapore) studies Asian dementia through perfusion, blood-brain-barrier change, multi-omics, imaging, digital assessment and therapeutic research. Chapter 13 connects biology, behaviour and lived outcomes without collapsing them into one score.
Measure exposures over time and test feasible interventions. A defensible study defines the cognitive domain, population, language and culture, health context, measurement conditions, follow-up interval and meaningful outcome. Researchers combine validated assessments with imaging, biomarkers, clinical observation and longitudinal data because each view has blind spots. Pre-specified analysis, calibration and external validation help distinguish a durable signal from chance or overfitting.
Risk reduction does not guarantee prevention for one person. The boundary protects people. Memory lapses, a game score, an image or a biomarker cannot diagnose dementia in isolation; cognitive performance may be affected by education, language, sleep, mood, medicines, sensory ability and other health conditions. Screening and research are not clinical diagnosis. Qualified professionals and current health services govern personal assessment, treatment and support.
Study “Study prevention and modifiable risk” with a fictional longitudinal cohort. Define baseline, repeated measures, possible confounders, missing data, meaningful change and the decision that a result could support. Draw a prevention pathway. Do not score or label a real person, promise early detection, or suggest changing treatment. The classroom task is to examine evidence and communication boundaries.
Did you know? A test can detect change reliably at group level yet remain too uncertain for an individual decision. Chapter 13 shows why dementia research needs repeat measurement, population context, appropriate reference standards and evidence that a tool improves a real pathway, not only a high accuracy number reported from one dataset.
Relevant capabilities include neuroscience, geriatrics, psychology, neurology, imaging, vascular biology, genomics, data science, rehabilitation, nursing, social science and implementation research. Current programmes determine entry routes. School science prepares learners through systems thinking, careful measurement, probability, respectful interviewing and the habit of separating correlation, mechanism, prediction and clinical utility.
Drug studies need biological rationale, dose, safety and meaningful endpoints. Dementia research asks how cognition changes, why different disease processes emerge and which measurements can support useful prevention, detection or care. LKCMedicine says Dementia Research Centre (Singapore) studies Asian dementia through perfusion, blood-brain-barrier change, multi-omics, imaging, digital assessment and therapeutic research. Chapter 14 connects biology, behaviour and lived outcomes without collapsing them into one score.
Use appropriate comparators and monitor adverse effects. A defensible study defines the cognitive domain, population, language and culture, health context, measurement conditions, follow-up interval and meaningful outcome. Researchers combine validated assessments with imaging, biomarkers, clinical observation and longitudinal data because each view has blind spots. Pre-specified analysis, calibration and external validation help distinguish a durable signal from chance or overfitting.
Biomarker response may not improve daily function. The boundary protects people. Memory lapses, a game score, an image or a biomarker cannot diagnose dementia in isolation; cognitive performance may be affected by education, language, sleep, mood, medicines, sensory ability and other health conditions. Screening and research are not clinical diagnosis. Qualified professionals and current health services govern personal assessment, treatment and support.
Study “Research therapeutics” with a fictional longitudinal cohort. Define baseline, repeated measures, possible confounders, missing data, meaningful change and the decision that a result could support. Build an endpoint ladder. Do not score or label a real person, promise early detection, or suggest changing treatment. The classroom task is to examine evidence and communication boundaries.
Did you know? A test can detect change reliably at group level yet remain too uncertain for an individual decision. Chapter 14 shows why dementia research needs repeat measurement, population context, appropriate reference standards and evidence that a tool improves a real pathway, not only a high accuracy number reported from one dataset.
Relevant capabilities include neuroscience, geriatrics, psychology, neurology, imaging, vascular biology, genomics, data science, rehabilitation, nursing, social science and implementation research. Current programmes determine entry routes. School science prepares learners through systems thinking, careful measurement, probability, respectful interviewing and the habit of separating correlation, mechanism, prediction and clinical utility.
A useful tool must lead to confirmatory assessment, support and accountable follow-up. Dementia research asks how cognition changes, why different disease processes emerge and which measurements can support useful prevention, detection or care. LKCMedicine says Dementia Research Centre (Singapore) studies Asian dementia through perfusion, blood-brain-barrier change, multi-omics, imaging, digital assessment and therapeutic research. Chapter 15 connects biology, behaviour and lived outcomes without collapsing them into one score.
Map referral, capacity, communication and outcomes. A defensible study defines the cognitive domain, population, language and culture, health context, measurement conditions, follow-up interval and meaningful outcome. Researchers combine validated assessments with imaging, biomarkers, clinical observation and longitudinal data because each view has blind spots. Pre-specified analysis, calibration and external validation help distinguish a durable signal from chance or overfitting.
Earlier information can cause harm if no pathway follows. The boundary protects people. Memory lapses, a game score, an image or a biomarker cannot diagnose dementia in isolation; cognitive performance may be affected by education, language, sleep, mood, medicines, sensory ability and other health conditions. Screening and research are not clinical diagnosis. Qualified professionals and current health services govern personal assessment, treatment and support.
Study “Connect detection to care” with a fictional longitudinal cohort. Define baseline, repeated measures, possible confounders, missing data, meaningful change and the decision that a result could support. Design a care cascade. Do not score or label a real person, promise early detection, or suggest changing treatment. The classroom task is to examine evidence and communication boundaries.
Did you know? A test can detect change reliably at group level yet remain too uncertain for an individual decision. Chapter 15 shows why dementia research needs repeat measurement, population context, appropriate reference standards and evidence that a tool improves a real pathway, not only a high accuracy number reported from one dataset.
Relevant capabilities include neuroscience, geriatrics, psychology, neurology, imaging, vascular biology, genomics, data science, rehabilitation, nursing, social science and implementation research. Current programmes determine entry routes. School science prepares learners through systems thinking, careful measurement, probability, respectful interviewing and the habit of separating correlation, mechanism, prediction and clinical utility.
Regional collaboration can improve representation and harmonise methods across Southeast Asia. Dementia research asks how cognition changes, why different disease processes emerge and which measurements can support useful prevention, detection or care. LKCMedicine says Dementia Research Centre (Singapore) studies Asian dementia through perfusion, blood-brain-barrier change, multi-omics, imaging, digital assessment and therapeutic research. Chapter 16 connects biology, behaviour and lived outcomes without collapsing them into one score.
Align definitions while preserving local context. A defensible study defines the cognitive domain, population, language and culture, health context, measurement conditions, follow-up interval and meaningful outcome. Researchers combine validated assessments with imaging, biomarkers, clinical observation and longitudinal data because each view has blind spots. Pre-specified analysis, calibration and external validation help distinguish a durable signal from chance or overfitting.
A pooled average can hide important site differences. The boundary protects people. Memory lapses, a game score, an image or a biomarker cannot diagnose dementia in isolation; cognitive performance may be affected by education, language, sleep, mood, medicines, sensory ability and other health conditions. Screening and research are not clinical diagnosis. Qualified professionals and current health services govern personal assessment, treatment and support.
Study “Strengthen regional evidence” with a fictional longitudinal cohort. Define baseline, repeated measures, possible confounders, missing data, meaningful change and the decision that a result could support. Plan a harmonisation table. Do not score or label a real person, promise early detection, or suggest changing treatment. The classroom task is to examine evidence and communication boundaries.
Did you know? A test can detect change reliably at group level yet remain too uncertain for an individual decision. Chapter 16 shows why dementia research needs repeat measurement, population context, appropriate reference standards and evidence that a tool improves a real pathway, not only a high accuracy number reported from one dataset.
Relevant capabilities include neuroscience, geriatrics, psychology, neurology, imaging, vascular biology, genomics, data science, rehabilitation, nursing, social science and implementation research. Current programmes determine entry routes. School science prepares learners through systems thinking, careful measurement, probability, respectful interviewing and the habit of separating correlation, mechanism, prediction and clinical utility.
The receipt links population, measure, biology, model, outcome and decision boundary. Dementia research asks how cognition changes, why different disease processes emerge and which measurements can support useful prevention, detection or care. LKCMedicine says Dementia Research Centre (Singapore) studies Asian dementia through perfusion, blood-brain-barrier change, multi-omics, imaging, digital assessment and therapeutic research. Chapter 17 connects biology, behaviour and lived outcomes without collapsing them into one score.
Archive protocols, versions, missingness, uncertainty and limits. A defensible study defines the cognitive domain, population, language and culture, health context, measurement conditions, follow-up interval and meaningful outcome. Researchers combine validated assessments with imaging, biomarkers, clinical observation and longitudinal data because each view has blind spots. Pre-specified analysis, calibration and external validation help distinguish a durable signal from chance or overfitting.
Documentation cannot turn research into a personal diagnosis. The boundary protects people. Memory lapses, a game score, an image or a biomarker cannot diagnose dementia in isolation; cognitive performance may be affected by education, language, sleep, mood, medicines, sensory ability and other health conditions. Screening and research are not clinical diagnosis. Qualified professionals and current health services govern personal assessment, treatment and support.
Study “Keep a dementia-research receipt” with a fictional longitudinal cohort. Define baseline, repeated measures, possible confounders, missing data, meaningful change and the decision that a result could support. Assemble a one-page receipt. Do not score or label a real person, promise early detection, or suggest changing treatment. The classroom task is to examine evidence and communication boundaries.
Did you know? A test can detect change reliably at group level yet remain too uncertain for an individual decision. Chapter 17 shows why dementia research needs repeat measurement, population context, appropriate reference standards and evidence that a tool improves a real pathway, not only a high accuracy number reported from one dataset.
Relevant capabilities include neuroscience, geriatrics, psychology, neurology, imaging, vascular biology, genomics, data science, rehabilitation, nursing, social science and implementation research. Current programmes determine entry routes. School science prepares learners through systems thinking, careful measurement, probability, respectful interviewing and the habit of separating correlation, mechanism, prediction and clinical utility.
| Evidence move | What to record | What it cannot prove alone |
|---|---|---|
| Dementia research question | Population, variable, unit and time window | A definition does not establish a result |
| Comparison | Baseline, comparator and uncertainty | Association alone does not establish cause |
| Decision boundary | Owner, trigger, review and escalation | A model does not replace accountable judgement |
Contents · Previous chapter · Continue in the Science Learning Hub
Primary sources and verification date
Source status for Why Science? | Dementia Research Centre (Singapore) was checked 11 November 2026. Research priorities, programmes, recruitment, facilities and participation routes can change; use the linked official owner for the current edition.
- Dementia Research Centre (Singapore) — Official current focus on Asian dementia, biomarkers, imaging, digital tools and therapeutics.
- LKCMedicine research centres — Official current centre listing and research context.
- LKCMedicine 15 years of discovery and research impact — Official institutional account of the centre’s dementia-research direction.
- LKCMedicine Population and Global Health — Official population-health programme context for prevention and implementation research.
Continue the evidence journey
Dementia research becomes useful when biological, cognitive and lived outcomes converge, tools are validated in the people they may serve, and every result remains inside a responsible care pathway. Families considering science tuition or a science tutor can use the same evidence questions: what will the learner practise, what feedback will they receive, and how will progress be shown?