Role of Ethics in Developing AI-Based Applications in Medicine: Insights From Expert Interviews and Discussion of Implications

Effects of the "AI-TA" Mobile App With Intelligent Design on Psychological and Related Symptoms of Young Survivors of Breast Cancer: Randomized Controlled Trial

BACKGROUND

Young women often face substantial psychological challenges in the initial years following cancer diagnosis, leading to a comparatively lower quality of life than older survivors. While mobile apps have emerged as potential interventions, their effectiveness remains inconclusive due to the diversity in intervention types and variation in follow-up periods. Furthermore, there is a particular dearth of evidence regarding the efficacy of these apps' intelligent features in addressing psychological distress with these apps.

OBJECTIVE

This study aims to evaluate the effectiveness of a mobile app with intelligent design called "AI-TA" on cancer-related psychological health and ongoing symptoms with a randomized controlled design.

METHODS

Women aged 18 to 45 years diagnosed with breast cancer were randomly assigned to the intervention or control group. The intervention was AI-TA, which included 2-way web-based follow-up every 2 weeks. Both intention-to-treat (ITT) and per-protocol (PP) analyses employed repeated measurement analysis of variance. The participants' background features, primary outcomes (psychological distress and frequency, self-efficacy, and social support), and secondary outcomes (quality of life) were measured using multiple instruments at 3 time points (baseline, 1-month intervention, and 3-month intervention).

RESULTS

A total of 124 participants were randomly allocated to the control group (n=62, 50%) or intervention group (n=62, 50%). In total, 92.7% (115/124) of the participants completed the intervention. Significant improvements in psychological symptoms (Memorial Symptom Assessment Scale-Short Form) were observed in the ITT group from baseline to 1-month intervention relative to the control group (ITT vs control: 1.17 vs 1.23; P<.001), which persisted at 3-month follow-up (ITT vs control: 0.68 vs 0.91; P<.001). Both the ITT and PP groups exhibited greater improvements in self-efficacy (Cancer Behavior Inventory-Brief Version) than the control group at 1-month (ITT vs PP vs control: 82.83 vs 77.12 vs 65.35; P<.001) and 3-month intervention (ITT vs PP vs control: 92.83 vs 89.30 vs 85.65; P<.001). However, the change in social support (Social Support Rating Scale) did not increase significantly until 3-month intervention (ITT vs control: 50.09 vs 45.10; P=.002) (PP vs control: 49.78 vs 45.10; P<.001). All groups also experienced beneficial effects on quality of life (Functional Assessment of Cancer Therapy-Breast), which persisted at 3-month follow-up (P<.001).

CONCLUSIONS

The intelligent mobile app AI-TA incorporating intelligent design shows promise for reducing psychological and cancer-related symptoms among young survivors of breast cancer.

TRIAL REGISTRATION

Chinese Clinical Trial Registry ChiCTR2200058823; https://www.chictr.org.cn/showproj.html?proj=151195.

Machine Learning Methods Using Artificial Intelligence Deployed on Electronic Health Record Data for Identification and Referral of At-Risk Patients From Primary Care Physicians to Eye Care Specialists: Retrospective, Case-Controlled Study

BACKGROUND

Identification and referral of at-risk patients from primary care practitioners (PCPs) to eye care professionals remain a challenge. Approximately 1.9 million Americans suffer from vision loss as a result of undiagnosed or untreated ophthalmic conditions. In ophthalmology, artificial intelligence (AI) is used to predict glaucoma progression, recognize diabetic retinopathy (DR), and classify ocular tumors; however, AI has not yet been used to triage primary care patients for ophthalmology referral.

OBJECTIVE

This study aimed to build and compare machine learning (ML) methods, applicable to electronic health records (EHRs) of PCPs, capable of triaging patients for referral to eye care specialists.

METHODS

Accessing the Optum deidentified EHR data set, 743,039 patients with 5 leading vision conditions (age-related macular degeneration [AMD], visually significant cataract, DR, glaucoma, or ocular surface disease [OSD]) were exact-matched on age and gender to 743,039 controls without eye conditions. Between 142 and 182 non-ophthalmic parameters per patient were input into 5 ML methods: generalized linear model, L1-regularized logistic regression, random forest, Extreme Gradient Boosting (XGBoost), and J48 decision tree. Model performance was compared for each pathology to select the most predictive algorithm. The area under the curve (AUC) was assessed for all algorithms for each outcome.

RESULTS

XGBoost demonstrated the best performance, showing, respectively, a prediction accuracy and an AUC of 78.6% (95% CI 78.3%-78.9%) and 0.878 for visually significant cataract, 77.4% (95% CI 76.7%-78.1%) and 0.858 for exudative AMD, 79.2% (95% CI 78.8%-79.6%) and 0.879 for nonexudative AMD, 72.2% (95% CI 69.9%-74.5%) and 0.803 for OSD requiring medication, 70.8% (95% CI 70.5%-71.1%) and 0.785 for glaucoma, 85.0% (95% CI 84.2%-85.8%) and 0.924 for type 1 nonproliferative diabetic retinopathy (NPDR), 82.2% (95% CI 80.4%-84.0%) and 0.911 for type 1 proliferative diabetic retinopathy (PDR), 81.3% (95% CI 81.0%-81.6%) and 0.891 for type 2 NPDR, and 82.1% (95% CI 81.3%-82.9%) and 0.900 for type 2 PDR.

CONCLUSIONS

The 5 ML methods deployed were able to successfully identify patients with elevated odds ratios (ORs), thus capable of patient triage, for ocular pathology ranging from 2.4 (95% CI 2.4-2.5) for glaucoma to 5.7 (95% CI 5.0-6.4) for type 1 NPDR, with an average OR of 3.9. The application of these models could enable PCPs to better identify and triage patients at risk for treatable ophthalmic pathology. Early identification of patients with unrecognized sight-threatening conditions may lead to earlier treatment and a reduced economic burden. More importantly, such triage may improve patients' lives.