Double vision is a major manifestation in moderate to severe graves' orbitopathy, but it correlates negatively with inflammatory signs and proptosis

Use of extreme gradient boosting, light gradient boosting machine, and deep neural networks to evaluate the activity stage of extraocular muscles in thyroid-associated ophthalmopathy

PURPOSE

To develop a machine learning model to evaluate the activity stage of extraocular muscles in thyroid-associated ophthalmopathy (TAO).

METHODS

This study retrospectively analysed data from patients with TAO who underwent contrast-enhanced magnetic resonance imaging (MRI) from 2015 to 2022. Three independent machine learning models, namely, extreme gradient boosting (XGBoost), light gradient boosting machine (LightGBM), and deep neural networks (DNNs), were constructed using common clinical features. The performance of these models was compared using evaluation metrics such as the area under the receiver operating curve (AUC), accuracy, precision, recall, and F1 score. The importance of features was explained using Shapley additive explanations (SHAP).

RESULTS

A total of 2561 eyes of 1479 TAO patients were included in this study. The original dataset was randomly divided into a training set (80%, n = 2048) and a test set (20%, n = 513). In the performance evaluation of the test set, the LightGBM model had the best diagnostic performance (AUC 0.9260). According to the SHAP results, features such as conjunctival congestion, swollen caruncles, oedema of the upper eyelid, course of TAO, and intraocular pressure had the most significant impact on the LightGBM model.

CONCLUSION

This study used contrast-enhanced MRI as an objective evaluation criterion and constructed a LightGBM model based on readily accessible clinical data. The model had good classification performance, making it a promising artificial intelligence (AI)-assisted tool to help community hospitals evaluate the inflammatory activity of extraocular muscles in TAO patients in a timely manner.

Effect of orbital wall resection areas in the treatment of patients with endocrine orbitopathy

In patients treated by orbital wall decompression for endocrine orbitopathy (EO) there is limited evidence on the effect of orbital wall resections. Thus, the aim of this study was to evaluate the effect of one, two, and three-wall resections on orbital parameters to determine if any such correlations exist. Preoperative and postoperative data from all patients at a tertiary care centre who underwent decompression surgery from 2010 - 2020 were digitally analysed. The effect of the number and area of resected walls on orbital area, orbital volume, and Hertel value, and the effect of lateral rim advancement (LARA) were determined. A total of 131 orbital areas showed an increase from a mean (SD) preoperative area of 42.0 (4.6) cm² to 47.3 (6.1) cm² postoperatively (p<0.001). In total, the mean (SD) area of osseous wall removed in all patients was 6.2 (1.7) cm² at the lateral orbit (n = 129), 6.7 (2.3) cm² at the orbital floor (n = 123), and 5.8 (1.8) cm² at the medial orbital wall (n =30). The mean (SD) orbital volume increased by 6.0 (3.0) cm³ after decompression. There was also a significant reduction in exophthalmos of 7.3 (3.2) mm (from 25.2 (3.9) to 17.9 (3.5), p<0.001). LARA was performed in 50 patients. Changes in volume and area, and reduction in exophthalmos were not significantly different with or without LARA. The postoperative effects of orbital wall resection are predictable and exhibit a relation with six units of change. Two-wall resection is the most common intervention.

Orbital Signaling in Graves' Orbitopathy

Graves' orbitopathy (GO) is a complex and poorly understood disease in which extensive remodeling of orbital tissue is dominated by adipogenesis and hyaluronan production. The resulting proptosis is disfiguring and underpins the majority of GO signs and symptoms. While there is strong evidence for the thyrotropin receptor (TSHR) being a thyroid/orbit shared autoantigen, the insulin-like growth factor 1 receptor (IGF1R) is also likely to play a key role in the disease. The pathogenesis of GO has been investigated extensively in the last decade with further understanding of some aspects of the disease. This is mainly derived by using in vitro and ex vivo analysis of the orbital tissues. Here, we have summarized the features of GO pathogenesis involving target autoantigens and their signaling pathways.

Prevalence of thyroid eye disease in Graves' disease: A meta-analysis and systematic review

OBJECTIVE

Thyroid eye disease (TED) is a debilitating condition that frequently manifests in patients suffering from Graves' disease (GD). This study aims to analyse the prevalence of TED among GD patients, with a focus on geographical region-specific rates.

METHODS

Medline and Embase were searched for articles examining TED prevalence on April 2020, and articles were retrieved and sieved. Statistical analysis was performed after Freeman-Tukey double arcsine transformation. Thereafter, results were pooled with random effects by DerSimonian and Laird model.

RESULTS

Fifty-seven articles involving 26,804 patients were included in the review. The overall pooled prevalence of TED was 40% (CI: 0.32 to 0.48) and by continent was 38% (CI: 0.31 to 0.46) for Europe, 44% (CI: 0.32 to 0.56) for Asia, 27% (CI: 0.06 to 0.56) for North America and 58% (CI: 0.55 to 0.61) for Oceania. The prevalence of TED in Southeast Asia was 35% (CI: 0.24 to 0.47) and Middle East 48% (CI: 0.19 to 0.78). Subgroup analysis showed regions with predominantly Caucasians (37%; CI: 0.28 to 0.46) had a lower prevalence of TED compared to Asians (45%; CI: 0.33 to 0.58). The pooled prevalence of lid retraction was 57% (CI: 0.39 to 0.74), proptosis 57% (CI: 0.48 to 0.65), diplopia 36% (CI: 0.24 to 0.48) and ocular hypertension 13% (CI: 0.06 to 0.19).

CONCLUSION

A substantial proportion of patients with GD have TED and often manifest as lid retraction, proptosis and diplopia. Early detection through active screening might help to mitigate the progression of TED and its associated complications.

Contextual Deep Regression Network for Volume Estimation in Orbital CT

Diseases of the optic nerve cause structural changes observable through clinical computed tomography (CT) imaging. Previous work has shown that multi-atlas methods can be used to segment and extract volumetric measurements from the optic nerve, which are associated with visual disability and disease. In this work, we trained a weakly supervised convolutional neural network to learn optic nerve volumes directly, without segmentation. Furthermore, we explored the role of contextual electronic medical record (EMR) information, specifically ICD-9 codes, to improve optic nerve volume estimation. We constructed a merged network to combine data from imaging as well as EMR and demonstrated that context improved volume prediction, with a 15% increase in explained-variance ( R ²). Finally, we compared disease prediction models using volumes learned from multi-atlas, CNN, and contextual-CNN. We observed that the predicted optic nerve volume from merge-CNN had an AUC of 0.74 for classification of disease, as compared to an AUC of 0.54 using the multi-atlas metric. This is the first work to show that a contextually derived volume biomarker is more accurate than volume estimations through multi-atlas or weakly supervised image CNN. These results highlight the potential for image processing improvements by incorporating non-imaging data.

Electronic Medical Record Context Signatures Improve Diagnostic Classification Using Medical Image Computing

Composite models that combine medical imaging with electronic medical records (EMR) improve predictive power when compared to traditional models that use imaging alone. The digitization of EMR provides potential access to a wealth of medical information, but presents new challenges in algorithm design and inference. Previous studies, such as Phenome Wide Association Study (PheWAS), have shown that EMR data can be used to investigate the relationship between genotypes and clinical conditions. Here, we introduce Phenome-Disease Association Study to extend the statistical capabilities of the PheWAS software through a custom Python package, which creates diagnostic EMR signatures to capture system-wide co-morbidities for a disease population within a given time interval. We investigate the effect of integrating these EMR signatures with radiological data to improve diagnostic classification in disease domains known to have confounding factors because of variable and complex clinical presentation. Specifically, we focus on two studies: First, a study of four major optic nerve related conditions; and second, a study of diabetes. Addition of EMR signature vectors to radiologically derived structural metrics improves the area under the curve (AUC) for diagnostic classification using elastic net regression, for diseases of the optic nerve. For glaucoma, the AUC improves from 0.71 to 0.83, for intrinsic optic nerve disease it increases from 0.72 to 0.91, for optic nerve edema it increases from 0.95 to 0.96, and for thyroid eye disease from 0.79 to 0.89. The EMR signatures recapitulate known comorbidities with diabetes, such as abnormal glucose, but do not significantly modulate image-derived features. In summary, EMR signatures present a scalable and readily applicable.