A Novel Meibomian Gland Morphology Analytic System Based on a Convolutional Neural Network

Artificial Intelligence in Optometry: Current and Future Perspectives

With the global shortage of eye care professionals and the increasing burden of vision impairment, particularly in low- and middle-income countries, there is an urgent need for innovative solutions to bridge gaps in eye care services. Advances in artificial intelligence (AI) over recent decades have significantly impacted healthcare, including the field of optometry. When integrated into optometric workflows, AI has the potential to streamline decision-making processes and enhance system efficiency. To realize this potential, it is essential to develop AI models that can improve each stage of the patient care workflow, including screening, detection, diagnosis, and management. This review explores the application of AI in optometry, focusing on its potential to optimize various aspects of patient care. We examined AI models across key areas in optometry. Our analysis considered crucial parameters, including model selection, sample sizes for training and validation, evaluation metrics, and the explainability of the models. This comprehensive review identified both the strengths and weaknesses of existing AI models. The majority of image-based studies utilized CNN or transfer learning models, while clinical data-based studies primarily employed RF, SVM, and XGBoost. In general, AI models trained on large datasets achieved higher accuracy. However, many optometry-focused models faced limitations due to insufficient sample sizes-28% of studies were trained on fewer than 500 samples, 18% used fewer than 200 samples, and over half validated their models on fewer than 500 samples, with 38% validating on fewer than 200. Additionally, some studies that used the same data for both training and validation experienced overfitting, leading to reduced accuracy. Notably, 20% of the included studies reported accuracy below 80%, limiting their practical applicability in clinical settings. This review provides optometrists with valuable insights into the strengths and weaknesses of AI models in the field, aiding in their informed implementation in clinical settings.

Strip and boundary detection multi-task learning network for segmentation of meibomian glands

BACKGROUND

Automatic segmentation of meibomian glands in near-infrared meibography images is basis of morphological parameter analysis, which plays a crucial role in facilitating the diagnosis of meibomian gland dysfunction (MGD). The special strip shape and the adhesion between glands make the automatic segmentation of meibomian glands very challenging.

PURPOSE

A strip and boundary detection multi-task learning network (SBD-MTLNet) based on encoder-decoder structure is proposed to realize the automatic segmentation of meibomian glands.

METHODS

A strip mixed attention module (SMAM) is proposed to enhance the network's ability to recognize the strip shape of glands. To alleviate the problem of adhesion between glands, a boundary detection auxiliary network (BDA-Net) is proposed, which introduces boundary features to assist gland segmentation. A self-adaptive interactive information fusion module (SIIFM) based on reverse attention mechanism is proposed to realize information complementation between meibomian gland segmentation and boundary detection tasks. The proposed SBD-MTLNet has been evaluated on an in-house dataset (453 images) and a public dataset MGD-1K (1000 images). Due to the limited number of images, a five-fold cross validation strategy is adopted.

RESULTS

Average dice coefficient of the proposed SBD-MTLNet reaches 81.08% and 84.32% on the in-house dataset and the public one, respectively. Comprehensive experimental results demonstrate the effectiveness the proposed SBD-MTLNet, outperforming other state-of-the-art methods.

CONCLUSIONS

The proposed SBD-MTLNet can focus more on the shape characteristics of the meibomian glands and the boundary contour information between the adjacent glands via multi-task learning strategy. The segmentation results of the proposed method can be used for the quantitative morphological characteristics analysis of meibomian glands, which has potential for the auxiliary diagnosis of MGD in clinic.

Meibomian gland alterations in allergic conjunctivitis: insights from a novel quantitative analysis algorithm

Purpose

To investigate the changes in meibomian gland (MG) structure in allergic conjunctivitis (AC) patients using an intelligent quantitative analysis algorithm and to explore the relationship between these changes and clinical parameters.

Methods

A total of 252 eyes from patients with AC and 200 eyes from normal controls were examined. Infrared meibography was performed using the non-contact mode of the Keratograph 5M. MG images were analyzed using a deep learning-based a quantitative analysis algorithm to evaluate gland length, area, dropout ratio, and deformation. Clinical parameters, including tear meniscus height, tear break up time (TBUT), conjunctival hyperemia, and Ocular Surface Disease Index (OSDI) scores, were assessed and correlated with changes in the structure of MG.

Results

The average MG length in AC patients was 4.48 ± 1.04 mm, shorter compared to the control group (4.72 ± 0.94 mm). The average length of the central 5 glands in AC patients was 4.94 ± 1.67 mm, which was also shorter than the control group's central 5 glands (5.38 ± 1.42 mm). Furthermore, the central 5 glands' area in AC patients (1.61 ± 0.64 mm2) was reduced compared to the control group (1.79 ± 0.62 mm2). Tear meniscus height was lower in the allergy group (0.26 ± 0.10 mm) compared to the control group (0.44 ± 0.08 mm) (P < 0.05). The non-invasive first tear film break-up time was shorter in the allergy group (8.65 ± 6.31 s) than in the control group (10.48 ± 2.58 s) (P < 0.05). Conjunctival congestion was higher in the allergy group (1.1 ± 0.52) compared to the control group (0.97 ± 0.30) (P < 0.05). The OSDI score in the allergy group (8.33 ± 7.6) was higher than that in the control group (4.00 ± 0.50) (P < 0.05). Correlation analysis revealed that the gland dropout ratio was positively associated with male gender and negatively associated with age and OSDI scores. Additionally, despite an increased number of MG, tear film stability was not improved.

Conclusion

Through the intelligent quantitative algorithm, we found that AC leads to significant changes in MG structure, particularly affecting gland length and central area.

Advances in artificial intelligence for meibomian gland evaluation: A comprehensive review

Meibomian gland dysfunction (MGD) is increasingly recognized as a critical contributor to evaporative dry eye, significantly impacting visual quality. With a global prevalence estimated at 35.8 %, it presents substantial challenges for clinicians. Conventional manual evaluation techniques for MGD face limitations characterized by inefficiencies, high subjectivity, limited big data processing capabilities, and a dearth of quantitative analytical tools. With rapidly advancing artificial intelligence (AI) techniques revolutionizing ophthalmology, studies are now leveraging sophisticated AI methodologies--including computer vision, unsupervised learning, and supervised learning--to facilitate comprehensive analyses of meibomian gland (MG) evaluations. These evaluations employ various techniques, including slit lamp examination, infrared imaging, confocal microscopy, and optical coherence tomography. This paradigm shift promises enhanced accuracy and consistency in disease evaluation and severity classification. While AI has achieved preliminary strides in meibomian gland evaluation, ongoing advancements in system development and clinical validation are imperative. We review the evolution of MG evaluation, juxtapose AI-driven methods with traditional approaches, elucidate the specific roles of diverse AI technologies, and explore their practical applications using various evaluation techniques. Moreover, we delve into critical considerations for the clinical deployment of AI technologies and envisages future prospects, providing novel insights into MG evaluation and fostering technological and clinical progress in this arena.

Applications of Artificial Intelligence in Diagnosis of Dry Eye Disease: A Systematic Review and Meta-Analysis

PURPOSE

Clinical diagnosis of dry eye disease is based on a subjective Ocular Surface Disease Index questionnaire or various objective tests, however, these diagnostic methods have several limitations.

METHODS

We conducted a comprehensive review of articles discussing various applications of artificial intelligence (AI) models in the diagnosis of the dry eye disease by searching PubMed, Web of Science, Scopus, and Google Scholar databases up to December 2022. We initially extracted 2838 articles, and after removing duplicates and applying inclusion and exclusion criteria based on title and abstract, we selected 47 eligible full-text articles. We ultimately selected 17 articles for the meta-analysis after applying inclusion and exclusion criteria on the full-text articles. We used the Standards for Reporting of Diagnostic Accuracy Studies to evaluate the quality of the methodologies used in the included studies. The performance criteria for measuring the effectiveness of AI models included area under the receiver operating characteristic curve, sensitivity, specificity, and accuracy. We calculated the pooled estimate of accuracy using the random-effects model.

RESULTS

The meta-analysis showed that pooled estimate of accuracy was 91.91% (95% confidence interval: 87.46-95.49) for all studies. The mean (±SD) of area under the receiver operating characteristic curve, sensitivity, and specificity were 94.1 (±5.14), 89.58 (±6.13), and 92.62 (±6.61), respectively.

CONCLUSIONS

This study revealed that AI models are more accurate in diagnosing dry eye disease based on some imaging modalities and suggested that AI models are promising in augmenting dry eye clinics to assist physicians in diagnosis of this ocular surface condition.

Artificial Intelligence in Dry Eye Disease: A Narrative Review

Dry eye disease (DED) is a multifactorial condition affecting millions worldwide, characterized by discomfort, visual disturbance, and potential damage to the ocular surface. The complexity of its diagnosis and management, driven by the diversity of symptoms and underlying causes, presents significant challenges to clinicians. Artificial intelligence (AI) has emerged as a transformative tool in healthcare, offering potential solutions to these challenges through its data analysis, pattern recognition, and predictive modeling capabilities. This narrative review explores the role of AI in diagnosing, treating, and managing dry eye disease. AI-driven tools such as machine learning algorithms, imaging technologies, and diagnostic platforms are examined for their ability to enhance diagnostic accuracy, personalize treatment approaches, and optimize patient outcomes. Furthermore, the review addresses the limitations of AI technologies in ophthalmology, including the need for robust clinical validation, data privacy concerns, and the ethical considerations of integrating AI into clinical practice. The findings suggest that while AI holds promise for improving the care of patients with DED, ongoing research and development are crucial to realizing its full potential.

Dual-Mode Imaging System for Early Detection and Monitoring of Ocular Surface Diseases

The global prevalence of ocular surface diseases (OSDs), such as dry eyes, conjunctivitis, and subconjunctival hemorrhage (SCH), is steadily increasing due to factors such as aging populations, environmental influences, and lifestyle changes. These diseases affect millions of individuals worldwide, emphasizing the importance of early diagnosis and continuous monitoring for effective treatment. Therefore, we present a deep learning-enhanced imaging system for the automated, objective, and reliable assessment of these three representative OSDs. Our comprehensive pipeline incorporates processing techniques derived from dual-mode infrared (IR) and visible (RGB) images. It employs a multi-stage deep learning model to enable accurate and consistent measurement of OSDs. This proposed method has achieved a 98.7% accuracy with an F1 score of 0.980 in class classification and a 96.2% accuracy with an F1 score of 0.956 in SCH region identification. Furthermore, our system aims to facilitate early diagnosis of meibomian gland dysfunction (MGD), a primary factor causing dry eyes, by quantitatively analyzing the meibomian gland (MG) area ratio and detecting gland morphological irregularities with an accuracy of 88.1% and an F1 score of 0.781. To enhance convenience and timely OSD management, we are integrating a portable IR camera for obtaining meibography during home inspections. Our system demonstrates notable improvements in expanding dual-mode image-based diagnosis for broader applicability, effectively enhancing patient care efficiency. With its automation, accuracy, and compact design, this system is well-suited for early detection and ongoing assessment of OSDs, contributing to improved eye healthcare in an accessible and comprehensible manner.

Red filter meibography by smartphones in patients with meibomian gland dysfunction: a validity and reliability study

OBJECTIVE

The objective of this study is to determine the validity and reliability of the red filter meibography by smartphone compared with infrared in assessing meibomian gland drop-out.

METHODS AND ANALYSIS

An analytical cross-sectional study was done with a total of 35 subjects (68 eyes) with suspected MGD based on symptoms and lid morphological abnormalities. Meibomian glands were photographed using two smartphones (Samsung S9 and iPhone XR) on a slit-lamp with added red filter. Images were assessed subjectively using meiboscore by the two raters and drop-out percentages were assessed by ImageJ.

RESULTS

There was no agreement in meiboscore and a minimal level of agreement in drop-out percentages between red filter meibography and infrared. Inter-rater reliability showed no agreement between two raters. Intra-rater reliability demonstrated weak agreement in rater 1 and no agreement in rater 2.

CONCLUSION

Validity of the red filter meibography technique by smartphones is not yet satisfactory in evaluating drop-out. Further improvement on qualities of images must be done and research on subjective assessment was deemed necessary due to poor results of intrarater and inter-rater reliability.

A Deep Learning Approach for Meibomian Gland Appearance Evaluation

Purpose

To develop and evaluate a deep learning algorithm for Meibomian gland characteristics calculation.

Design

Evaluation of diagnostic technology.

Subjects

A total of 1616 meibography images of both the upper (697) and lower (919) eyelids from a total of 282 individuals.

Methods

Images were collected using the LipiView II device. All the provided data were split into 3 sets: the training, validation, and test sets. Data partitions used proportions of 70/10/20% and included data from 2 optometry settings. Each set was separately partitioned with these proportions, resulting in a balanced distribution of data from both settings. The images were divided based on patient identifiers, such that all images collected for one participant could end up only in one set. The labeled images were used to train a deep learning model, which was subsequently used for Meibomian gland segmentation. The model was then applied to calculate individual Meibomian gland metrics. Interreader agreement and agreement between manual and automated methods for Meibomian gland segmentation were also carried out to assess the accuracy of the automated approach.

Main Outcome Measures

Meibomian gland metrics, including length ratio, area, tortuosity, intensity, and width, were measured. Additionally, the performance of the automated algorithms was evaluated using the aggregated Jaccard index.

Results

The proposed semantic segmentation-based approach achieved average aggregated Jaccard index of mean 0.4718 (95% confidence interval [CI], 0.4680-0.4771) for the 'gland' class and a mean of 0.8470 (95% CI, 0.8432-0.8508) for the 'eyelid' class. The result for object detection-based approach was a mean of 0.4476 (95% CI, 0.4426-0.4533). Both artificial intelligence-based algorithms underestimated area, length ratio, tortuosity, widthmean, widthmedian, width10th, and width90th. Meibomian gland intensity was overestimated by both algorithms compared with the manual approach. The object detection-based algorithm seems to be as reliable as the manual approach only for Meibomian gland width10th calculation.

Conclusions

The proposed approach can successfully segment Meibomian glands; however, to overcome problems with gland overlap and lack of image sharpness, the proposed method requires further development. The study presents another approach to utilizing automated, artificial intelligence-based methods in Meibomian gland health assessment that may assist clinicians in the diagnosis, treatment, and management of Meibomian gland dysfunction.

Financial Disclosures

The authors have no proprietary or commercial interest in any materials discussed in this article.

The influence of overnight orthokeratology and soft contact lens on the meibomian gland evaluated using an artificial intelligence analytic system

PURPOSE

To test the changes of meibomian gland (MG) morphology using an artificial intelligence (AI) analytic system in asymptomatic children wearing overnight orthokeratology (OOK) and soft contact lens (SCL).

METHODS

A retrospective study was conducted including 89 participants treated with OOK and 70 participants with SCL. Tear meniscus height (TMH), noninvasive tear breakup time (NIBUT), and meibography were obtained using Keratograph 5 M. MG tortuosity, height, width, density, and vagueness value were measured using an artificial intelligence (AI) analytic system.

RESULTS

In an average of 20.80 ± 10.83 months follow-up, MG width of the upper eyelid significantly increased and MG vagueness value significantly decreased after OOK and SCL treatment (all P < 0.05). MG tortuosity of the upper eyelid significantly increased after OOK treatment (P < 0.05). TMH and NIBUT did not differ significantly pre- and post- OOK and SCL treatment (all P > 0.05). The results from the GEE model demonstrated that OOK treatment positively affected MG tortuosity of both upper and lower eyelids (P < 0.001; P = 0.041, respectively) and MG width of the upper eyelid (P = 0.038), while it negatively affected MG density of the upper eyelid (P = 0.036) and MG vagueness value of both upper and lower eyelids (P < 0.001; P < 0.001, respectively). SCL treatment positively affected MG width of both upper and lower eyelids (P < 0.001; P = 0.049, respectively) as well as MG height of the lower eyelid (P = 0.009) and tortuosity of the upper eyelid, (P = 0.034) while it negatively affected MG vagueness value of both upper and lower eyelids (P < 0.001; P < 0.001, respectively). However, no significant relationship was found between the treatment duration and TMH, NIBUT, MG morphological parameters in OOK group. SCL treatment duration negatively affected MG height of the lower eyelid (P = 0.002).

CONCLUSIONS

OOK and SCL treatment in asymptomatic children can influence MG morphology. The AI analytic system may be an effective method to facilitate the quantitative detection of MG morphological changes.

Artificial intelligence-assisted diagnosis of ocular surface diseases

With the rapid development of computer technology, the application of artificial intelligence (AI) in ophthalmology research has gained prominence in modern medicine. Artificial intelligence-related research in ophthalmology previously focused on the screening and diagnosis of fundus diseases, particularly diabetic retinopathy, age-related macular degeneration, and glaucoma. Since fundus images are relatively fixed, their standards are easy to unify. Artificial intelligence research related to ocular surface diseases has also increased. The main issue with research on ocular surface diseases is that the images involved are complex, with many modalities. Therefore, this review aims to summarize current artificial intelligence research and technologies used to diagnose ocular surface diseases such as pterygium, keratoconus, infectious keratitis, and dry eye to identify mature artificial intelligence models that are suitable for research of ocular surface diseases and potential algorithms that may be used in the future.

Is histogram manipulation always beneficial when trying to improve model performance across devices? Experiments using a Meibomian gland segmentation model

Meibomian gland dysfunction (MGD) is caused by abnormalities of the meibomian glands (MG) and is one of the causes of evaporative dry eye (DED). Precise MG segmentation is crucial for MGD-related DED diagnosis because the morphological parameters of MG are of importance. Deep learning has achieved state-of-the-art performance in medical image segmentation tasks, especially when training and test data come from the same distribution. But in practice, MG images can be acquired from different devices or hospitals. When testing image data from different distributions, deep learning models that have been trained on a specific distribution are prone to poor performance. Histogram specification (HS) has been reported as an effective method for contrast enhancement and improving model performance on images of different modalities. Additionally, contrast limited adaptive histogram equalization (CLAHE) will be used as a preprocessing method to enhance the contrast of MG images. In this study, we developed and evaluated the automatic segmentation method of the eyelid area and the MG area based on CNN and automatically calculated MG loss rate. This method is evaluated in the internal and external testing sets from two meibography devices. In addition, to assess whether HS and CLAHE improve segmentation results, we trained the network model using images from one device (internal testing set) and tested on images from another device (external testing set). High DSC (0.84 for MG region, 0.92 for eyelid region) for the internal test set was obtained, while for the external testing set, lower DSC (0.69-0.71 for MG region, 0.89-0.91 for eyelid region) was obtained. Also, HS and CLAHE were reported to have no statistical improvement in the segmentation results of MG in this experiment.

Automation of dry eye disease quantitative assessment: A review

Dry eye disease (DED) is a common eye condition worldwide and a primary reason for visits to the ophthalmologist. DED diagnosis is performed through a combination of tests, some of which are unfortunately invasive, non‐reproducible and lack accuracy. The following review describes methods that diagnose and measure the extent of eye dryness, enabling clinicians to quantify its severity. Our aim with this paper is to review classical methods as well as those that incorporate automation. For only four ways of quantifying DED, we take a deeper look into what main elements can benefit from automation and the different ways studies have incorporated it. Like numerous medical fields, Artificial Intelligence (AI) appears to be the path towards quality DED diagnosis. This review categorises diagnostic methods into the following: classical, semi‐automated and promising AI‐based automated methods.

Uneven Index: A Digital Biomarker to Prompt Demodex Blepharitis Based on Deep Learning

Purpose: To evaluate ocular surface manifestations and morphological changes in meibomian glands (MGs) based on artificial intelligence (AI) analysis in patients with Demodex blepharitis.

Methods: In this retrospective study, 115 subjects were enrolled, including 64 subjects with Demodex blepharitis and 51 subjects without Demodex blepharitis as control group. Morphological indexes were evaluated for height, width, tortuosity, MG density, total variation, and the three types of corrected total variation as Uneven indexes.

Results: There were no statistically significant differences in all MGs’ average tortuosity and width between the two groups. The average height of all MGs and MG density were significantly lower in the Demodex blepharitis group than control group. The total variation and two types of Uneven indexes were significantly higher in the Demodex blepharitis group than in the control group. Especially the Uneven Index of total variation/MG density had an AUC of 0.822. And the sensitivity and specificity were 59.4% and 92.2%, respectively, at a cut-off value of 3971.667. In addition, Demodex blepharitis was associated with significantly lower meibum quality and expressibility, severe atrophy of MGs, a higher ocular surface disease index (OSDI), and more instability of the tear film.

Conclusion:Demodex mites are strongly associated with morphological changes in the MGs and may cause uneven gland atrophy. Therefore, the novel characteristic parameter, the Uneven index, may serve as a digital biomarker to evaluate uneven atrophy of MGs and prompt Demodex blepharitis.

Systematic Bibliometric and Visualized Analysis of Research Hotspots and Trends on the Application of Artificial Intelligence in Ophthalmic Disease Diagnosis

Background: Artificial intelligence (AI) has been used in the research of ophthalmic disease diagnosis, and it may have an impact on medical and ophthalmic practice in the future. This study explores the general application and research frontier of artificial intelligence in ophthalmic disease detection. Methods: Citation data were downloaded from the Web of Science Core Collection database to evaluate the extent of the application of Artificial intelligence in ophthalmic disease diagnosis in publications from 1 January 2012, to 31 December 2021. This information was analyzed using CiteSpace.5.8. R3 and Vosviewer. Results: A total of 1,498 publications from 95 areas were examined, of which the United States was determined to be the most influential country in this research field. The largest cluster labeled "Brownian motion" was used prior to the application of AI for ophthalmic diagnosis from 2007 to 2017, and was an active topic during this period. The burst keywords in the period from 2020 to 2021 were system, disease, and model. Conclusion: The focus of artificial intelligence research in ophthalmic disease diagnosis has transitioned from the development of AI algorithms and the analysis of abnormal eye physiological structure to the investigation of more mature ophthalmic disease diagnosis systems. However, there is a need for further studies in ophthalmology and computer engineering.

Meibomian Gland Density: An Effective Evaluation Index of Meibomian Gland Dysfunction Based on Deep Learning and Transfer Learning

We aimed to establish an artificial intelligence (AI) system based on deep learning and transfer learning for meibomian gland (MG) segmentation and evaluate the efficacy of MG density in the diagnosis of MG dysfunction (MGD). First, 85 eyes of 85 subjects were enrolled for AI system-based evaluation effectiveness testing. Then, from 2420 randomly selected subjects, 4006 meibography images (1620 upper eyelids and 2386 lower eyelids) graded by three experts according to the meiboscore were analyzed for MG density using the AI system. The updated AI system achieved 92% accuracy (intersection over union, IoU) and 100% repeatability in MG segmentation after 4 h of training. The processing time for each meibography was 100 ms. We discovered a significant and linear correlation between MG density and ocular surface disease index questionnaire (OSDI), tear break-up time (TBUT), lid margin score, meiboscore, and meibum expressibility score (all p < 0.05). The area under the curve (AUC) was 0.900 for MG density in the total eyelids. The sensitivity and specificity were 88% and 81%, respectively, at a cutoff value of 0.275. MG density is an effective index for MGD, particularly supported by the AI system, which could replace the meiboscore, significantly improve the accuracy of meibography analysis, reduce the analysis time and doctors’ workload, and improve the diagnostic efficiency.

Research on an Intelligent Lightweight-Assisted Pterygium Diagnosis Model Based on Anterior Segment Images

AIMS

The lack of primary ophthalmologists in China results in the inability of basic-level hospitals to diagnose pterygium patients. To solve this problem, an intelligent-assisted lightweight pterygium diagnosis model based on anterior segment images is proposed in this study.

METHODS

Pterygium is a common and frequently occurring disease in ophthalmology, and fibrous tissue hyperplasia is both a diagnostic biomarker and a surgical biomarker. The model diagnosed pterygium based on biomarkers of pterygium. First, a total of 436 anterior segment images were collected; then, two intelligent-assisted lightweight pterygium diagnosis models (MobileNet 1 and MobileNet 2) based on raw data and augmented data were trained via transfer learning. The results of the lightweight models were compared with the clinical results. The classic models (AlexNet, VGG16 and ResNet18) were also used for training and testing, and their results were compared with the lightweight models. A total of 188 anterior segment images were used for testing. Sensitivity, specificity, F1-score, accuracy, kappa, area under the concentration-time curve (AUC), 95% CI, size, and parameters are the evaluation indicators in this study.

RESULTS

There are 188 anterior segment images that were used for testing the five intelligent-assisted pterygium diagnosis models. The overall evaluation index for the MobileNet2 model was the best. The sensitivity, specificity, F1-score, and AUC of the MobileNet2 model for the normal anterior segment image diagnosis were 96.72%, 98.43%, 96.72%, and 0976, respectively; for the pterygium observation period anterior segment image diagnosis, the sensitivity, specificity, F1-score, and AUC were 83.7%, 90.48%, 82.54%, and 0.872, respectively; for the surgery period anterior segment image diagnosis, the sensitivity, specificity, F1-score, and AUC were 84.62%, 93.50%, 85.94%, and 0.891, respectively. The kappa value of the MobileNet2 model was 77.64%, the accuracy was 85.11%, the model size was 13.5 M, and the parameter size was 4.2 M.

CONCLUSION

This study used deep learning methods to propose a three-category intelligent lightweight-assisted pterygium diagnosis model. The developed model can be used to screen patients for pterygium problems initially, provide reasonable suggestions, and provide timely referrals. It can help primary doctors improve pterygium diagnoses, confer social benefits, and lay the foundation for future models to be embedded in mobile devices.