Evidence-based Canadian guidelines for tele-retina screening for diabetic retinopathy: recommendations from the Canadian Retina Research Network (CR2N) Tele-Retina Steering Committee

A systematic review of generalization research in medical image classification

Numerous Deep Learning (DL) classification models have been developed for a large spectrum of medical image analysis applications, which promises to reshape various facets of medical practice. Despite early advances in DL model validation and implementation, which encourage healthcare institutions to adopt them, a fundamental questions remain: how can these models effectively handle domain shift? This question is crucial to limit DL models performance degradation. Medical data are dynamic and prone to domain shift, due to multiple factors. Two main shift types can occur over time: (1) covariate shift mainly arising due to updates to medical equipment and (2) concept shift caused by inter-grader variability. To mitigate the problem of domain shift, existing surveys mainly focus on domain adaptation techniques, with an emphasis on covariate shift. More generally, no work has reviewed the state-of-the-art solutions while focusing on the shift types. This paper aims to explore existing domain generalization methods for DL-based classification models through a systematic review of literature. It proposes a taxonomy based on the shift type they aim to solve. Papers were searched and gathered on Scopus till 10 April 2023, and after the eligibility screening and quality evaluation, 77 articles were identified. Exclusion criteria included: lack of methodological novelty (e.g., reviews, benchmarks), experiments conducted on a single mono-center dataset, or articles not written in English. The results of this paper show that learning based methods are emerging, for both shift types. Finally, we discuss future challenges, including the need for improved evaluation protocols and benchmarks, and envisioned future developments to achieve robust, generalized models for medical image classification.

Implementation of Artificial Intelligence-Based Diabetic Retinopathy Screening in a Tertiary Care Hospital in Quebec: Prospective Validation Study

BACKGROUND

Diabetic retinopathy (DR) affects about 25% of people with diabetes in Canada. Early detection of DR is essential for preventing vision loss.

OBJECTIVE

We evaluated the real-world performance of an artificial intelligence (AI) system that analyzes fundus images for DR screening in a Quebec tertiary care center.

METHODS

We prospectively recruited adult patients with diabetes at the Centre hospitalier de l'Université de Montréal (CHUM) in Montreal, Quebec, Canada. Patients underwent dual-pathway screening: first by the Computer Assisted Retinal Analysis (CARA) AI system (index test), then by standard ophthalmological examination (reference standard). We measured the AI system's sensitivity and specificity for detecting referable disease at the patient level, along with its performance for detecting any retinopathy and diabetic macular edema (DME) at the eye level, and potential cost savings.

RESULTS

This study included 115 patients. CARA demonstrated a sensitivity of 87.5% (95% CI 71.9-95.0) and specificity of 66.2% (95% CI 54.3-76.3) for detecting referable disease at the patient level. For any retinopathy detection at the eye level, CARA showed 88.2% sensitivity (95% CI 76.6-94.5) and 71.4% specificity (95% CI 63.7-78.1). For DME detection, CARA had 100% sensitivity (95% CI 64.6-100) and 81.9% specificity (95% CI 75.6-86.8). Potential yearly savings from implementing CARA at the CHUM were estimated at CAD $245,635 (US $177,643.23, as of July 26, 2024) considering 5000 patients with diabetes.

CONCLUSIONS

Our study indicates that integrating a semiautomated AI system for DR screening demonstrates high sensitivity for detecting referable disease in a real-world setting. This system has the potential to improve screening efficiency and reduce costs at the CHUM, but more work is needed to validate it.

MAPLES-DR: MESSIDOR Anatomical and Pathological Labels for Explainable Screening of Diabetic Retinopathy

Reliable automatic diagnosis of Diabetic Retinopathy (DR) and Macular Edema (ME) is an invaluable asset in improving the rate of monitored patients among at-risk populations and in enabling earlier treatments before the pathology progresses and threatens vision. However, the explainability of screening models is still an open question, and specifically designed datasets are required to support the research. We present MAPLES-DR (MESSIDOR Anatomical and Pathological Labels for Explainable Screening of Diabetic Retinopathy), which contains, for 198 images of the MESSIDOR public fundus dataset, new diagnoses for DR and ME as well as new pixel-wise segmentation maps for 10 anatomical and pathological biomarkers related to DR. This paper documents the design choices and the annotation procedure that produced MAPLES-DR, discusses the interobserver variability and the overall quality of the annotations, and provides guidelines on using the dataset in a machine learning context.

Health transition and eye care policy planning for people with diabetic retinopathy in south Asia

The prevalence of type 2 diabetes (T2D), associated systemic disorders, diabetic retinopathy (DR) and current health policies in south Asian countries were analysed to assess country-specific preparedness to meet the 2030 Sustainable Development Goals. The south Asian countries were classified by human development index, socio-demographic index, multidimensional poverty indices, and eye health resources for epidemiological resource-level analysis. In south Asia, the prevalence of diagnosed and undiagnosed T2D in adults aged 40 years or above, was higher in Pakistan (26.3%) and Afghanistan (71.4%), respectively; India has the highest absolute number of people with DR, and Afghanistan has the highest prevalence of DR (50.6%). In this region, out-of-pocket spending is high (∼77%). This Health Policy is a situational analysis of data available on the prevalence of DR and common eye diseases in people with T2D in south Asia and available resources to suggest tailored health policies to local needs. The common issues in the region are insufficient human resources for eye health, unequal distribution of available workforce, and inadequate infrastructure. Addressing these challenges of individuals with T2D and DR, a 10-point strategy is suggested to improve infrastructure, augment human resources, reduce out-of-pocket spending, employ targeted screening, and encourage public-private partnerships.

Diabetic Disease of the Eye in Canada: Consensus Statements from a Retina Specialist Working Group

Despite advances in systemic care, diabetic disease of the eye (DDE) remains the leading cause of blindness worldwide. There is a critical gap of up-to-date, evidence-based guidance for ophthalmologists in Canada that includes evidence from recent randomized controlled trials. Previous guidance has not always given special consideration to applying treatments and managing DDE in the context of the healthcare system. This consensus statement aims to assist practitioners in the field by providing a spectrum of acceptable opinions on DDE treatment and management from recognized experts in the field. In compiling evidence and generating consensus, a working group of retinal specialists in Canada addressed clinical questions surrounding the four themes of disease, patient, management, and collaboration. The working group reviewed literature representing the highest level of evidence on DDE and shared their opinions on topics surrounding the epidemiology and pathophysiology of diabetic retinopathy and diabetic macular edema; diagnosis and monitoring; considerations around diabetes medication use; strategic considerations for management given systemic comorbidities, ocular comorbidities, and pregnancy; treatment goals and modalities for diabetic macular edema, non-proliferative and proliferative diabetic retinopathy, and retinal detachment; and interdisciplinary collaboration. Ultimately, this work highlighted that the retinal examination in DDE not only informs the treating ophthalmologist but can serve as a global index for disease progression across many tissues of the body. It highlighted further that DDE can be treated regardless of diabetic control, that a systemic approach to patient care will result in the best health outcomes, and prevention of visual complications requires a multidisciplinary management approach. Ophthalmologists must tailor their clinical approach to the needs and circumstances of individual patients and work within the realities of their healthcare setting.

Guidelines, Consensus Statements, and Standards for the Use of Artificial Intelligence in Medicine: Systematic Review

BACKGROUND

The application of artificial intelligence (AI) in the delivery of health care is a promising area, and guidelines, consensus statements, and standards on AI regarding various topics have been developed.

OBJECTIVE

We performed this study to assess the quality of guidelines, consensus statements, and standards in the field of AI for medicine and to provide a foundation for recommendations about the future development of AI guidelines.

METHODS

We searched 7 electronic databases from database establishment to April 6, 2022, and screened articles involving AI guidelines, consensus statements, and standards for eligibility. The AGREE II (Appraisal of Guidelines for Research & Evaluation II) and RIGHT (Reporting Items for Practice Guidelines in Healthcare) tools were used to assess the methodological and reporting quality of the included articles.

RESULTS

This systematic review included 19 guideline articles, 14 consensus statement articles, and 3 standard articles published between 2019 and 2022. Their content involved disease screening, diagnosis, and treatment; AI intervention trial reporting; AI imaging development and collaboration; AI data application; and AI ethics governance and applications. Our quality assessment revealed that the average overall AGREE II score was 4.0 (range 2.2-5.5; 7-point Likert scale) and the mean overall reporting rate of the RIGHT tool was 49.4% (range 25.7%-77.1%).

CONCLUSIONS

The results indicated important differences in the quality of different AI guidelines, consensus statements, and standards. We made recommendations for improving their methodological and reporting quality.

TRIAL REGISTRATION

PROSPERO International Prospective Register of Systematic Reviews (CRD42022321360); https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=321360.

Prevalence and predictors for being unscreened for diabetic retinopathy: a population-based study over a decade

OBJECTIVE

To determine the population-level predictors for being unscreened for diabetic retinopathy (DR) among individuals with diabetes in a developed country.

DESIGN

A retrospective population-based repeated-cross-sectional study.

PARTICIPANTS

All individuals with diabetes (types 1 and 2) aged ≥20 years in the universal health care system in Ontario were identified in the 2011-2013 and 2017-2019 time periods.

METHODS

The Mantel-Haenszel test was used for the relative risk (RR) comparison of subcategories stratified by the 2 cross-sectional time periods.

RESULTS

A total of 1 145 645 and 1 346 578 individuals with diabetes were identified in 2011-2013 and 2017-2019, respectively. The proportion of patients unscreened for DR declined very slightly from 35% (n = 405 967) in 2011-2013 to 34% (n = 455 027) in 2017-2019 of the population with diabetes (RR = 0.967; 95% CI, 0.964-0.9693; p < 0.0001). Young adults aged 20-39 years of age had the highest proportion of unscreened patients (62% and 58% in 2011-2013 and 2017-2019, respectively). Additionally, those who had a lower income quintile (RR = 1.039; 95% CI, 1.036-1.044; p < 0.0001), were recent immigrants (RR = 1.286; 95% CI, 1.280-1.293; p < 0.0001), lived in urban areas (RR = 1.149; 95% CI, 1.145-1.154; p < 0.0001), had a mental health history (RR = 1.117; 95% CI, 1.112-1.122; p < 0.0001), or lacked a connection to a primary care provider (RR = 1.656; 95% CI, 1.644-1.668; p < 0.0001) had a higher risk of being unscreened.

CONCLUSIONS

This population-based study suggests that over 1 decade, 33% of individuals with diabetes are unscreened for DR, and young age, low income, immigration, residing in a large city, mental health illness, and no primary care access are the main predictors.

Care Coordination of Older Adults With Diabetes: A Scoping Review

OBJECTIVES

Care coordination is a common intervention to support older adults with diabetes and their caregivers, and provides individualized, integrated health and social care. However, the optimal approach of care coordination is not well described. In this scoping review we synthesized evidence regarding the implementation of traditional and virtual care coordination for older adults with diabetes to inform future research and best practices.

METHODS

The Joanna Briggs Institute scoping review methods were used. A systematic search was conducted in CINAHL, Embase, EmCare, and Medline, as well as a targeted grey literature search, and a hand-search of reference lists. Screening and data extraction were completed by 3 independent reviewers.

RESULTS

Forty-two articles were included in the synthesis. Included studies operationalized care coordination in different ways. The most commonly implemented elements of care coordination were regular communication and monitoring. In contrast, coordination between health-care teams and the community, individualized planning, and caregiver involvement were less often reported. Outcomes to evaluate the impact of care coordination were predominantly diabetes-centric, and less often person-centred. In addition, evidence indicates that older adults value a trusting relationship with their care coordinator.

CONCLUSIONS

Studies assessing care coordination for older adults with diabetes have shown positive outcomes. To inform best practices, future intervention research for this population should focus on evaluating the impact of comprehensive care planning, system navigation across the health and social care sectors, the care coordinator and patient relationship and caregiver support.

Handheld fundus camera performance, image quality and outcomes of diabetic retinopathy grading in a pilot screening study

PURPOSE

To compare the performance and image quality of the handheld fundus camera to standard table-top fundus cameras in diabetic retinopathy (DR) screening. The reliability and diagnostic accuracy of DR grading performed by an ophthalmologist and a photographer reader were evaluated.

MATERIALS AND METHODS

157 patients with diabetes, attending screening or follow-up of DR, were evaluated by fundus photographs taken in mydriasis by Optomed Aurora and Canon or Zeiss Visucam fundus cameras. The image quality and the severity of DR were evaluated independently by an ophthalmologist and experienced photographer. The sensitivity, specificity and reliability of the assessments were determined.

RESULTS

1884 fundus images from 314 eyes were analysed. In 53% of all eyes, DR was not present. 10% had mild non-proliferative diabetic retinopathy (NPDR), 16% moderate NPDR, 6% severe NPDR and 16% proliferative diabetic retinopathy (PDR). The DR grading outcomes by Aurora highly equalled to those of Canon or Zeiss (κ = 0.93, 95% CI 0.91 to 0.94), and there was almost perfect agreement in grading between the ophthalmologist and photographer (κ = 0.96, 95% CI 0.95 to 0.97). The image quality of Aurora was sufficient for reliable assessment according to both graders in 84-88% of the cases.

CONCLUSION

The Optomed Aurora fundus camera seems appropriate for DR screening. The sufficient image quality and high diagnostic accuracy for DR grading are supportive for a less expensive and easily transportable screening system for DR. Immediate image grading carried out by a photographer would further improve and speed up the screening process in all settings.

Exploring Ophthalmologists' Adoption of Telemedicine during the COVID-19 Pandemic: A Mixed Methods Study

INTRODUCTION

The COVID-19 pandemic promoted hitherto unseen uptake of telemedicine by ophthalmologists. We performed a mixed methods study to explore patters of utilization during the pandemic and perceived future utility.

METHODS

Ophthalmologists practicing in Canada between March and July 2020 were invited to complete an online questionnaire assessing demographics, clinical practice characteristics and telemedicine utilization prior to and during the pandemic. Descriptive and bivariate statistics were used to analyze the data. Agglomerative hierarchical cluster analysis was used to identify groups who varied on the types of visits offered using telemedicine. Ten one-on-one interviews were conducted and analyzed using thematic content analysis to explain trends observed in the survey data.

RESULTS

Seventy-three ophthalmologists completed the survey. Six percent reported using telemedicine prior to the pandemic compared to 80% during the pandemic. A significant majority (81%) primarily used the telephone for telemedicine visits. Overall, visit volumes during the pandemic declined to 40% of pre-pandemic levels, with a smaller decline for ophthalmologists who used telemedicine than those who did not. Those who used telemedicine for all visit types were more likely to use telemedicine software and to anticipate a modest-to-large role for telemedicine in their future practice.

DISCUSSION

For many ophthalmologists, integrating telemedicine into clinical practice may have partially offset the disruption to normal clinical activities during the pandemic. While the majority saw telemedicine as a temporary solution, a sizeable minority appear to have made considerable use of the technology and see an ongoing role for it once regular clinical activities resume.

Recently updated global diabetic retinopathy screening guidelines: commonalities, differences, and future possibilities

Diabetic retinopathy (DR) is a global health burden. Screening for sight-threatening DR (STDR) is the first cost-effective step to decrease this burden. We analyzed the similarities and variations between the recent country-specific and the International Council of Ophthalmology (ICO) DR guideline to identify gaps and suggest possible solutions for future universal screening. We selected six representative national DR guidelines, one from each World Health Organization region, including Canada (North America), England (Europe), India (South- East Asia), Kenya (Africa), New Zealand (Western Pacific), and American Academy of Ophthalmology Preferred Practice Pattern (used in Latin America and East Mediterranean). We weighed the newer camera and artificial intelligence (AI) technology against the traditional screening methodologies. All guidelines agree that screening for DR and STDR in people with diabetes is currently led by an ophthalmologist; few engage non-ophthalmologists. Significant variations exist in the screening location and referral timelines. Screening with digital fundus photography has largely replaced traditional slit-lamp examination and ophthalmoscopy. The use of mydriatic digital 2-or 4-field fundus photography is the current norm; there is increasing interest in using non-mydriatic fundus cameras. The use of automated DR grading and tele-screening is currently sparse. Country-specific guidelines are necessary to align with national priorities and human resources. International guidelines such as the ICO DR guidelines remain useful in countries where no guidelines exist. Validation studies on AI and tele-screening call for urgent policy decisions to integrate DR screening into universal health coverage to reduce this global public health burden.

The Role of Telemedicine, In-Home Testing and Artificial Intelligence to Alleviate an Increasingly Burdened Healthcare System: Diabetic Retinopathy

In the presence of the ever-increasing incidence of diabetes mellitus (DM), the prevalence of diabetic eye disease (DED) is also growing. Despite many improvements in diabetic care, DM remains a leading cause of visual impairment in working-age patients. So far, prevention has been the best way to protect vision. The sooner we diagnose DED, the more effective the treatment is. Thus, diabetic retinopathy (DR) screening, especially with imaging techniques, is a method of choice for vision protection. To alleviate the burden of diabetic patients who need ophthalmic care, telemedicine and in-home testing are used, supported by artificial intelligence (AI) algorithms. This is why we decided to evaluate current image teleophthalmology methods used for DR screening. We searched the PubMed platform for papers published over the last 5 years (2015–2020) using the following key words: telemedicine in diabetic retinopathy screening, diabetic retinopathy screening, automated diabetic retinopathy screening, artificial intelligence in diabetic retinopathy screening, smartphone diabetic retinopathy testing. We have included 118 original articles meeting the above criteria, discussing imaging diabetic retinopathy screening methods. We have found that fundus cameras, stable or mobile, are most commonly used for retinal photography, with portable fundus cameras also relatively common. Other possibilities involve the use of ultra-wide-field (UWF) imaging and even optical coherence tomography (OCT) devices for DR screening. Also, the role of smartphones is increasingly recognized in the field. Retinal fundus images are assessed by humans instantly or remotely, while AI algorithms seem to be useful tools facilitating retinal image assessment. The common use of smartphones and availability of relatively cheap, easy-to-use adapters for retinal photographs augmented by AI algorithms make it possible for eye fundus photographs to be taken by non-specialists and in non-medical setting. This opens the way for in-home testing conducted on a much larger scale in the future. In conclusion, based on current DR screening techniques, we can suggest that the future practice of eye care specialists will be widely supported by AI algorithms, and this way will be more effective.

Use of Machine Learning Approaches in Clinical Epidemiological Research of Diabetes

PURPOSE OF REVIEW

Machine learning approaches-which seek to predict outcomes or classify patient features by recognizing patterns in large datasets-are increasingly applied to clinical epidemiology research on diabetes. Given its novelty and emergence in fields outside of biomedical research, machine learning terminology, techniques, and research findings may be unfamiliar to diabetes researchers. Our aim was to present the use of machine learning approaches in an approachable way, drawing from clinical epidemiological research in diabetes published from 1 Jan 2017 to 1 June 2020.

RECENT FINDINGS

Machine learning approaches using tree-based learners-which produce decision trees to help guide clinical interventions-frequently have higher sensitivity and specificity than traditional regression models for risk prediction. Machine learning approaches using neural networking and "deep learning" can be applied to medical image data, particularly for the identification and staging of diabetic retinopathy and skin ulcers. Among the machine learning approaches reviewed, researchers identified new strategies to develop standard datasets for rigorous comparisons across older and newer approaches, methods to illustrate how a machine learner was treating underlying data, and approaches to improve the transparency of the machine learning process. Machine learning approaches have the potential to improve risk stratification and outcome prediction for clinical epidemiology applications. Achieving this potential would be facilitated by use of universal open-source datasets for fair comparisons. More work remains in the application of strategies to communicate how the machine learners are generating their predictions.

Telemedicine in ophthalmology in view of the emerging COVID-19 outbreak

Purpose

Technological advances in recent years have resulted in the development and implementation of various modalities and techniques enabling medical professionals to remotely diagnose and treat numerous medical conditions in diverse medical fields, including ophthalmology. Patients who require prolonged isolation until recovery, such as those who suffer from COVID-19, present multiple therapeutic dilemmas to their caregivers. Therefore, utilizing remote care in the daily workflow would be a valuable tool for the diagnosis and treatment of acute and chronic ocular conditions in this challenging clinical setting. Our aim is to review the latest technological and methodical advances in teleophthalmology and highlight their implementation in screening and managing various ocular conditions. We present them as well as potential diagnostic and treatment applications in view of the recent SARS-CoV-2 virus outbreak.

Methods

A computerized search from January 2017 up to March 2020 of the online electronic database PubMed was performed, using the following search strings: “telemedicine,” “telehealth,” and “ophthalmology.” More generalized complementary contemporary research data regarding the COVID-19 pandemic was also obtained from the PubMed database.

Results

A total of 312 records, including COVID-19-focused studies, were initially identified. After exclusion of non-relevant, non-English, and duplicate studies, a total of 138 records were found eligible. Ninety records were included in the final qualitative analysis.

Conclusion

Teleophthalmology is an effective screening and management tool for a range of adult and pediatric acute and chronic ocular conditions. It is mostly utilized in screening of retinal conditions such as retinopathy of prematurity, diabetic retinopathy, and age-related macular degeneration; in diagnosing anterior segment condition; and in managing glaucoma. With improvements in image processing, and better integration of the patient’s medical record, teleophthalmology should become a more accepted modality, all the more so in circumstances where social distancing is inflicted upon us.