Prediction of visual impairment in retinitis pigmentosa using deep learning and multimodal fundus images

Künstliche Intelligenz (KI) zur Früherkennung von Netzhauterkrankungen

Künstliche Intelligenz (KI) hat sich zu einem transformativen Werkzeug auf dem Gebiet der Augenheilkunde entwickelt und revolutioniert die Diagnose und Behandlung von Krankheiten. Diese Arbeit gibt einen umfassenden Überblick über KI-Anwendungen bei verschiedenen Netzhauterkrankungen und zeigt ihr Potenzial, die Effizienz von Vorsorgeuntersuchungen zu erhöhen, Frühdiagnosen zu erleichtern und die Patientenergebnisse zu verbessern. Wir erklären die grundlegenden Konzepte der KI, einschließlich des maschinellen Lernens (ML) und des Deep Learning (DL), und deren Anwendung in der Augenheilkunde und heben die Bedeutung von KI-basierten Lösungen bei der Bewältigung der Komplexität und Variabilität von Netzhauterkrankungen hervor. Wir gehen auch auf spezifische Anwendungen der KI im Zusammenhang mit Netzhauterkrankungen wie diabetischer Retinopathie (DR), altersbedingter Makuladegeneration (AMD), makulärer Neovaskularisation, Frühgeborenen-Retinopathie (ROP), retinalem Venenverschluss (RVO), hypertensiver Retinopathie (HR), Retinopathia pigmentosa, Morbus Stargardt, Morbus Best (Best’sche vitelliforme Makuladystrophie) und Sichelzellenretinopathie ein. Wir konzen­trieren uns auf die aktuelle Landschaft der KI-Technologien, einschließlich verschiedener KI-Modelle, ihrer Leistungsmetriken und klinischen Implikationen. Darüber hinaus befassen wir uns mit den Herausforderungen und Schwierigkeiten bei der Integration von KI in die klinische Praxis, einschließlich des «Black-Box-Phänomens», der Verzerrungen bei der Darstellung von Daten und der Einschränkungen im Zusammenhang mit der ganzheitlichen Bewertung von Patienten. Abschließend wird die kollaborative Rolle der KI an der Seite des medizinischen Fachpersonals hervorgehoben, wobei ein synergetischer Ansatz für die Erbringung von Gesundheitsdienstleistungen befürwortet wird. Es wird betont, wie wichtig es ist, KI als Ergänzung und nicht als Ersatz für menschliche Expertise einzusetzen, um ihr Potenzial zu maximieren, die Gesundheitsversorgung zu revolutionieren, Ungleichheiten in der Gesundheitsversorgung zu verringern und die Patientenergebnisse in der sich entwickelnden medizinischen Landschaft zu verbessern.

Applications of artificial intelligence to inherited retinal diseases: A systematic review

Artificial intelligence(AI)-based methods have been extensively used for the detection and management of various common retinal conditions, but their targeted development for inherited retinal diseases (IRD) is still nascent. In the context of limited availability of retinal subspecialists, genetic testing and genetic counseling, there is a high need for accurate and accessible diagnostic methods. The currently available AI studies, aiming for detection, classification, and prediction of IRD, remain mainly retrospective and include relatively limited numbers of patients due to their scarcity. We summarize the latest findings and clinical implications of machine-learning algorithms in IRD, highlighting the achievements and challenges of AI to assist ophthalmologists in their clinical practice.

Classification of fundus autofluorescence images based on macular function in retinitis pigmentosa using convolutional neural networks

PURPOSE

To determine whether convolutional neural networks (CNN) can classify the severity of central vision loss using fundus autofluorescence (FAF) images and color fundus images of retinitis pigmentosa (RP), and to evaluate the utility of those images for severity classification.

STUDY DESIGN

Retrospective observational study.

METHODS

Medical charts of patients with RP who visited Nagoya University Hospital were reviewed. Eyes with atypical RP or previous surgery were excluded. The mild group was comprised of patients with a mean deviation value of > - 10 decibels, and the severe group of < - 20 decibels, in the Humphrey field analyzer 10-2 program. CNN models were created by transfer learning of VGG16 pretrained with ImageNet to classify patients as either mild or severe, using FAF images or color fundus images.

RESULTS

Overall, 165 patients were included in this study; 80 patients were classified into the severe and 85 into the mild group. The test data comprised 40 patients in each group, and the images of the remaining patients were used as training data, with data augmentation by rotation and flipping. The highest accuracies of the CNN models when using color fundus and FAF images were 63.75% and 87.50%, respectively.

CONCLUSION

Using FAF images may enable the accurate assessment of central vision function in RP. FAF images may include more parameters than color fundus images that can evaluate central visual function.

When do patients with retinitis pigmentosa present to ophthalmologists? A multi-centre retrospective study

BACKGROUND

Planned gene therapies for retinitis pigmentosa (RP) depend on viable photoreceptors for efficacy. Understanding disease severity at presentation, and drivers that influence time to presentation is important when planning interventions. We examined features that influence RP severity at initial presentation.

METHODS

Multi-centre retrospective cohort study of RP patients at initial presentation. Disease severity was scored using ellipsoid zone (EZ) width on SD-OCT and logistic regression used to determine risk factors for advanced disease at presentation.

RESULTS

A total of 146 unrelated RP patients were included. Median age at onset and presentation was 40.5 (range 1-74) and 50.1 (range 3.9-81.8), respectively. Severe disease (<5° of remaining EZ width) was present in 28.1% of cases at presentation. Patients with family history of RP had greater odds of severe disease (OR 3.29, 95% CI 1.56, 6.95; p = 0.002), while male gender, race, age, syndromic features, and socioeconomic status did not. Patients with affected siblings (median EZ width 6.2°; p = 0.01), but not affected parents (median EZ width 9.4°; p = 0.99), presented with severe EZ loss compared to patients without family history (median EZ width 13.1°). Patients with affected siblings had delayed presentation (≥5 years; OR 5.76, 95% CI 1.817, 18.262; p = 0.003) compared to patients without family history.

CONCLUSIONS

Family history influences the stage of disease at which RP patients initially seek ophthalmology review. This has implications for patient counselling and the number of patients who may benefit from future therapies.

Deep learning aided measurement of outer retinal layer metrics as biomarkers for inherited retinal degenerations: opportunities and challenges

PURPOSE OF REVIEW

The purpose of this review was to provide a summary of currently available retinal imaging and visual function testing methods for assessing inherited retinal degenerations (IRDs), with the emphasis on the application of deep learning (DL) approaches to assist the determination of structural biomarkers for IRDs.

RECENT FINDINGS

(clinical trials for IRDs; discover effective biomarkers as endpoints; DL applications in processing retinal images to detect disease-related structural changes).

SUMMARY

Assessing photoreceptor loss is a direct way to evaluate IRDs. Outer retinal layer structures, including outer nuclear layer, ellipsoid zone, photoreceptor outer segment, RPE, are potential structural biomarkers for IRDs. More work may be needed on structure and function relationship.

Determinants of Visual Impairment Among Chinese Middle-Aged and Older Adults: Risk Prediction Model Using Machine Learning Algorithms

Background

Visual impairment (VI) is a prevalent global health issue, affecting over 2.2 billion people worldwide, with nearly half of the Chinese population aged 60 years and older being affected. Early detection of high-risk VI is essential for preventing irreversible vision loss among Chinese middle-aged and older adults. While machine learning (ML) algorithms exhibit significant predictive advantages, their application in predicting VI risk among the general middle-aged and older adult population in China remains limited.

Objective

This study aimed to predict VI and identify its determinants using ML algorithms.

Methods

We used 19,047 participants from 4 waves of the China Health and Retirement Longitudinal Study (CHARLS) that were conducted between 2011 and 2018. To envisage the prevalence of VI, we generated a geographical distribution map. Additionally, we constructed a model using indicators of a self-reported questionnaire, a physical examination, and blood biomarkers as predictors. Multiple ML algorithms, including gradient boosting machine, distributed random forest, the generalized linear model, deep learning, and stacked ensemble, were used for prediction. We plotted receiver operating characteristic and calibration curves to assess the predictive performance. Variable importance analysis was used to identify key predictors.

Results

Among all participants, 33.9% (6449/19,047) had VI. Qinghai, Chongqing, Anhui, and Sichuan showed the highest VI rates, while Beijing and Xinjiang had the lowest. The generalized linear model, gradient boosting machine, and stacked ensemble achieved acceptable area under curve values of 0.706, 0.710, and 0.715, respectively, with the stacked ensemble performing best. Key predictors included hearing impairment, self-expectation of health status, pain, age, hand grip strength, depression, night sleep duration, high-density lipoprotein cholesterol, and arthritis or rheumatism.

Conclusions

Nearly one-third of middle-aged and older adults in China had VI. The prevalence of VI shows regional variations, but there are no distinct east-west or north-south distribution differences. ML algorithms demonstrate accurate predictive capabilities for VI. The combination of prediction models and variable importance analysis provides valuable insights for the early identification and intervention of VI among Chinese middle-aged and older adults.

Multimodality Fusion Strategies in Eye Disease Diagnosis

Multimodality fusion has gained significance in medical applications, particularly in diagnosing challenging diseases like eye diseases, notably diabetic eye diseases that pose risks of vision loss and blindness. Mono-modality eye disease diagnosis proves difficult, often missing crucial disease indicators. In response, researchers advocate multimodality-based approaches to enhance diagnostics. This study is a unique exploration, evaluating three multimodality fusion strategies-early, joint, and late-in conjunction with state-of-the-art convolutional neural network models for automated eye disease binary detection across three datasets: fundus fluorescein angiography, macula, and combination of digital retinal images for vessel extraction, structured analysis of the retina, and high-resolution fundus. Findings reveal the efficacy of each fusion strategy: type 0 early fusion with DenseNet121 achieves an impressive 99.45% average accuracy. InceptionResNetV2 emerges as the top-performing joint fusion architecture with an average accuracy of 99.58%. Late fusion ResNet50V2 achieves a perfect score of 100% across all metrics, surpassing both early and joint fusion. Comparative analysis demonstrates that late fusion ResNet50V2 matches the accuracy of state-of-the-art feature-level fusion model for multiview learning. In conclusion, this study substantiates late fusion as the optimal strategy for eye disease diagnosis compared to early and joint fusion, showcasing its superiority in leveraging multimodal information.

Cataract surgery in retinitis pigmentosa

Background

Retinitis pigmentosa (RP) is an inherited retinal dystrophy characterized by progressive vision loss due to photoreceptor degeneration. Complicated cataract formation, particularly posterior subcapsular cataract (PSC), frequently occurs in RP and exacerbates the visual impairment. Cataract surgery may improve vision; however, the distinctive challenges of RP require specific considerations. This mini-review aims to provide a comprehensive overview of the RP-related cataract.

Methods

A comprehensive literature review was conducted via PubMed/MEDLINE, spanning the period from January 1976 to June 2024, using the keywords "cataract," "cataract surgery," "cystoid macular edema," "hereditary retinal dystrophy," "retinitis pigmentosa," "posterior subcapsular cataract," "posterior capsular opacification," "zonular weakness," and "artificial intelligence." We aimed to evaluate cataract surgery in patients with RP, focusing on cataract formation, its surgical management, postoperative complications, patient follow-up, and visual outcomes. Relevant review articles, clinical trials, and case reports with related reference lists of these articles were included.

Results

A total of 53 articles were examined in detail, including those identified through focused keyword searches and the reference lists of these articles. Cataract surgery in patients with RP generally results in substantial visual improvement. However, surgery can be complicated, particularly by zonular weakness and subluxation of the crystalline lens. These risks can be reduced by using capsular tension rings and employing meticulous surgical technique. Furthermore, postoperative complications, such as cystoid macular edema and posterior capsular opacification, are common. Despite these challenges, regular postoperative follow-up and appropriate management can help mitigate complications. Integrity of the ellipsoid zone and external limiting membrane on preoperative optical coherence tomographic examination are the main predictors of visual outcomes following cataract surgery; however, outcomes can vary. Though many patients experience significant visual improvement, some may experience limited benefits due to pre-existing advanced retinal degeneration.

Conclusions

Cataract surgery may offer meaningful visual benefits in patients with RP; however, careful preoperative evaluation and meticulous surgical technique are required to address the possible challenges. Attentive postoperative care and follow-up are essential to optimize visual outcomes. Early surgical intervention can significantly improve the quality of life in selected candidates, and tailored approaches are necessary in patients with RP requiring cataract surgery. Further studies on the potential application of artificial intelligence to monitor postoperative recovery and detect complications may improve surgical outcomes and enhance patient care.

Correlations between the Michigan Retinal Degeneration Questionnaire and visual function parameters in patients with retinitis pigmentosa

PURPOSE

To validate the use of best-corrected visual acuity (BCVA), low-luminance visual acuity (LLVA), low-luminance deficit (LLD; the difference between BCVA and LLVA), mean macular sensitivity and fixation stability as parameters of vision-related quality of life based on the novel Michigan Retinal Degeneration Questionnaire (MRDQ) in retinitis pigmentosa (RP) patients.

METHODS

In this prospective cross sectional study, 30 patients with RP (47% female) were included with a median age of 41.0 years (interquartile range: 24.1-58.3 years). BCVA, LLVA and LLD were measured with Early Treatment Diabetic Retinopathy Study (ETDRS) charts. Mesopic microperimetry was performed to measure mean macular sensitivity and fixation stability. Patients completed a Dutch translation of the MRDQ which results in an experienced disability (Θ-)score of seven domains. Spearman's rank correlation was used.

RESULTS

BCVA correlated significantly to the MRDQ domain of central vision (r = 0.657; p < 0.001) and colour vision (r = 0.524; p = 0.003). Lower LLVA significantly correlated to higher experienced disability in the MRDQ domains for central vision (=0.550; p = 0.002) and contrast sensitivity (r = 0.502; p = 0.005). LLD was significantly correlated to the MRDQ domains of scotopic function (r = -0.484; p = 0.007) and mesopic peripheral function (r = -0.533; p = 0.002). Lower mean macular sensitivity was significantly associated with high experienced disability in all domains except for photosensitivity.

CONCLUSIONS

The majority of the MRDQ domains is strongly associated with visual function parameters. These findings show that visual function measurements, especially LLVA, LLD and mean macular sensitivity on microperimetry, reflect vision-related quality of life and can be used as relevant outcome measures in clinical trials for RP.

Artificial Intelligence (AI) for Early Diagnosis of Retinal Diseases

Artificial intelligence (AI) has emerged as a transformative tool in the field of ophthalmology, revolutionizing disease diagnosis and management. This paper provides a comprehensive overview of AI applications in various retinal diseases, highlighting its potential to enhance screening efficiency, facilitate early diagnosis, and improve patient outcomes. Herein, we elucidate the fundamental concepts of AI, including machine learning (ML) and deep learning (DL), and their application in ophthalmology, underscoring the significance of AI-driven solutions in addressing the complexity and variability of retinal diseases. Furthermore, we delve into the specific applications of AI in retinal diseases such as diabetic retinopathy (DR), age-related macular degeneration (AMD), Macular Neovascularization, retinopathy of prematurity (ROP), retinal vein occlusion (RVO), hypertensive retinopathy (HR), Retinitis Pigmentosa, Stargardt disease, best vitelliform macular dystrophy, and sickle cell retinopathy. We focus on the current landscape of AI technologies, including various AI models, their performance metrics, and clinical implications. Furthermore, we aim to address challenges and pitfalls associated with the integration of AI in clinical practice, including the "black box phenomenon", biases in data representation, and limitations in comprehensive patient assessment. In conclusion, this review emphasizes the collaborative role of AI alongside healthcare professionals, advocating for a synergistic approach to healthcare delivery. It highlights the importance of leveraging AI to augment, rather than replace, human expertise, thereby maximizing its potential to revolutionize healthcare delivery, mitigate healthcare disparities, and improve patient outcomes in the evolving landscape of medicine.

Estimation of Visual Function Using Deep Learning From Ultra-Widefield Fundus Images of Eyes With Retinitis Pigmentosa

Importance

There is no widespread effective treatment to halt the progression of retinitis pigmentosa. Consequently, adequate assessment and estimation of residual visual function are important clinically.

Objective

To examine whether deep learning can accurately estimate the visual function of patients with retinitis pigmentosa by using ultra-widefield fundus images obtained on concurrent visits.

Design, Setting, and Participants

Data for this multicenter, retrospective, cross-sectional study were collected between January 1, 2012, and December 31, 2018. This study included 695 consecutive patients with retinitis pigmentosa who were examined at 5 institutions. Each of the 3 types of input images-ultra-widefield pseudocolor images, ultra-widefield fundus autofluorescence images, and both ultra-widefield pseudocolor and fundus autofluorescence images-was paired with 1 of the 31 types of ensemble models constructed from 5 deep learning models (Visual Geometry Group-16, Residual Network-50, InceptionV3, DenseNet121, and EfficientNetB0). We used 848, 212, and 214 images for the training, validation, and testing data, respectively. All data from 1 institution were used for the independent testing data. Data analysis was performed from June 7, 2021, to December 5, 2022.

Main Outcomes and Measures

The mean deviation on the Humphrey field analyzer, central retinal sensitivity, and best-corrected visual acuity were estimated. The image type-ensemble model combination that yielded the smallest mean absolute error was defined as the model with the best estimation accuracy. After removal of the bias of including both eyes with the generalized linear mixed model, correlations between the actual values of the testing data and the estimated values by the best accuracy model were examined by calculating standardized regression coefficients and P values.

Results

The study included 1274 eyes of 695 patients. A total of 385 patients were female (55.4%), and the mean (SD) age was 53.9 (17.2) years. Among the 3 types of images, the model using ultra-widefield fundus autofluorescence images alone provided the best estimation accuracy for mean deviation, central sensitivity, and visual acuity. Standardized regression coefficients were 0.684 (95% CI, 0.567-0.802) for the mean deviation estimation, 0.697 (95% CI, 0.590-0.804) for the central sensitivity estimation, and 0.309 (95% CI, 0.187-0.430) for the visual acuity estimation (all P < .001).

Conclusions and Relevance

Results of this study suggest that the visual function estimation in patients with retinitis pigmentosa from ultra-widefield fundus autofluorescence images using deep learning might help assess disease progression objectively. Findings also suggest that deep learning models might monitor the progression of retinitis pigmentosa efficiently during follow-up.