Explainable Machine Learning Approach as a Tool to Understand Factors Used to Select the Refractive Surgery Technique on the Expert Level

Artificial intelligence in the diagnosis and management of refractive errors

Refractive error is among the leading causes of visual impairment globally. The diagnosis and management of refractive error has traditionally relied on comprehensive eye examinations by eye care professionals, but access to these specialized services has remained limited in many areas of the world. Given this, artificial intelligence (AI) has shown immense potential in transforming the diagnosis and management of refractive error. We review AI applications across various aspects of refractive error care - from axial length prediction using fundus images to risk stratification for myopia progression. AI algorithms can be trained to analyze clinical data to detect refractive error as well as predict associated risks of myopia progression. For treatments such as implantable collamer and orthokeratology lenses, AI models facilitate vault size prediction and optimal lens fitting with high accuracy. Furthermore, AI has demonstrated promise in optimizing surgical planning and outcomes for refractive procedures. Emerging digital technologies such as telehealth, smartphone applications, and virtual reality integrated with AI present novel avenues for refractive error screening. We discuss key challenges, including limited validation datasets, lack of data standardization, image quality issues, population heterogeneity, practical deployment, and ethical considerations regarding patient privacy that need to be addressed before widespread clinical implementation.

Analyzing the Effect of Surgical and Corneal Parameters on the Postoperative Refractive Outcomes of Smile in Myopic Eyes Based on Machine Learning

PURPOSE

To analyze the influence of individual parameters on the postoperative refractive outcomes of small incision lenticule extraction (SMILE) in myopic eyes using machine learning.

DESIGN

Retrospective Clinical Cohort Study METHODS: This study included 477 patients (922 eyes) of SMILE and divided them into two groups based on postoperative spherical equivalent (SE) ≤ -0.50D to analyze the factors influencing postoperative refractive outcomes. The XGBoost model took 72 clinical parameters (34 biomechanical, 31 morphological, 4 surgical-related, and 3 preoperative refractive parameters) as features; randomly selected 42 eyes from the good refractive outcomes group and all eyes (a total of 42 eyes) from the poor refractive outcomes group to conduct 100 times influence factors analysis.

RESULTS

The 10 most important factors influencing postoperative refractive outcomes included 3 surgery-related parameters (PTA, LTmax, and RST), 3 corneal biomechanical parameters (HC Time, Deflection Amp Max (ms), and SSI), 2 corneal morphological parameters (R Per F and Rs F), and 2 preoperative refractive parameters (SE and SD). When PTA ≥ 25.09%, LTmax ≥ 139μm, SE ≤ -7.00D, or SD ≤ -6.75D, the postoperative SE significantly increased (all P < 0.05), with averages of -0.183D, -0.171D, -0.188D, and -0.184D. After controlling for age, intraocular pressure, and corneal thickness, the postoperative SE significantly increased when HC time ≥ 17.422 and Deflection Amp Max (ms) ≥ 16.616, reaching -0.209D and -0.202D.

CONCLUSIONS

More corneal tissue ablation, longer HC Time and Deflection Amp Max (ms), lower SSI, and higher preoperative refractive error may result in poor postoperative refractive outcomes.

Application of multimodal large language models for safety indicator calculation and contraindication prediction in laser vision correction

This study demonstrates the potential of multimodal large language models in calculating safety indicators and predicting contraindications for laser vision correction. ChatGPT-4 effectively analyzed ocular data, calculated key indicators, generated calculator codes, and outperformed traditional machine learning models and indicators in handling unstructured data and corneal topography. Its modality-independent system enabled efficient and accurate data analysis. Despite longer processing times, ChatGPT-4's performance highlights its potential as a decision-support tool, offering advancements in improving safety.

Automated and code-free development of a risk calculator using ChatGPT-4 for predicting diabetic retinopathy and macular edema without retinal imaging

BACKGROUND

Diabetic retinopathy (DR) and macular edema (DME) are critical causes of vision loss in patients with diabetes. In many communities, access to ophthalmologists and retinal imaging equipment is limited, making screening for diabetic retinal complications difficult in primary health care centers. We investigated whether ChatGPT-4, an advanced large-language-model chatbot, can develop risk calculators for DR and DME using health check-up tabular data without the need for retinal imaging or coding experience.

METHODS

Data-driven prediction models were developed using medical history and laboratory blood test data from diabetic patients in the Korea National Health and Nutrition Examination Surveys (KNHANES). The dataset was divided into training (KNHANES 2017-2020) and validation (KNHANES 2021) datasets. ChatGPT-4 was used to build prediction formulas for DR and DME and developed a web-based risk calculator tool. Logistic regression analysis was performed by ChatGPT-4 to predict DR and DME, followed by the automatic generation of Hypertext Markup Language (HTML) code for the web-based tool. The performance of the models was evaluated using areas under the curves of receiver operating characteristic curve (ROC-AUCs).

RESULTS

ChatGPT-4 successfully developed a risk calculator for DR and DME, operational on a web browser without any coding experience. The validation set showed ROC-AUCs of 0.786 and 0.835 for predicting DR and DME, respectively. The performance of the ChatGPT-4 developed models was comparable to those created using various machine-learning tools.

CONCLUSION

By utilizing ChatGPT-4 with code-free prompts, we overcame the technical barriers associated with using coding skills for developing prediction models, making it feasible to build a risk calculator for DR and DME prediction. Our approach offers an easily accessible tool for the risk prediction of DM and DME in diabetic patients during health check-ups, without the need for retinal imaging. Based on this automatically developed risk calculator using ChatGPT-4, health care workers will be able to effectively screen patients who require retinal examinations using only medical history and laboratory data. Future research should focus on validating this approach in diverse populations and exploring the integration of more comprehensive clinical data to enhance predictive performance.

Artificial Intelligence in Surgery: A Systematic Review of Use and Validation

Background: Artificial Intelligence (AI) holds promise for transforming healthcare, with AI models gaining increasing clinical use in surgery. However, new AI models are developed without established standards for their validation and use. Before AI can be widely adopted, it is crucial to ensure these models are both accurate and safe for patients. Without proper validation, there is a risk of integrating AI models into practice without sufficient evidence of their safety and accuracy, potentially leading to suboptimal patient outcomes. In this work, we review the current use and validation methods of AI models in clinical surgical settings and propose a novel classification system. Methods: A systematic review was conducted in PubMed and Cochrane using the keywords "validation", "artificial intelligence", and "surgery", following PRISMA guidelines. Results: The search yielded a total of 7627 articles, of which 102 were included for data extraction, encompassing 2,837,211 patients. A validation classification system named Surgical Validation Score (SURVAS) was developed. The primary applications of models were risk assessment and decision-making in the preoperative setting. Validation methods were ranked as high evidence in only 45% of studies, and only 14% of the studies provided publicly available datasets. Conclusions: AI has significant applications in surgery, but validation quality remains suboptimal, and public data availability is limited. Current AI applications are mainly focused on preoperative risk assessment and are suggested to improve decision-making. Classification systems such as SURVAS can help clinicians confirm the degree of validity of AI models before their application in practice.

Artificial intelligence in the anterior segment of eye diseases

Ophthalmology is a subject that highly depends on imaging examination. Artificial intelligence (AI) technology has great potential in medical imaging analysis, including image diagnosis, classification, grading, guiding treatment and evaluating prognosis. The combination of the two can realize mass screening of grass-roots eye health, making it possible to seek medical treatment in the mode of "first treatment at the grass-roots level, two-way referral, emergency and slow treatment, and linkage between the upper and lower levels". On the basis of summarizing the AI technology carried out by scholars and their teams all over the world in the field of ophthalmology, quite a lot of studies have confirmed that machine learning can assist in diagnosis, grading, providing optimal treatment plans and evaluating prognosis in corneal and conjunctival diseases, ametropia, lens diseases, glaucoma, iris diseases, etc. This paper systematically shows the application and progress of AI technology in common anterior segment ocular diseases, the current limitations, and prospects for the future.

Comparison of early visual outcomes after SMILE using VISUMAX 800 and VISUMAX 500 for myopia: a retrospective matched case-control study

VISUMAX 800 was introduced to improve the patient experience and clinical outcomes of small incision lenticule extraction (SMILE). This was a retrospective, matched, and case-control study (1:2) controlled for preoperative central corneal thickness and refractive error that compared early refractive and visual outcomes after SMILE using VISUMAX 800 and VISUMAX 500 to treat myopia. We included 50 eyes that underwent the VISUMAX 800 SMILE and 100 eyes that underwent the VISUMAX 500 SMILE. SMILE using VISUMAX 800 was performed using the CentraLign aid for vertex centration. Cyclotorsion was controlled by an OcuLign assistant in the VISUMAX 800 group after corneal marking. Corneal higher-order aberrations (HOAs) were evaluated using a Pentacam 1 month after surgery. No differences were observed in the pre- and post-operative refractive and visual outcomes at 1 day, 1 month, and 6 months after surgery. VISUMAX 800 induced less total HOAs than VISUMAX 500 (P = 0.036). No statistically significant differences were observed in the amounts of induced spherical aberrations or vertical and horizontal comas. No differences were observed in the 1 month and 6 months refractive and visual outcomes between two SMILE procedures, except for VISUMAX 800, which resulted in lower postoperative total HOAs than VISUMAX 500.

Prediction of substantial closed-globe injuries in orbital wall fractures

PURPOSE

To determine risk factors for substantial closed-globe injuries in orbital fractures (SCGI) and to develop the best multivariate model for the prediction of SCGI.

METHODS

A retrospective study was performed on patients diagnosed with orbital fractures at Farabi Hospital between 2016 and 2022. Patients with a comprehensive ophthalmologic examination and orbital CT scan were included. Predictive signs or imaging findings for SCGI were identified by logistic regression (LR) analysis. Support vector machine (SVM), random forest regression (RFR), and extreme gradient boosting (XGBoost) were also trained using a fivefold cross-validation method.

RESULTS

A total of 415 eyes from 403 patients were included. Factors associated with an increased risk of SCGI were reduced uncorrected visual acuity (UCVA), increased difference between UCVA of the traumatic eye from the contralateral eye, older age, male sex, grade of periorbital soft tissue trauma, trauma in the occupational setting, conjunctival hemorrhage, extraocular movement restriction, number of fractured walls, presence of medial wall fracture, size of fracture, intraorbital emphysema and retrobulbar hemorrhage. The area under the curve of the receiver operating characteristic for LR, SVM, RFR, and XGBoost for the prediction of SCGI was 57.2%, 68.8%, 63.7%, and 73.1%, respectively.

CONCLUSIONS

Clinical and radiographic findings could be utilized to efficiently predict SCGI. XGBoost outperforms the logistic regression model in the prediction of SCGI and could be incorporated into clinical practice.

Role of Machine Learning and Artificial Intelligence in the Diagnosis and Treatment of Refractive Errors for Enhanced Eye Care: A Systematic Review

A significant contributor to blindness and visual impairment globally is uncorrected refractive error. To plan effective interventions, eye care professionals must promptly identify people at a high risk of acquiring myopia, and monitor disease progress. Artificial intelligence (AI) and machine learning (ML) have enormous potential to improve diagnosis and treatment. This systematic review explores the current state of ML and AI applications in the diagnoses and treatment of refractory errors in optometry. A systematic review and meta-analysis of studies evaluating the diagnostic performance of AI-based tools in PubMed was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. To find relevant studies on the use of ML or AI in the diagnosis or treatment of refractive errors in optometry, a thorough search was conducted in various electronic databases such as PubMed, Google Scholar, and Web of Science. The search was limited to studies published between January 2015 and December 2022. The search terms used were "refractive errors," "myopia," "optometry," "machine learning," "ophthalmology," and "artificial intelligence." A total of nine studies met the inclusion criteria and were included in the final analysis. ML is increasingly being utilized for automating clinical data processing as AI technology progresses, making the formerly labor-intensive work possible. AI models that primarily use a neural network demonstrated exceptional efficiency and performance in the analysis of vast medical data, rivaling board-certified, healthcare professionals. Several studies showed that ML models could support diagnosis and clinical decision-making. Moreover, an ML algorithm predicted future refraction values in patients with myopia. AI and ML models have great potential to improve the diagnosis and treatment of refractive errors in optometry.

Optimizing Refractive Outcomes of SMILE: Artificial Intelligence versus Conventional State-of-the-Art Nomograms

PURPOSE

AI (artificial intelligence)-based methodologies have become established tools for researchers and physicians in the entire field of ophthalmology. However, the potential of AI to optimize the refractive outcome of keratorefractive surgery by means of machine learning (ML)-based nomograms has not been exhausted yet. In this study, we wanted to comprehensively compare state-of-the-art conventional nomograms for Small-Incision-Lenticule-Extraction (SMILE) with a novel ML-based nomogram regarding both their spherical and astigmatic predictability.

METHODS

A total of 1,342 eyes were analyzed for creation of three different nomograms based on a linear model (LM), a generalized additive mixed model (GAMM) and an artificial-neuronal-network (ANN), respectively. A total of 16 patient- and treatment-related features were included. Each model was trained by 895 eyes and validated by the remaining 447 eyes. Predictability was assessed by the difference between attempted and achieved change in spherical equivalent (SE) and the difference between target induced astigmatism (TIA) and surgically induced astigmatism (SIA). The root mean squared error (RMSE) of each model was computed as a measure of overall model performance.

RESULTS

The RMSE of LM, GAMM and ANN were 0.355, 0.348 and 0.367 for the prediction of SE and 0.279, 0.278 and 0.290 for the astigmatic correction, respectively. By applying the created models, the theoretical yield of eyes within ±0.50 D of SE from target refraction improved from 82 to 83% (LM), 84% (GAMM) and 83% (ANN), respectively. Astigmatic outcomes showed an improvement of eyes within ±0.50 D from TIA from 90 to 93% (LM), 93% (GAMM) and 92% (ANN), respectively. Subjective manifest refraction was the single most influential covariate in all models.

CONCLUSION

Machine learning endorsed the validity of state-of-the-art linear and non-linear SMILE nomograms. However, improving the accuracy of subjective manifest refraction seems warranted for optimizing ±0.50 D SE predictability beyond an apparent methodological 90% limit.

Choice of refractive surgery types for myopia assisted by machine learning based on doctors' surgical selection data

In recent years, corneal refractive surgery has been widely used in clinics as an effective means to restore vision and improve the quality of life. When choosing myopia-refractive surgery, it is necessary to comprehensively consider the differences in equipment and technology as well as the specificity of individual patients, which heavily depend on the experience of ophthalmologists. In our study, we took advantage of machine learning to learn about the experience of ophthalmologists in decision-making and assist them in the choice of corneal refractive surgery in a new case. Our study was based on the clinical data of 7,081 patients who underwent corneal refractive surgery between 2000 and 2017 at the Department of Ophthalmology, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences. Due to the long data period, there were data losses and errors in this dataset. First, we cleaned the data and deleted the samples of key data loss. Then, patients were divided into three groups according to the type of surgery, after which we used SMOTE technology to eliminate imbalance between groups. Six statistical machine learning models, including NBM, RF, AdaBoost, XGBoost, BP neural network, and DBN were selected, and a ten-fold cross-validation and grid search were used to determine the optimal hyperparameters for better performance. When tested on the dataset, the multi-class RF model showed the best performance, with agreement with ophthalmologist decisions as high as 0.8775 and Macro F1 as high as 0.8019. Furthermore, the results of the feature importance analysis based on the SHAP technique were consistent with an ophthalmologist's practical experience. Our research will assist ophthalmologists in choosing appropriate types of refractive surgery and will have beneficial clinical effects.

Development of a Machine-Learning-Based Tool for Overnight Orthokeratology Lens Fitting

Purpose

Orthokeratology (ortho-K) is widely used to control myopia. Overnight ortho-K lens fitting with the selection of appropriate parameters is an important technique for achieving successful reductions in myopic refractive error. In this study, we developed a machine-learning model that could select ortho-K lens parameters at an expert level.

Methods

Machine-learning models were established to predict the optimal ortho-K parameters, including toric lens option (toric or non-toric), overall diameter (OAD; 10.5 or 11.0 mm), base curve (BC), return zone depth (RZD), landing zone angle (LZA), and lens sagittal depth (LensSag). The analysis included 547 eyes of 297 Korean adolescents with myopia or astigmatism. The dataset was randomly divided into training (80%, n = 437 eyes) and validation (20%, n = 110 eyes) sets at the patient level. The model was trained based on clinical ortho-K lens fitting performed by highly experienced experts and ophthalmic measurements.

Results

The final machine-learning models showed accuracies of 92.7% and 86.4% for predicting the toric lens option and OAD, respectively. The mean absolute errors for the BC, RZD, LZA, and LensSag predictions were 0.052 mm, 2.727 µm, 0.118°, and 5.215 µm, respectively. The machine-learning model outperformed the manufacturer's conventional initial lens selector in predicting BC and RZD.

Conclusions

We developed an expert-level machine-learning-based model for determining comprehensive ortho-K lens parameters. We also created a web-based application.

Translational Relevance

This model may provide more accurate fitting parameters for lenses than those of conventional calculations, thus reducing the need to rely on trial and error.

Predicting Implantable Collamer Lens Vault Using Machine Learning Based on Various Preoperative Biometric Factors

Purpose

To predict the vault size after Implantable Collamer Lens (ICL) V4c implantation using machine learning methods and to compare the predicted vault with the conventional manufacturer's nomogram.

Methods

This study included 707 patients (707 eyes) who underwent ICL V4c implantation at the Department of Ophthalmology, Peking Union Medical College Hospital, from September 2019 to January 2022. Random Forest Regression (RFR), XGBoost, and linear regression (LR) were used to predict the vault size 1 week after ICL V4c implantation. The mean absolute error (MAE), median absolute error (MedAE), root mean square error (RMSE), symmetric mean absolute percentage error (SMAPE), and Bland-Altman plot were utilized to compare the prediction performance of these machine learning methods.

Results

The dataset was divided into a training set of 180 patients (180 eyes) and a test set of 527 patients (527 eyes). XGBoost had the lowest prediction error, with mean MAE, RMSE, and SMAPE values of 121.70 µm, 148.87 µm, and 19.13%, respectively. The Bland‒Altman plots of RFR and XGBoost showed better prediction consistency than LR. However, XGBoost showed narrower 95% limits of agreement (LoA) than RFR, ranging from -307.12 to 256.59 µm.

Conclusions

XGBoost demonstrated better predictive performance than RFR and LR, as it had the lowest prediction error and the narrowest 95% LoA. Machine learning may be applicable for vault prediction, and it might be helpful for reducing the complications and the secondary surgery rate.

Translational Relevance

Using the proposed machine learning model, surgeons can consider the postoperative vault to reduce the surgical complications.

Artificial Intelligence in Ophthalmic Surgery: Current Applications and Expectations

Artificial Intelligence (AI) has found rapidly growing applications in ophthalmology, achieving robust recognition and classification in most kind of ocular diseases. Ophthalmic surgery is one of the most delicate microsurgery, requiring high fineness and stability of surgeons. The massive demand of the AI assist ophthalmic surgery will constitute an important factor in boosting accelerate precision medicine. In clinical practice, it is instrumental to update and review the considerable evidence of the current AI technologies utilized in the investigation of ophthalmic surgery involved in both the progression and innovation of precision medicine. Bibliographic databases including PubMed and Google Scholar were searched using keywords such as "ophthalmic surgery", "surgical selection", "candidate screening", and "robot-assisted surgery" to find articles about AI technology published from 2018 to 2023. In addition to the Editorials and letters to the editor, all types of approaches are considered. In this paper, we will provide an up-to-date review of artificial intelligence in eye surgery, with a specific focus on its application to candidate screening, surgery selection, postoperative prediction, and real-time intraoperative guidance.

Artificial intelligence-aided diagnosis and treatment in the field of optometry

With the rapid development of computer technology, the application of artificial intelligence (AI) to ophthalmology has gained prominence in modern medicine. As modern optometry is closely related to ophthalmology, AI research on optometry has also increased. This review summarizes current AI research and technologies used for diagnosis in optometry, related to myopia, strabismus, amblyopia, optical glasses, contact lenses, and other aspects. The aim is to identify mature AI models that are suitable for research on optometry and potential algorithms that may be used in future clinical practice.

Explainable artificial intelligence in ophthalmology

PURPOSE OF REVIEW

Despite the growing scope of artificial intelligence (AI) and deep learning (DL) applications in the field of ophthalmology, most have yet to reach clinical adoption. Beyond model performance metrics, there has been an increasing emphasis on the need for explainability of proposed DL models.

RECENT FINDINGS

Several explainable AI (XAI) methods have been proposed, and increasingly applied in ophthalmological DL applications, predominantly in medical imaging analysis tasks.

SUMMARY

We summarize an overview of the key concepts, and categorize some examples of commonly employed XAI methods. Specific to ophthalmology, we explore XAI from a clinical perspective, in enhancing end-user trust, assisting clinical management, and uncovering new insights. We finally discuss its limitations and future directions to strengthen XAI for application to clinical practice.

A machine learning-based algorithm for estimating the original corneal curvature based on corneal topography after orthokeratology

OBJECTIVE

To estimate the original corneal curvature after orthokeratology by applying a machine learning-based algorithm.

METHODS

A total of 497 right eyes of 497 patients undergoing overnight orthokeratology for myopia for more than 1 year were enrolled in this retrospective study. All patients were fitted with lenses from Paragon CRT. Corneal topography was obtained by a Sirius corneal topography system (CSO, Italy). Original flat K (K1) and original steep K (K2) were set as the targets of calculation. The importance of each variable was explored by Fisher's criterion. Two machine learning models were established to allow adaptation to more situations. Bagging Tree, Gaussian process, support vector machine (SVM), and decision tree were used for prediction.

RESULTS

K2 after one year of orthokeratology (K2after) was most important in the prediction of K1 and K2. Bagging Tree performed best in both models 1 and 2 for K1 prediction (R = 0.812, RMSE = 0.855 in model 1 and R = 0.812, RMSE = 0.858 in model 2) and K2 prediction (R = 0.831, RMSE = 0.898 in model 1 and R = 0.837, RMSE = 0.888 in model 2). In model 1, the difference was 0.006 ± 1.34 D (p = 0.93) between the predictive value of K1 and the true value of K1 (K1before) and was 0.005 ± 1.51 D(p = 0.94) between the predictive value of K2 and the true value of K2 (K2before). In model 2, the difference was -0.056 ± 1.75 D (p = 0.59) between the predictive value of K1 and K1before and was 0.017 ± 2.01 D(p = 0.88) between the predictive value of K2 and K2before.

CONCLUSION

Bagging Tree performed best in predicting K1 and K2. Machine learning can be applied to predict the corneal curvature for those who cannot provide the initial corneal parameters in the outpatient clinic, providing a relatively certain degree of reference for the refitting of the Ortho-k lenses.

Predicting Postoperative Anterior Chamber Angle for Phakic Intraocular Lens Implantation Using Preoperative Anterior Segment Metrics

Purpose

The anterior chamber angle (ACA) is a critical factor in posterior chamber phakic intraocular lens (EVO Implantable Collamer Lens [ICL]) implantation. Herein, we predicted postoperative ACAs to select the optimal ICL size to reduce narrow ACA-related complications.

Methods

Regression models were constructed using pre-operative anterior segment optical coherence tomography metrics to predict postoperative ACAs, including trabecular-iris angles (TIAs) and scleral-spur angles (SSAs) at 500 µm and 750 µm from the scleral spur (TIA500, TIA750, SSA500, and SSA750). Data from three expert surgeons were assigned to the development (N = 430 eyes) and internal validation (N = 108 eyes) datasets. Additionally, data from a novice surgeon (N = 42 eyes) were used for external validation.

Results

Postoperative ACAs were highly predictable using the machine-learning (ML) technique (extreme gradient boosting regression [XGBoost]), with mean absolute errors (MAEs) of 4.42 degrees, 3.77 degrees, 5.25 degrees, and 4.30 degrees for TIA500, TIA750, SSA500, and SSA750, respectively, in internal validation. External validation also showed MAEs of 3.93 degrees, 3.86 degrees, 5.02 degrees, and 4.74 degrees for TIA500, TIA750, SSA500, and SSA750, respectively. Linear regression using the pre-operative anterior chamber depth, anterior chamber width, crystalline lens rise, TIA, and ICL size also exhibited good performance, with no significant difference compared with XGBoost in the validation sets.

Conclusions

We developed linear regression and ML models to predict postoperative ACAs for ICL surgery anterior segment metrics. These will prevent surgeons from overlooking the risks associated with the narrowing of the ACA.

Translational Relevance

Using the proposed algorithms, surgeons can consider the postoperative ACAs to increase surgical accuracy and safety.

Advances in artificial intelligence models and algorithms in the field of optometry

The rapid development of computer science over the past few decades has led to unprecedented progress in the field of artificial intelligence (AI). Its wide application in ophthalmology, especially image processing and data analysis, is particularly extensive and its performance excellent. In recent years, AI has been increasingly applied in optometry with remarkable results. This review is a summary of the application progress of different AI models and algorithms used in optometry (for problems such as myopia, strabismus, amblyopia, keratoconus, and intraocular lens) and includes a discussion of the limitations and challenges associated with its application in this field.

Application of explainable artificial intelligence for healthcare: A systematic review of the last decade (2011-2022)

BACKGROUND AND OBJECTIVES

Artificial intelligence (AI) has branched out to various applications in healthcare, such as health services management, predictive medicine, clinical decision-making, and patient data and diagnostics. Although AI models have achieved human-like performance, their use is still limited because they are seen as a black box. This lack of trust remains the main reason for their low use in practice, especially in healthcare. Hence, explainable artificial intelligence (XAI) has been introduced as a technique that can provide confidence in the model's prediction by explaining how the prediction is derived, thereby encouraging the use of AI systems in healthcare. The primary goal of this review is to provide areas of healthcare that require more attention from the XAI research community.

METHODS

Multiple journal databases were thoroughly searched using PRISMA guidelines 2020. Studies that do not appear in Q1 journals, which are highly credible, were excluded.

RESULTS

In this review, we surveyed 99 Q1 articles covering the following XAI techniques: SHAP, LIME, GradCAM, LRP, Fuzzy classifier, EBM, CBR, rule-based systems, and others.

CONCLUSION

We discovered that detecting abnormalities in 1D biosignals and identifying key text in clinical notes are areas that require more attention from the XAI research community. We hope this is review will encourage the development of a holistic cloud system for a smart city.

Classification of expert-level therapeutic decisions for degenerative cervical myelopathy using ensemble machine learning algorithms

Background

Therapeutic decisions for degenerative cervical myelopathy (DCM) are complex and should consider various factors. We aimed to develop machine learning (ML) models for classifying expert-level therapeutic decisions in patients with DCM.

Methods

This retrospective cross-sectional study included patients diagnosed with DCM, and the diagnosis of DCM was confirmed clinically and radiologically. The target outcomes were defined as conservative treatment, anterior surgical approaches (ASA), and posterior surgical approaches (PSA). We performed the following classifications using ML algorithms: multiclass, one-versus-rest, and one-versus-one. Two ensemble ML algorithms were used: random forest (RF) and extreme gradient boosting (XGB). The area under the receiver operating characteristic curve (AUC-ROC) was the primary metric. We also identified the variable importance for each classification.

Results

In total, 304 patients were included (109 conservative, 66 ASA, 125 PSA, and 4 combined surgeries). For multiclass classification, the AUC-ROC of RF and XGB models were 0.91 and 0.92, respectively. In addition, ML models showed AUC-ROC values of >0.9 for all types of binary classifications. Variable importance analysis revealed that the modified Japanese Orthopaedic Association score and central motor conduction time were the two most important variables for distinguishing between conservative and surgical treatments. When classifying ASA and PSA, the number of involved levels, age, and body mass index were important contributing factors.

Conclusion

ML-based classification of DCM therapeutic options is valid and feasible. This study can be a basis for establishing generalizable ML-based surgical decision models for DCM. Further studies are needed with a large multicenter database.

Ultrasound-based radiomics XGBoost model to assess the risk of central cervical lymph node metastasis in patients with papillary thyroid carcinoma: Individual application of SHAP

Objectives

A radiomics-based explainable eXtreme Gradient Boosting (XGBoost) model was developed to predict central cervical lymph node metastasis (CCLNM) in patients with papillary thyroid carcinoma (PTC), including positive and negative effects.

Methods

A total of 587 PTC patients admitted at Binzhou Medical University Hospital from 2017 to 2021 were analyzed retrospectively. The patients were randomized into the training and test cohorts with an 8:2 ratio. Radiomics features were extracted from ultrasound images of the primary PTC lesions. The minimum redundancy maximum relevance algorithm and the least absolute shrinkage and selection operator regression were used to select CCLNM positively-related features and radiomics scores were constructed. Clinical features, ultrasound features, and radiomics score were screened out by the Boruta algorithm, and the XGBoost model was constructed from these characteristics. SHapley Additive exPlanations (SHAP) was used for individualized and visualized interpretation. SHAP addressed the cognitive opacity of machine learning models.

Results

Eleven radiomics features were used to calculate the radiomics score. Five critical elements were used to build the XGBoost model: capsular invasion, radiomics score, diameter, age, and calcification. The area under the curve was 91.53% and 90.88% in the training and test cohorts, respectively. SHAP plots showed the influence of each parameter on the XGBoost model, including positive (i.e., capsular invasion, radiomics score, diameter, and calcification) and negative (i.e., age) impacts. The XGBoost model outperformed the radiologist, increasing the AUC by 44%.

Conclusions

The radiomics-based XGBoost model predicted CCLNM in PTC patients. Visual interpretation using SHAP made the model an effective tool for preoperative guidance of clinical procedures, including positive and negative impacts.

Machine learning predicting myopic regression after corneal refractive surgery using preoperative data and fundus photography

PURPOSE

Myopic regression after surgery is the most common long-term complication of refractive surgery, but it is difficult to identify myopic regression without long-term observation. This study aimed to develop machine learning models to identify high-risk patients for refractive regression based on preoperative data and fundus photography.

METHODS

This retrospective study assigned subjects to the training (n = 1606 eyes) and validation (n = 403 eyes) datasets with chronological data splitting. Machine learning models with ResNet50 (for image analysis) and XGBoost (for integration of all variables and fundus photography) were developed based on subjects who underwent corneal refractive surgery. The primary outcome was the predictive performance for the presence of myopic regression at 4 years of follow-up examination postoperatively.

RESULTS

By integrating all factors and fundus photography, the final combined machine learning model showed good performance to predict myopic regression of more than 0.5 D (area under the receiver operating characteristic curve [ROC-AUC], 0.753; 95% confidence interval [CI], 0.710-0.793). The performance of the final model was better than the single ResNet50 model only using fundus photography (ROC-AUC, 0.673; 95% CI, 0.627-0.716). The top-five most important input features were fundus photography, preoperative anterior chamber depth, planned ablation thickness, age, and preoperative central corneal thickness.

CONCLUSION

Our machine learning algorithm provides an efficient strategy to identify high-risk patients with myopic regression without additional labor, cost, and time. Surgeons might benefit from preoperative risk assessment of myopic regression, patient counseling before surgery, and surgical option decisions.

Applications of Artificial Intelligence in Myopia: Current and Future Directions

With the continuous development of computer technology, big data acquisition and imaging methods, the application of artificial intelligence (AI) in medical fields is expanding. The use of machine learning and deep learning in the diagnosis and treatment of ophthalmic diseases is becoming more widespread. As one of the main causes of visual impairment, myopia has a high global prevalence. Early screening or diagnosis of myopia, combined with other effective therapeutic interventions, is very important to maintain a patient's visual function and quality of life. Through the training of fundus photography, optical coherence tomography, and slit lamp images and through platforms provided by telemedicine, AI shows great application potential in the detection, diagnosis, progression prediction and treatment of myopia. In addition, AI models and wearable devices based on other forms of data also perform well in the behavioral intervention of myopia patients. Admittedly, there are still some challenges in the practical application of AI in myopia, such as the standardization of datasets; acceptance attitudes of users; and ethical, legal and regulatory issues. This paper reviews the clinical application status, potential challenges and future directions of AI in myopia and proposes that the establishment of an AI-integrated telemedicine platform will be a new direction for myopia management in the post-COVID-19 period.

Applications of Explainable Artificial Intelligence in Diagnosis and Surgery

In recent years, artificial intelligence (AI) has shown great promise in medicine. However, explainability issues make AI applications in clinical usages difficult. Some research has been conducted into explainable artificial intelligence (XAI) to overcome the limitation of the black-box nature of AI methods. Compared with AI techniques such as deep learning, XAI can provide both decision-making and explanations of the model. In this review, we conducted a survey of the recent trends in medical diagnosis and surgical applications using XAI. We have searched articles published between 2019 and 2021 from PubMed, IEEE Xplore, Association for Computing Machinery, and Google Scholar. We included articles which met the selection criteria in the review and then extracted and analyzed relevant information from the studies. Additionally, we provide an experimental showcase on breast cancer diagnosis, and illustrate how XAI can be applied in medical XAI applications. Finally, we summarize the XAI methods utilized in the medical XAI applications, the challenges that the researchers have met, and discuss the future research directions. The survey result indicates that medical XAI is a promising research direction, and this study aims to serve as a reference to medical experts and AI scientists when designing medical XAI applications.

Explainable artificial intelligence in forensics: Realistic explanations for number of contributor predictions of DNA profiles

Machine learning obtains good accuracy in determining the number of contributors (NOC) in short tandem repeat (STR) mixture DNA profiles. However, the models used so far are not understandable to users as they only output a prediction without any reasoning for that conclusion. Therefore, we leverage techniques from the field of explainable artificial intelligence (XAI) to help users understand why specific predictions are made. Where previous attempts at explainability for NOC estimation have relied upon using simpler, more understandable models that achieve lower accuracy, we use techniques that can be applied to any machine learning model. Our explanations incorporate SHAP values and counterfactual examples for each prediction into a single visualization. Existing methods for generating counterfactuals focus on uncorrelated features. This makes them inappropriate for the highly correlated features derived from STR data for NOC estimation, as these techniques simulate combinations of features that could not have resulted from an STR profile. For this reason, we have constructed a new counterfactual method, Realistic Counterfactuals (ReCo), which generates realistic counterfactual explanations for correlated data. We show that ReCo outperforms state-of-the-art methods on traditional metrics, as well as on a novel realism score. A user evaluation of the visualization shows positive opinions of end-users, which is ultimately the most appropriate metric in assessing explanations for real-world settings.

Noncontrast Cardiac Magnetic Resonance Imaging Predictors of Heart Failure Hospitalization in Heart Failure With Preserved Ejection Fraction

BACKGROUND

Heart failure patients with preserved ejection fraction (HFpEF) are at increased risk of future hospitalization. Contrast agents are often contra-indicated in HFpEF patients due to the high prevalence of concomitant kidney disease. Therefore, the prognostic value of a noncontrast cardiac magnetic resonance imaging (MRI) for HF-hospitalization is important.

PURPOSE

To develop and test an explainable machine learning (ML) model to investigate incremental value of noncontrast cardiac MRI for predicting HF-hospitalization.

STUDY TYPE

Retrospective.

POPULATION

A total of 203 HFpEF patients (mean, 64 ± 12 years, 48% women) referred for cardiac MRI were randomly split into training validation (143 patients, ~70%) and test sets (60 patients, ~30%).

FIELD STRENGTH

A 1.5 T, balanced steady-state free precession (bSSFP) sequence.

ASSESSMENT

Two ML models were built based on the tree boosting technique and the eXtreme Gradient Boosting model (XGBoost): 1) basic clinical ML model using clinical and echocardiographic data and 2) cardiac MRI-based ML model that included noncontrast cardiac MRI markers in addition to the basic model. The primary end point was defined as HF-hospitalization.

STATISTICAL TESTS

ML tool was used for advanced statistics, and the Elastic Net method for feature selection. Area under the receiver operating characteristic (ROC) curve (AUC) was compared between models using DeLong's test. To gain insight into the ML model, the SHapley Additive exPlanations (SHAP) method was leveraged. A P-value <0.05 was considered statistically significant.

RESULTS

During follow-up (mean, 50 ± 39 months), 85 patients (42%) reached the end point. The cardiac MRI-based ML model using the XGBoost algorithm provided a significantly superior prediction of HF-hospitalization (AUC: 0.81) compared to the basic model (AUC: 0.64). The SHAP analysis revealed left atrium (LA) and right atrium (RV) strains as top imaging markers contributing to its performance with cutoff values of 17.5% and -15%, respectively.

DATA CONCLUSIONS

Using an ML model, RV and LA strains measured in noncontrast cardiac MRI provide incremental value in predicting future hospitalization in HFpEF.

EVIDENCE LEVEL

3 TECHNICAL EFFICACY: Stage 2.

Artificial intelligence applications in different imaging modalities for corneal topography

Interpretation of topographical maps used to detect corneal ectasias requires a high level of expertise. Several artificial intelligence (AI) technologies have attempted to interpret topographic maps. The purpose of this study is to provide a review of AI algorithms in corneal topography from the perspectives of an eye care professional, a biomedical engineer, and a data scientist. A systematic literature review using Web of Science, Pubmed, and Google Scholar was performed from 2010 to 2020 on themes regarding imaging modalities, their parameters, purpose, and conclusions and their samples and performance related to AI in corneal topography. We provide a comprehensive summary of advances in corneal imaging and its applications in AI. Combined metrics from the Dual Scheimpflug and Placido device could be a good starting point to try AI models in corneal imaging systems. The range of area under the receiving operating curve for AI in keratoconus detection and classification was from 0.87 to 1, sensitivity was from 0.89 to 1, and specificity was from 0.82 to 1. A combination of different types of AI applications to corneal ectasia diagnosis is recommended.

Artificial Intelligence in Cornea, Refractive Surgery, and Cataract: Basic Principles, Clinical Applications, and Future Directions

ABSTRACT

Corneal diseases, uncorrected refractive errors, and cataract represent the major causes of blindness globally. The number of refractive surgeries, either cornea- or lens-based, is also on the rise as the demand for perfect vision continues to increase. With the recent advancement and potential promises of artificial intelligence (AI) technologies demonstrated in the realm of ophthalmology, particularly retinal diseases and glaucoma, AI researchers and clinicians are now channeling their focus toward the less explored ophthalmic areas related to the anterior segment of the eye. Conditions that rely on anterior segment imaging modalities, including slit-lamp photography, anterior segment optical coherence tomography, corneal tomography, in vivo confocal microscopy and/or optical biometers, are the most commonly explored areas. These include infectious keratitis, keratoconus, corneal grafts, ocular surface pathologies, preoperative screening before refractive surgery, intraocular lens calculation, and automated refraction, among others. In this review, we aimed to provide a comprehensive update on the utilization of AI in anterior segment diseases, with particular emphasis on the recent advancement in the past few years. In addition, we demystify some of the basic principles and terminologies related to AI, particularly machine learning and deep learning, to help improve the understanding, research and clinical implementation of these AI technologies among the ophthalmologists and vision scientists. As we march toward the era of digital health, guidelines such as CONSORT-AI, SPIRIT-AI, and STARD-AI will play crucial roles in guiding and standardizing the conduct and reporting of AI-related trials, ultimately promoting their potential for clinical translation.

Development of a Web-Based Ensemble Machine Learning Application to Select the Optimal Size of Posterior Chamber Phakic Intraocular Lens

Purpose

Selecting the optimal lens size by predicting the postoperative vault can reduce complications after implantation of an implantable collamer lens with a central-hole (ICL with KS-aquaport). We built a web-based machine learning application that incorporated clinical measurements to predict the postoperative ICL vault and select the optimal ICL size.

Methods

We applied the stacking ensemble technique based on eXtreme Gradient Boosting (XGBoost) and a light gradient boosting machine to pre-operative ocular data from two eye centers to predict the postoperative vault. We assigned the Korean patient data to a training (N = 2756 eyes) and internal validation (N = 693 eyes) datasets (prospective validation). Japanese patient data (N = 290 eyes) were used as an independent external dataset from different centers to validate the model.

Results

We developed an ensemble model that showed statistically better performance with a lower mean absolute error for ICL vault prediction (106.88 µm and 143.69 µm in the internal and external validation, respectively) than the other machine learning techniques and the classic ICL sizing methods did when applied to both validation datasets. Considering the lens size selection accuracy, our proposed method showed the best performance for both reference datasets (75.9% and 67.4% in the internal and external validation, respectively).

Conclusions

Applying the ensemble approach to a large dataset of patients who underwent ICL implantation resulted in a more accurate prediction of vault size and selection of the optimal ICL size.

Translational Relevance

We developed a web-based application for ICL sizing to facilitate the use of machine learning calculators for clinicians.

Computer-aided detection and abnormality score for the outer retinal layer in optical coherence tomography

BACKGROUND

To develop computer-aided detection (CADe) of ORL abnormalities in the retinal pigmented epithelium, interdigitation zone and ellipsoid zone via optical coherence tomography (OCT).

METHODS

In this retrospective study, healthy participants with normal ORL, and patients with abnormality of ORL including choroidal neovascularisation (CNV) or retinitis pigmentosa (RP) were included. First, an automatic segmentation deep learning (DL) algorithm, CADe, was developed for the three outer retinal layers using 120 handcraft masks of ORL. This automatic segmentation algorithm generated 4000 segmentations, which included 2000 images with normal ORL and 2000 (1000 CNV and 1000 RP) images with focal or wide defects in ORL. Second, based on the automatically generated segmentation images, a binary classifier (normal vs abnormal) was developed. Results were evaluated by area under the receiver operating characteristic curve (AUC).

RESULTS

The DL algorithm achieved an AUC of 0.984 (95% CI 0.976 to 0.993) for individual image evaluation in the internal test set of 797 images. In addition, performance analysis of a publicly available external test set (n=968) had an AUC of 0.957 (95% CI 0.944 to 0.970) and a second clinical external test set (n=1124) had an AUC of 0.978 (95% CI 0.970 to 0.986). Moreover, the CADe highlighted well normal parts of ORL and omitted highlights in abnormal ORLs of CNV and RP.

CONCLUSION

The CADe can use OCT images to segment ORL and differentiate between normal ORL and abnormal ORL. The CADe classifier also performs visualisation and may aid future physician diagnosis and clinical applications.

Feasibility study to improve deep learning in OCT diagnosis of rare retinal diseases with few-shot classification

Deep learning (DL) has been successfully applied to the diagnosis of ophthalmic diseases. However, rare diseases are commonly neglected due to insufficient data. Here, we demonstrate that few-shot learning (FSL) using a generative adversarial network (GAN) can improve the applicability of DL in the optical coherence tomography (OCT) diagnosis of rare diseases. Four major classes with a large number of datasets and five rare disease classes with a few-shot dataset are included in this study. Before training the classifier, we constructed GAN models to generate pathological OCT images of each rare disease from normal OCT images. The Inception-v3 architecture was trained using an augmented training dataset, and the final model was validated using an independent test dataset. The synthetic images helped in the extraction of the characteristic features of each rare disease. The proposed DL model demonstrated a significant improvement in the accuracy of the OCT diagnosis of rare retinal diseases and outperformed the traditional DL models, Siamese network, and prototypical network. By increasing the accuracy of diagnosing rare retinal diseases through FSL, clinicians can avoid neglecting rare diseases with DL assistance, thereby reducing diagnosis delay and patient burden.

Toward automated severe pharyngitis detection with smartphone camera using deep learning networks

PURPOSE

Severe pharyngitis is frequently associated with inflammations caused by streptococcal pharyngitis, which can cause immune-mediated and post-infectious complications. The recent global pandemic of coronavirus disease (COVID-19) encourages the use of telemedicine for patients with respiratory symptoms. This study, therefore, purposes automated detection of severe pharyngitis using a deep learning framework with self-taken throat images.

METHODS

A dataset composed of two classes of 131 throat images with pharyngitis and 208 normal throat images was collected. Before the training classifier, we constructed a cycle consistency generative adversarial network (CycleGAN) to augment the training dataset. The ResNet50, Inception-v3, and MobileNet-v2 architectures were trained with transfer learning and validated using a randomly selected test dataset. The performance of the models was evaluated based on the accuracy and area under the receiver operating characteristic curve (ROC-AUC).

RESULTS

The CycleGAN-based synthetic images reflected the pragmatic characteristic features of pharyngitis. Using the synthetic throat images, the deep learning model demonstrated a significant improvement in the accuracy of the pharyngitis diagnosis. ResNet50 with GAN-based augmentation showed the best ROC-AUC of 0.988 for pharyngitis detection in the test dataset. In the 4-fold cross-validation using the ResNet50, the highest detection accuracy and ROC-AUC achieved were 95.3% and 0.992, respectively.

CONCLUSION

The deep learning model for smartphone-based pharyngitis screening allows fast identification of severe pharyngitis with a potential of the timely diagnosis of pharyngitis. In the recent pandemic of COVID-19, this framework will help patients with upper respiratory symptoms to improve convenience in diagnosis and reduce transmission.