Predicting Postoperative Vision for Macular Hole with Automated Image Analysis

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.

The Effect of Macular Hole Duration on Surgical Outcomes: An Individual Participant Data Study of Randomized Controlled Trials

TOPIC

To define the effect of symptom duration on outcomes in people undergoing surgery for idiopathic full-thickness macular holes (iFTMHs) by means of an individual participant data (IPD) study of randomized controlled trials (RCTs). The outcomes assessed were primary iFTMH closure and postoperative best-corrected visual acuity (BCVA).

CLINICAL RELEVANCE

Idiopathic full-thickness macular holes are visually disabling with a prevalence of up to 0.5%. Untreated BCVA is typically reduced to 20/200. Surgery can close holes and improve vision. Symptom duration is thought to affect outcomes with surgery, but the effect is unclear.

METHODS

A systematic review identified eligible RCTs that included adults with iFTMH undergoing vitrectomy with gas tamponade in which symptom duration, primary iFTMH closure, and postoperative BCVA were recorded. Bibliographic databases were searched for articles published between 2000 and 2020. Individual participant data were requested from eligible studies.

RESULTS

Twenty eligible RCTs were identified. Data were requested from all studies and obtained from 12, representing 940 eyes in total. Median symptom duration was 6 months (interquartile range, 3-10). Primary closure was achieved in 81.5% of eyes. There was a linear relationship between predicted probability of closure and symptom duration. Multilevel logistic regression showed each additional month of duration was associated with 0.965 times lower odds of closure (95% confidence interval [CI], 0.935-0.996, P = 0.026). Internal limiting membrane (ILM) peeling, ILM flap use, better preoperative BCVA, face-down positioning, and smaller iFTMH size were associated with increased odds of primary closure. Median postoperative BCVA in eyes achieving primary closure was 0.48 logarithm of the minimum angle of resolution (logMAR) (20/60). Multilevel logistic regression showed for eyes achieving primary iFTMH closure, each additional month of symptom duration was associated with worsening BCVA by 0.008 logMAR units (95% CI, 0.005-0.011, P < 0.001) (i.e., ∼1 Early Treatment Diabetic Retinopathy Study letter loss per 2 months). ILM flaps, intraocular tamponade using long-acting gas, better preoperative BCVA, smaller iFTMH size, and phakic status were also associated with improved postoperative BCVA.

CONCLUSIONS

Symptom duration was independently associated with both anatomic and visual outcomes in persons undergoing surgery for iFTMH. Time to surgery should be minimized and care pathways designed to enable this.

Development and validation of a deep learning system to classify aetiology and predict anatomical outcomes of macular hole

AIMS

To develop a deep learning (DL) model for automatic classification of macular hole (MH) aetiology (idiopathic or secondary), and a multimodal deep fusion network (MDFN) model for reliable prediction of MH status (closed or open) at 1 month after vitrectomy and internal limiting membrane peeling (VILMP).

METHODS

In this multicentre retrospective cohort study, a total of 330 MH eyes with 1082 optical coherence tomography (OCT) images and 3300 clinical data enrolled from four ophthalmic centres were used to train, validate and externally test the DL and MDFN models. 266 eyes from three centres were randomly split by eye-level into a training set (80%) and a validation set (20%). In the external testing dataset, 64 eyes were included from the remaining centre. All eyes underwent macular OCT scanning at baseline and 1 month after VILMP. The area under the receiver operated characteristic curve (AUC), accuracy, specificity and sensitivity were used to evaluate the performance of the models.

RESULTS

In the external testing set, the AUC, accuracy, specificity and sensitivity of the MH aetiology classification model were 0.965, 0.950, 0.870 and 0.938, respectively; the AUC, accuracy, specificity and sensitivity of the postoperative MH status prediction model were 0.904, 0.825, 0.977 and 0.766, respectively; the AUC, accuracy, specificity and sensitivity of the postoperative idiopathic MH status prediction model were 0.947, 0.875, 0.815 and 0.979, respectively.

CONCLUSION

Our DL-based models can accurately classify the MH aetiology and predict the MH status after VILMP. These models would help ophthalmologists in diagnosis and surgical planning of MH.

Predicting Visual Improvement After Macular Hole Surgery: A Combined Model Using Deep Learning and Clinical Features

Purpose

The purpose of this study was to assess the feasibility of deep learning (DL) methods to enhance the prediction of visual acuity (VA) improvement after macular hole (MH) surgery from a combined model using DL on high-definition optical coherence tomography (HD-OCT) B-scans and clinical features.

Methods

We trained a DL convolutional neural network (CNN) using pre-operative HD-OCT B-scans of the macula and combined with a logistic regression model of pre-operative clinical features to predict VA increase ≥15 Early Treatment Diabetic Retinopathy Study (ETDRS) letters at 6 months post-vitrectomy in closed MHs. A total of 121 MHs with 242 HD-OCT B-scans and 484 clinical data points were used to train, validate, and test the model. Prediction of VA increase was evaluated using the area under the receiver operating characteristic curve (AUROC) and F1 scores. We also extracted the weight of each input feature in the hybrid model.

Results

All performances are reported on the held-out test set, matching results obtained with cross-validation. Using a regression on clinical features, the AUROC was 80.6, with an F1 score of 79.7. For the CNN, relying solely on the HD-OCT B-scans, the AUROC was 72.8 ± 14.6, with an F1 score of 61.5 ± 23.7. For our hybrid regression model using clinical features and CNN prediction, the AUROC was 81.9 ± 5.2, with an F1 score of 80.4 ± 7.7. In the hybrid model, the baseline VA was the most important feature (weight = 59.1 ± 6.9%), while the weight of HD-OCT prediction was 9.6 ± 4.2%.

Conclusions

Both the clinical data and HD-OCT models can predict postoperative VA improvement in patients undergoing vitrectomy for a MH with good discriminative performances. Combining them into a hybrid model did not significantly improve performance.

Translational Relevance

OCT-based DL models can predict postoperative VA improvement following vitrectomy for MH but fusing those models with clinical data might not provide improved predictive performance.

Benchmarking automated detection of the retinal external limiting membrane in a 3D spectral domain optical coherence tomography image dataset of full thickness macular holes

In this article, we present a new benchmark for the segmentation of the retinal external limiting membrane (ELM) using an image dataset of spectral domain optical coherence tomography (OCT) scans in a patient population with idiopathic full-thickness macular holes. Specifically, the dataset used contains OCT images from one eye of 107 patients with an idiopathic full-thickness macular hole. In total, the dataset contains 5243 individual 2-dimensional (2-D) OCT image slices, with each patient contributing 49 individual spectral-domain OCT tagged image slices. We display precise image-wise binary annotations to segment the ELM line. The OCT images present high variations in image contrast, motion, brightness, and speckle noise which can affect the robustness of applied algorithms, so we performed an extensive OCT imaging and annotation data quality analysis. Imaging data quality control included noise, blurriness and contrast scores, motion estimation, darkness and average pixel scores, and anomaly detection. Annotation quality was measured using gradient mapping of ELM line annotation confidence, and idiopathic full-thickness macular hole detection. Finally, we compared qualitative and quantitative results with seven state-of-the-art machine learning-based segmentation methods to identify the ELM line with an automated system.