Association Between Dementia and Optical Coherence Tomography Scan Quality

Reuben Jyong Kiat Foo, Damon Wing Kee Wong, Nur Fidyana Binte Abdul Gani, Bingyao Tan, Munirah Ismail, Gerhard Garhöfer, Laetitia Hinterhuber, Narayanaswamy Venketasubramanian, Christopher Chen, Leopold Schmetterer, Jacqueline ChuaAging 2025 · DOI

Abstract

It is generally assumed that dementia affects the quality of optical coherence tomography (OCT) scans. However, the magnitude of this effect and its independence from other factors require further clarification. In this cross-sectional study, our aim was to evaluate the association between cognitive impairment and OCT scan quality, adjusting for key confounders, in a multiethnic cohort. 541 participants aged 50 years or older were recruited from memory clinics and the community at the National University Hospital and St. Luke's Hospital, Singapore. They were then stratified into three groups: no cognitive impairment (NCI, n=112), cognitive impairment without dementia (CIND, n=235), and dementia (n=194); OCT scan quality was subsequently assessed based on the presence and severity of artifacts. We found that dementia patients were nearly three times more likely to produce poor-quality OCT scans compared to NCI participants (adjusted odds ratio [OR]=2.90; 95% CI, 1.24-6.80). Lower cognitive scores, including Mini-Mental State Examination (MMSE) (OR=0.92; 95% CI, 0.88-0.96), Montreal Cognitive Assessment (MoCA) (OR=0.90; 95% CI, 0.86-0.94), and higher Clinical Dementia Rating (CDR) scores (OR=2.11; 95% CI, 1.43-3.10), were also independently associated with poor scan quality. In conclusion, cognitive impairment, particularly dementia, substantially increases the likelihood of poor-quality OCT scans, even after accounting for key demographic and clinical factors. Hence, strategies tailored to improve imaging in this population are essential for enhancing diagnostic accuracy and patient care.

Introduction: The Challenge of Imaging the Aging Brain

Optical Coherence Tomography (OCT) has revolutionized ophthalmology by providing non-invasive, high-resolution cross-sectional images of the retina. However, the diagnostic utility of OCT is heavily dependent on patient cooperation. Subjects must maintain steady visual fixation and follow complex instructions—tasks that become increasingly difficult for individuals with cognitive impairment. While clinicians have long observed lower scan quality in patients with dementia, this study provides a rigorous quantification of that effect, independent of other age-related factors.

Methods and Artifact Classification

The study analyzed 541 participants (aged 50+) categorized into three groups: No Cognitive Impairment (NCI), Cognitive Impairment No Dementia (CIND), and Dementia. Using a Zeiss Cirrus HD-5000, researchers captured structural OCT scans and graded them for quality based on the presence of specific artifacts.

Figure 5: Classification of OCT scan artifacts, ranging from good scans (A) to common issues like motion (B), shadows (C), and low signal (H).

Scans were evaluated for several specific artifacts that compromise data integrity. These included motion displacements, shadows from vitreous floaters or blinking, and decentration (off-center scans). A single, blinded grader evaluated the scans to ensure consistency in quality assessment.

Key Results: Quantifying the Impact of Cognitive Decline

The analysis revealed a stark correlation between cognitive health and imaging success. Patients with dementia were nearly three times more likely to produce poor-quality scans than those in the NCI group (adjusted Odds Ratio [OR] = 2.90). For those specifically diagnosed with Alzheimer’s Disease, the odds increased to 3.16.

Figure 6: A detailed view of motion artifacts (arrows), which were identified as the primary cause of poor scan quality, occurring in 67% of failed scans.

Beyond categorical diagnosis, continuous cognitive scores (MMSE and MoCA) were reliable predictors of scan quality. Interestingly, the study also found that female participants were twice as likely to have poor-quality scans (OR = 2.03), suggesting a need for further research into gender-specific communication or biological factors in clinical imaging.

Clinical Implications and Future Directions

The high prevalence of motion artifacts (67% of poor scans) suggests that the primary barrier is behavioral rather than optical. This has significant implications for the development of AI-based diagnostic tools, which are often trained on "perfect" scans and may fail when applied to the lower-quality images typical of dementia patients.

To improve diagnostic accuracy in aging populations, the authors recommend:

  • Tailored Protocols: Scheduling scans during periods of high alertness (avoiding "sundowning") and using shorter acquisition times.
  • Hardware Innovations: Utilizing faster swept-source lasers to minimize the window for motion.
  • Software Solutions: Developing robust artifact-correction algorithms specifically designed for neurodegenerative cohorts.

Ultimately, a sudden decline in a patient's ability to cooperate with OCT imaging may serve as an early clinical indicator of cognitive change, warranting further neuropsychological evaluation.

Summary updated

Abstract

It is generally assumed that dementia affects the quality of optical coherence tomography (OCT) scans. However, the magnitude of this effect and its independence from other factors require further clarification. In this cross-sectional study, our aim was to evaluate the association between cognitive impairment and OCT scan quality, adjusting for key confounders, in a multiethnic cohort. 541 participants aged 50 years or older were recruited from memory clinics and the community at the National University Hospital and St. Luke's Hospital, Singapore. They were then stratified into three groups: no cognitive impairment (NCI, n=112), cognitive impairment without dementia (CIND, n=235), and dementia (n=194); OCT scan quality was subsequently assessed based on the presence and severity of artifacts. We found that dementia patients were nearly three times more likely to produce poor-quality OCT scans compared to NCI participants (adjusted odds ratio [OR]=2.90; 95% CI, 1.24-6.80). Lower cognitive scores, including Mini-Mental State Examination (MMSE) (OR=0.92; 95% CI, 0.88-0.96), Montreal Cognitive Assessment (MoCA) (OR=0.90; 95% CI, 0.86-0.94), and higher Clinical Dementia Rating (CDR) scores (OR=2.11; 95% CI, 1.43-3.10), were also independently associated with poor scan quality. In conclusion, cognitive impairment, particularly dementia, substantially increases the likelihood of poor-quality OCT scans, even after accounting for key demographic and clinical factors. Hence, strategies tailored to improve imaging in this population are essential for enhancing diagnostic accuracy and patient care.
BibTeX
@inproceedings{foo2025association,
  author = {Reuben Jyong Kiat Foo and Damon Wing Kee Wong and Nur Fidyana Binte Abdul Gani and Bingyao Tan and Munirah Ismail and Gerhard Garhöfer and Laetitia Hinterhuber and Narayanaswamy Venketasubramanian and Christopher Chen and Leopold Schmetterer and Jacqueline Chua},
  title = {{Association Between Dementia and Optical Coherence Tomography Scan Quality}},
  booktitle = {Aging},
  year = {2025},
  doi = {10.14336/ad.2024.1744},
}
Figure