Development of a Mobile Terminal-Based Intelligent Detection System for Multiple Anterior Segment Diseases of the Eye

• Smartphone-based on-device artificial intelligence system for anterior segment eye disease screening (Device): A structured-pruned one-stage object-detection model deployed as a standalone Android application, performing all image inference on-device without internet connectivity, designed to detect 16 anterior segment eye diseases from smartphone-captured images.

Primary outcomes

• Case-level diagnostic accuracy of the AI system compared with board-certified ophthalmologists [Day 1]: Case-level accuracy is defined as the proportion of images with fully correct diagnostic labels concordant with the reference standard. The AI system and board-certified ophthalmologists stratified by clinical seniority (junior: fewer than 5 years of independent practice; intermediate: 5 to 15 years; senior: more than 15 years) independently evaluate the same standardized set of smartphone-captured anterior segment images, sampled to ensure balanced representation of all 16 disease categories and normal eyes. The reference standard for each image is established by a senior ophthalmologist with more than 30 years of clinical experience who does not participate in the benchmarking exercise. Clinicians are masked to the AI system output and to each other's assessments throughout.
• Diagnostic accuracy of the AI system when operated by non-medical users [Day 1]: Diagnostic accuracy is defined as the proportion of images correctly classified by non-medical users operating the AI system independently. Non-medical users, including patients and their family members attending the outpatient clinic of Zhongshan Ophthalmic Center, are instructed to install the AI application on their own smartphones, follow the in-app guidelines, and capture anterior segment images of an accompanying person to receive a screening result. All inference is performed on-device without internet connectivity. AI-generated diagnostic outputs are compared with reference-standard diagnoses obtained from subsequent slit-lamp examinations performed by ophthalmologists at the same clinic.
• Sensitivity of the on-device AI system in population-level community screening [Day 1]: Sensitivity is defined as the proportion of participants with a given anterior segment disease who are correctly identified as positive by the on-device AI system (true positives divided by the sum of true positives and false negatives), calculated separately for each target disease category. Village staff without medical training use the on-device AI system to screen consecutively enrolled local residents. The reference standard is established through slit-lamp examinations performed by ophthalmologists.
• Incremental cost-effectiveness ratio of on-device versus cloud-based screening [Day 1]: The incremental cost-effectiveness ratio (ICER) is defined as the difference in lifetime costs between on-device and cloud-based screening divided by the difference in quality-adjusted life-years (QALYs) between the two strategies, estimated from a pre-specified decision-analytic model comprising a decision tree with a downstream Markov state-transition structure applied to a simulated cohort of 100,000 individuals.
• Per-inference electricity consumption of on-device versus cloud-based deployment [Day 1]: Electricity consumption in joules per inference cycle is measured over 10,000 inference cycles on both the on-device smartphone and a standardized cloud-server configuration (Intel Xeon Gold 6248 CPU, NVIDIA Tesla V100 GPU) using the Experiment Impact Tracker toolkit. Cloud-server measurements are averaged across low-traffic and high-traffic server conditions. Results are reported separately for the on-device and cloud-based deployment modalities and compared as a ratio.

Secondary outcomes

• Total image evaluation time of the AI system compared with board-certified ophthalmologists [Day 1]: Total time in seconds required to complete independent evaluation of all images in the standardized benchmarking set, recorded separately for the AI system and for the board-certified ophthalmologists. Reported as mean with standard deviation for both the AI system and the clinician group.
• Per-screening time learning curve among non-medical operators [Day 1]: The duration of each screening attempt in seconds is recorded for analysis. Participants repeat the screening process and self-evaluate their operating proficiency after each attempt. The procedure is concluded once a participant deems himself or herself proficient for three consecutive attempts. The learning curve is characterized by the change in mean per-screening duration across consecutive attempts.
• Screening duration comparing on-device and cloud-based deployment in resource-limited field settings [Day 1]: Mean screening duration in seconds per participant, recorded for each of the two deployment modalities: on-device inference and cloud-based inference using a functionally equivalent deployment of the identical model architecture with inference performed on a remote server. Local residents are divided into two equal groups and screened by village staff without medical training using either the on-device or the cloud-based system.
• User acceptability comparing on-device and cloud-based deployment in resource-limited field settings [Day 1]: User acceptability is assessed using a questionnaire investigating patient satisfaction, privacy concerns, and willingness to recommend the AI system, administered to participants in both the on-device and cloud-based deployment groups. Local residents are divided into two equal groups and screened by village staff without medical training using either the on-device or the cloud-based system. Questionnaire responses are compared between the two deployment groups.
• Diagnostic accuracy of the on-device AI system in population-level community screening [Day 1]: Diagnostic accuracy is defined as the proportion of participants for whom the AI system assigns a correct diagnostic label concordant with the reference standard. Village staff without medical training use the on-device AI system to screen consecutively enrolled local residents. The reference standard is established through slit-lamp examinations performed by ophthalmologists.
• Specificity of the on-device AI system in population-level community screening [Day 1]: Specificity is defined as the proportion of participants without a given anterior segment disease who are correctly identified as negative by the on-device AI system (true negatives divided by the sum of true negatives and false positives), calculated separately for each target disease category. Village staff without medical training use the on-device AI system to screen consecutively enrolled local residents. The reference standard is established through slit-lamp examinations performed by ophthalmologists.

Sex: All, ages: 18 Years to any, accepts healthy volunteers

Inclusion Criteria:

* Adults aged 18 years or older;
* Willing to participate and able to provide written informed consent prior to enrollment.

Exclusion Criteria:

* Unable to cooperate with anterior segment image capture (including smartphone-based photography or slit-lamp biomicroscopy).

• Longhui Li, Principal investigator, Zhongshan Ophthalmic Center, Sun Yat-sen University