Using artificial intelligence in medical school admissions screening to decrease inter- and intra-observer variability.

Objectives: Inter- and intra-observer variability is a concern for medical school admissions. Artificial intelligence (AI) may present an opportunity to apply a fair standard to all applicants systematically and yet maintain sensitivity to nuances that have been a part of traditional screening methods. Material and Methods: Data from 5 years of medical school applications were retrospectively accrued and analyzed. The applicants (m = 22 258 applicants) were split 60%-20%-20% into a training set (m = 13 354), validation set (m = 4452), and test set (m = 4452). An AI model was trained and evaluated with the ground truth being whether a given applicant was invited for an interview. In addition, a "real-world" evaluation was conducted simultaneously within an admissions cycle to observe how it would perform if utilized. Results: The algorithm had an accuracy of 95% on the training set, 88% on the validation set, and 88% on the test set. The area under the curve of the test set was 0.93. The SHapely Additive exPlanations (SHAP) values demonstrated that the model utilizes features in a concordant manner with current admissions rubrics. By using a combined human and AI evaluation process, the accuracy of the process was demonstrated to be 96% on the "real-world" evaluation with a negative predictive value of 0.97. Discussion and Conclusion: These results demonstrate the feasibility of an AI approach applied to medical school admissions screening decision-making. Model explainability and supplemental analyses help ensure that the model makes decisions as intended.

Targeting Cognitive Control Deficits With Neuroplasticity-Based Computerized Cognitive Remediation in Patients With Geriatric Major Depression: A Randomized, Double-Blind, Controlled Trial.

Late life major depression (LLD) is often accompanied by cognitive deficits. When patients have specific deficits in cognitive control functions (CCD), they are not only distressing and debilitating, they often predict poor clinical outcomes such as reduced response to SSRI/SNRI antidepressants, increased disability, suicide and all-cause mortality. We recently reported that in an open label trial, our treatment designed to target these specific CCD with neuroplasticity-based computerized cognitive remediation (nCCR) improved depression and CCD in patients who failed to remit with conventional antidepressant treatment. This study tested the hypothesis that in patients with LLD who have failed at least one trial of an SSRI/SNRI antidepressant at an adequate dose for at least 8 weeks, nCCR will improve both depressive symptoms and the CCD associated with poor antidepressant response (i.e. semantic strategy, inhibition of prepotent responses) more than an active control group. Participants were randomized (1:1) to receive either 30 hours/ 4 weeks of neuroplasticity based computerized cognitive remediation (nCCR) designed to target CCD, or the active control condition matched for duration, engagement, reward, computer presentation, and contact with study staff. All participants and raters were blinded. Mixed effects model analysis the time effect (week) (F(1,71.22)=25.2, p<0.0001) and treatment group X time interaction (F(1,61.8)=11.37, p=.002) reached significance indicating that the slope of decline in MADRS was steeper in the nCCR-GD group. Further, the nCCR group improved their semantic clustering strategy(t(28)=9.5; p=.006), as well as performance on the Stroop interference condition, and cognitive flexibility (Trails B). Further, results transferred to memory performance, which was not a function trained by nCCR. clinicaltrials.gov.

Executive Dysfunction Predicts Treatment Response to Neuroplasticity-Based Computerized Cognitive Remediation (nCCR-GD) in Elderly Patients with Major Depression.

OBJECTIVES: Executive dysfunction (ED) is a predictor of poor treatment response of late-life depression to pharmacotherapy. In response to the consistency of these findings, we designed neuroplasticity-based computerized cognitive remediation (nCCR-GD) intervention to target and improve ED in patients who failed to remit with antidepressant treatment. This study tests the hypothesis that ED at baseline will predict favorable treatment response to nCCR-GD. METHODS: 11 elderly patients with treatment-resistant major depression were treated with a 30-hour, 4-week, unblinded, nCCR-GD treatment trial. Neuropsychological performance was assessed at baseline and after treatment ceased. RESULTS: ED at baseline was associated with greater reduction in Montgomery-Asberg Depression Rating Scale score over the 4-week treatment ß = -0.74, F(2,8) = 10.85, p = 0.009, R(2) = 0.55. CONCLUSIONS: ED predicts favorable treatment response to nCCR-GD in older adults suffering from major depression resistant to antidepressants. This finding is opposed to studies testing pharmacotherapy where ED predicts poorer treatment response.

Neuroplasticity-based computerized cognitive remediation for treatment-resistant geriatric depression.

Executive dysfunction (ED) in geriatric depression (GD) is common, predicts poor clinical outcomes and often persists despite remission of symptoms. Here we develop a neuroplasticity-based computerized cognitive remediation-geriatric depression treatment (nCCR-GD) to target ED in GD. Our assumption is that remediation of these deficits may modulate the underlying brain network abnormalities shared by ED and depression. We compare nCCR-GD to a gold-standard treatment (escitalopram: 20 mg per 12 weeks) in 11 treatment-resistant older adults with major depression; and 33 matched historical controls. We find that 91% of participants complete nCCR-GD. nCCR-GD is equally as effective at reducing depressive symptoms as escitalopram but does so in 4 weeks instead of 12. In addition, nCCR-GD improves measures of executive function more than the escitalopram. We conclude that nCCR-GD may be equally effective as escitalopram in treating GD. In addition, nCCR-GD participants showed greater improvement in executive functions than historical controls treated with escitalopram.