Self-Screening for Ocular Surface Malignancies Using a Smartphone

A newly developed diagnostic system, CaptureTumor (CaT), that is driven by artificial intelligence (AI), was designed specifically for smartphone deployment. According to the investigators led by Ruixin Wang, MD, PhD, the system “achieved diagnostic accuracy comparable with that of specialist-graded slit-lamp evaluation” for detecting rare ocular surface malignancies.¹ Wang is from the Zhongshan Ophthalmic Center, Sun Yat-sen University, WHO Collaborating Centre for Eye Care and Vision, State Key Laboratory of Ophthalmology, Guangdong Provincial Key Laboratory of Ophthalmology and Visual Science, Guangdong Provincial Clinical Research Center for Ocular Diseases, Guangzhou, China.

Wang and colleagues explained that “effective screening for rare diseases, a critical component of equitable care, continues to present substantial systemic challenges. Early detection is often compromised by limited specialist availability and low public awareness.² Studies report that approximately 42% of patients with rare diseases are initially misdiagnosed, experiencing a diagnostic odyssey that spans a mean of 7 years and involves 8 or more physicians before accurate referral.³ Ocular malignant surface tumors exemplify this clinical and public health dilemma.”

They underscored the importance of early detection of ocular surface malignancies, which threaten vision and survival. Such lesions can often be misdiagnosed as benign because of their subtle presentation and the absence of widely accessible screening tools, which can cause treatment delays and the need for extensive surgical intervention.

The research group set out to develop and validate a smartphone system for proactive self-screening of ocular surface malignancies in the general population. They conducted a nonrandomized clinical trial (NCT05645341) in centers across China over a 6-month period. They trained and validated a deep learning model using 12 years of multicenter slit-lamp images. They then optimized the system for smartphone-based photography and it was deployed through a widely disseminated mobile application.

The study participants used the CaT application to capture images of suspected ocular surface lesions. The application provided immediate binary (benign vs malignant) and multiclass risk stratification and triaged high-risk cases for expedited clinical referral, the investigators explained.

The primary outcome was the area under the receiver operating characteristic curve (AUC) for differentiating malignant from benign lesions; the secondary outcomes included sensitivity, specificity, and the number of histopathologically confirmed malignancies detected.

What did the CaT application find?

A total of 614 individuals (median age 46 years; 301 women) completed the screening at home using the application.

Wang and colleagues reported, “After optimizing the image quality, the smartphone-based CaT achieved an AUC of 0.905 (95% confidence interval [CI], 0.837-0.973), comparable with the performance of the slit-lamp-based model (AUC = 0.945; 95% CI, 0.918-0.972). During real-world screening, 20 malignancies were pathologically confirmed among the 614 participants, with 19 of 20 participants (95%) newly diagnosed, and no cases requiring enucleation. At the population level, CaT demonstrated an AUC of 0.977 (95% CI, 0.964-0.990), a sensitivity of 89.3% (95% CI, 86.7%-91.9%), and a specificity of 95.9% (95% CI, 94.2%-97.6%).”

They concluded, “We developed and validated CaT, an AI-driven smartphone-based system for screening ocular malignancies, and demonstrated its real-world potential in a nationwide implementation study. This mobile health model offers a potentially scalable, accessible, and affordable strategy for early detection of rare, vision- and life-threatening diseases. Further validation and long-term assessment will be essential to determine its sustained potential to influence patient outcomes and health systems worldwide.”

References
1. Wang R, Bi S, Lin D, et al. Smartphone-based proactive self-screening for ocular surface malignancies. A nonrandomized clinical trial. JAMA Ophthalmol. 2026; published online June 4. doi: 10.1001/jamaophthalmol.2026.1609
2. Wang JB, Guo JJ, Yang L, Zhang YD, Sun ZQ, Zhang YJ. Rare diseases and legislation in China. Lancet. 2010;375:708-709. doi:10.1016/S0140-6736(10)60240-1
3. Tambuyzer E, Vandendriessche B, Austin CP, et al. Therapies for rare diseases: therapeutic modalities, progress and challenges ahead. Nat Rev Drug Discov. 2020;19:93-111. doi:10.1038/s41573-019-0049-9