Key result
A deep neural network model using basic blood test data alone predicted cognitive impairment with 81.8% sensitivity and 91.3% specificity, performing similarly to models incorporating time-resolved near-infrared spectroscopy.
Why the study?
It was unclear whether prediction accuracy of cognitive function differs between NIRS and blood test data, between linear and non-linear models, or when combining both modalities.
Does a deep neural network using time-resolved near-infrared spectroscopy and basic blood test data accurately predict cognitive impairment in older adults?
Cross-Sectional (n=250)
No
Does a deep neural network using time-resolved near-infrared spectroscopy and basic blood test data accurately predict cognitive impairment in older adults?
Absolute Event Rate: 26.3% vs 29.1%
p-value: p=<0.001
A deep neural network using basic blood test data can accurately predict cognitive impairment, offering a cost-effective and scalable alternative to subjective screening tools like the MMSE.
ML prediction of cognition from NIRS or blood tests is hypothesis-generating; prospective validation required before any clinical role.
We have demonstrated that machine learning allows us to predict cognitive function in aged people using near-infrared spectroscopy (NIRS) data or basic blood test data. However, the following points are not yet clear: first, whether there are differences in prediction accuracy between NIRS and blood test data; second, whether there are differences in prediction accuracy for cognitive function in linear models and non-linear models; and third, whether there are changes in prediction accuracy when both NIRS and blood test data are added to the input layer. We used a linear regression model (LR) for the linear model and random forest (RF) and deep neural network (DNN) for the non-linear model. We studied 250 participants (mean age = 73.3 ± 12.6 years) and assessed cognitive function using the Mini Mental State Examination (MMSE) (mean MMSE scores = 22.9 ± 6.1). We used time-resolved NIRS (TNIRS) to measure absolute concentrations of hemoglobin and optical pathlength at rest in the bilateral prefrontal cortices. A basic blood test was performed on the same day. We compared predicted MMSE scores and grand truth MMSE scores; prediction accuracies were evaluated using mean absolute error (MAE) and mean absolute percentage error (MAPE). We found that (1) the DNN-based prediction using TNIRS data exhibited lower MAE and MAPE compared with those using blood test data, (2) the difference in MAPE between TNIRS and blood test data was only 0.3%, (3) adding TNIRS data to the blood test data of the input layer only improved MAPE by 1.0% compared to the use of blood test data alone, whereas the use of the blood test data alone exhibited the prediction accuracy with 81.8% sensitivity and 91.3% specificity (N = 202, repeated five-fold cross validation). Given these findings and the benefits of using blood test data (low cost and large-scale screening possible), we concluded that the DNN model using blood test data is still the most suitable for mass screening.
No takes yet. Share an insight, caveat, or question.
Oyama et al. (2022) conducted a cross-sectional in Cognitive impairment (n=250). Deep Neural Network (DNN) model using basic blood test data vs. Linear Regression model was evaluated on Mean Absolute Percentage Error (MAPE) for predicting MMSE scores (p=<0.001). A deep neural network model using basic blood test data alone predicted cognitive impairment with 81.8% sensitivity and 91.3% specificity, performing similarly to models incorporating time-resolved near-infrared spectroscopy.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: