Scoring algorithms for a computer-based cognitive screening tool: An illustrative example of overfitting machine learning approaches and the impact on estimates of classification accuracy.

Abstract

Computerized cognitive screening tools, such as the self-administered Computerized Assessment of Memory Cognitive Impairment (CAMCI), require little training and ensure standardized administration and could be an ideal test for primary care settings. We conducted a secondary analysis of a data set including 887 older adults (M age ϭ 72.7 years, SD ϭ 7.1 years; 32.1% male; M years education ϭ 13.4, SD ϭ 2.7 years) with CAMCI scores and independent diagnoses of mild cognitive impairment (MCI). A study by the CAMCI developers used a portion of this data set with a machine learning decision tree model and suggested that the CAMCI had high classification accuracy for MCI (sensitivity ϭ 0.86, specificity ϭ 0.94). We found similar support for accuracy (sensitivity ϭ 0.94, specificity ϭ 0.94) by overfitting a decision tree model, but we found evidence of lower accuracy in a cross-validation sample (sensitivity ϭ 0.62, specificity ϭ 0.66). A logistic regression model, however, discriminated modestly in both training (sensitivity ϭ 0.72, specificity ϭ 0.80) and cross-validation data sets (sensitivity ϭ 0.69, specificity ϭ 0.74). Evidence for strong accuracy when overfitting a decision tree model and substantially reduced accuracy in cross-validation samples was replicated across 500 bootstrapped samples. In contrast, the evidence for accuracy of the logistic regression model was similar in the training and cross-validation samples. The logistic regression model produced accuracy estimates consistent with other published CAMCI studies, suggesting evidence for classification accuracy of the CAMCI for MCI is likely modest. This case study illustrates the general need for cross-validation and careful evaluation of the generalizability of machine learning models. Public Significance StatementUsing an archival database, a decision tree machine learning method demonstrated overfitting and had substantially reduced evidence for classification accuracy measured in cross-validation, but the statistical method results in similar evidence of classification in training and cross-validation samples. The evidence for classification accuracy of the Computerized Assessment of Mild Cognitive Impairment for cognitive impairment is modest, and this has clinical implications.

Digital Cognitive Biomarker for Mild Cognitive Impairments and Dementia: A Systematic Review

“…After an initial screening of the titles and abstracts, 87 studies were retrieved for full-text reviewing. It is worth noting that, among the 87 studies, 1 study by Ursenbach and colleagues [ 34 ] was excluded because it was analyzed on the same dataset as a study by Saxton and colleagues [ 35 ] and did not report any novel findings related to our research question. Moreover, an earlier study by Müller and colleagues [ 36 ] was a preliminary study of a later study [ 37 ]; therefore, only the results of the later study were included for analysis.…”

Application of Artificial Intelligence in Community-Based Primary Health Care: Systematic Scoping Review and Critical Appraisal

“…The majority of the 90 included studies (68/90, 76%) investigated the use of AI in relation to a specific medical condition. Conditions studied were vascular diseases including hypertension, hypercholesteremia, peripheral arterial disease, and congestive heart failure (10/90, 11%) [ 40 - 49 ]; infectious diseases including influenza, herpes zoster, tuberculosis, urinary tract infections, and subcutaneous infections (8/90, 9%) [ 50 - 57 ]; type 2 diabetes (5/90, 6%) [ 58 - 62 ]; respiratory disorders including chronic obstructive pulmonary disease and asthma (6/90, 8%) [ 63 - 69 ]; orthopedic disorders including rheumatoid arthritis, gout, and lower back pain (5/90, 5%) [ 36 , 39 , 70 - 72 ]; neurological disorders including stroke, Parkinson disease, Alzheimer disease [ 73 - 75 ], and cognitive impairments (6/90, 5%) [ 76 , 77 ]; cancer including colorectal cancer, and head and neck cancer (4/90, 4%) [ 78 - 81 ]; psychological disorders including depression and schizophrenia (3/90, 3%) [ 82 - 84 ]; diabetic retinopathy (3/90, 3%) [ 85 - 87 ]; suicidal ideations (2/90, 2%) [ 88 , 89 ]; tropical diseases including malaria (2/90, 2%) [ 90 , 91 ]; renal disorders (2/90, 2%) [ 92 , 93 ]; autism spectrum disorder (2/90, 2%) [ 94 , 95 ]; venous disorders including deep vein thrombosis and venous ulcers (2/90, 2%) [ 96 , 97 ]; and other health conditions (8/90, 8%) [ 98 - 105 ].…”

Proceedings of the International Conference of Applied Business and Management (ICABM2020)

“…Some indications of overfitting in a model are high training accuracy and the training accuracy is far higher than the testing accuracy. While indications of underfitting are low training accuracy and the training accuracy is approaching the testing accuracy (Ursenbach et al, 2019). To prevent overfitting or underfitting and get a moderate hyperplane, we assess the shape of the hyperplane by plotting the model.…”