Although the clinical activity score (CAS) is a validated scoring system for identifying disease activity of thyroid-associated orbitopathy (TAO), it may produce differing results depending on the evaluator, and an experienced ophthalmologist is required for accurate evaluation. In this study, we developed a machine learning (ML)-assisted system to mimic an expert’s CAS assessment using digital facial images and evaluated its accuracy for predicting the CAS and diagnosing active TAO (CAS ≥ 3). An ML-assisted system was designed to assess five CAS components related to inflammatory signs (redness of the eyelids, redness of the conjunctiva, swelling of the eyelids, inflammation of the caruncle and/or plica, and conjunctival edema) in patients’ facial images and to predict the CAS by considering two components of subjective symptoms (spontaneous retrobulbar pain and pain on gaze). To train and test the system, 3,060 cropped images from 1020 digital facial images of TAO patients were used. The reference CAS for each image was scored by three ophthalmologists, each with > 15 years of clinical experience. We repeated the experiments for 30 randomly split training and test sets at a ratio of 8:2. The sensitivity and specificity of the ML-assisted system for diagnosing active TAO were 72.7% and 83.2% in the test set constructed from the entire dataset. For the test set constructed from the dataset with consistent results for the three ophthalmologists, the sensitivity and specificity for diagnosing active TAO were 88.1% and 86.9%. In the test sets from the entire dataset and from the dataset with consistent results, 40.0% and 49.9% of the predicted CAS values were the same as the reference CAS, respectively. The system predicted the CAS within 1 point of the reference CAS in 84.6% and 89.0% of cases when tested using the entire dataset and in the dataset with consistent results, respectively. An ML-assisted system estimated the clinical activity of TAO and detect inflammatory active TAO with reasonable accuracy. The accuracy could be improved further by obtaining more data. This ML-assisted system can help evaluate the disease activity consistently as well as accurately and enable the early diagnosis and timely treatment of active TAO.

Objective: To elucidate the clinical features of ocular motility and the risk factors for recurrence in idiopathic orbital myositis. Methods: The medical records of 31 patients diagnosed with idiopathic orbital inflammation between 2003 and 2019 were retrospectively reviewed. All patients were initially treated with corticosteroids. Treatment outcome and ocular motility were noted. Results: Twenty-six patients (84%) had unilateral involvement and five patients (16%) were bilateral. Of the 31 patients, 22 patients (71%) showed ocular motility limitation. The mean grading scale of extraocular muscle (EOM) limitation was −1.65 ± 1.80. EOM limitation was found in the same direction of the most affected muscle in 14 patients (64%), while 8 patients (36%) showed duction limitation in the opposite direction. Nine patients (35%) suffered from recurrence. Recurrence was more likely to occur in patients with multiple muscle involvement (p < 0.001). The interval to relapse of symptoms after discontinuation of steroids was significantly shorter in patients with multiple recurrences compared to those with a single recurrence (1.8 ± 0.8 weeks versus 6.0 ± 1.4 weeks, p = 0.020). Conclusions: Idiopathic orbital myositis showed variable degrees of ocular motility limitation, and limitation in the same direction of the action of the affected muscle was more frequent. Recurrent myositis was more likely to have multiple muscle involvement. Rapid relapse of symptoms after discontinuation of steroids was a significant indicator of multiple recurrences.

Purpose: To investigate the association between the bony nasolacrimal duct (NLD) size and outcomes of nasolacrimal silicone intubation for incomplete primary acquired nasolacrimal duct obstruction (PANDO). Methods: Patients who underwent silicone intubation for incomplete PANDO and had undergone facial computed tomography (CT) were included. Surgical success was judged by both epiphora improvement and normalized tear meniscus height (TMH; < 300 μm) on anterior segment optical coherence tomography at 3 months after tube removal. The area, major axis diameter, and minor axis diameter of the elliptic bony NLD sections were measured in 1.0 mm-thick axial CT images. These bony NLD sizes were analyzed for associations with surgical success and TMH normalization. Results: Eighty-one eyes of 48 patients were investigated. The smallest area and the smallest minor axis diameter were significantly larger in the success group (49 eyes), compared with those in the failure group (median smallest minor axis diameter: 4.7 mm vs. 3.8 mm, P = 0.008, Mann–Whitney U test). There was also a tendency for the TMH normalization rate to significantly increase as the smallest area and the smallest minor axis diameter increased (P = 0.028 and 0.037, respectively, Fisher’s 2 × 4 tests). Under multivariable logistic regression analysis using generalized estimating equation, a larger smallest minor axis diameter was associated with success of the nasolacrimal silicone intubation (odds ratio: 2.481, 95% confidence interval: 1.143–5.384). Conclusion: Surgical success of the nasolacrimal silicone intubation in incomplete PANDO is associated with a larger smallest minor axis diameter of the bony NLD. This finding will help understand the pathophysiology of surgical failure after nasolacrimal silicone intubation.

The biological behavior of sebaceous carcinoma (SeC) is relatively indolent; however, local invasion or distant metastasis is sometimes reported. Nevertheless, a lack of understanding of the genetic background of SeC makes it difficult to apply effective systemic therapy. This study was designed to investigate major genetic alterations in SeCs in Korean patients. A total of 29 samples, including 20 ocular SeCs (SeC-Os) and 9 extraocular SeCs (SeC-EOs), were examined. Targeted next-generation sequencing tests including 171 cancer-related genes were performed. TP53 and PIK3CA genes were frequently mutated in both SeC-Os and SeC-EOs with slight predominance in SeC-Os, whereas the NOTCH1 gene was more commonly mutated in SeC-EOs. In clinical correlation, mutations in RUNX1 and ATM were associated with development of distant metastases, and alterations in MSH6 and BRCA1 were associated with inferior progression-free survival (all p < 0.05). In conclusion, our study revealed distinct genetic alterations between SeC-Os and SeC-EOs and some important prognostic molecular markers. Mutations in potentially actionable genes, including EGFR, ERBB2, and mismatch repair genes, were noted, suggesting consideration of a clinical trial in intractable cases.

To investigate periorbital lipogranuloma cases that developed after autologous fat injection and to determine various treatment outcomes from these cases. This retrospective study involved 27 patients who presented with periocular mass (final diagnosis of lipogranuloma) and had history of facial autologous fat injection. The collected data included information on patient sex, age, clinical presentation, number and site of fat injections, interval between injections, duration from injection to symptom onset, fat harvesting site, use of cryopreservation, and treatment outcome. The most common presenting symptom was palpable mass (92.6%), followed by blepharoptosis and eyelid edema. The mean time from injection to symptom onset was 13.6 ± 29.2 months (range, 2 to 153 months). Patients were managed by intralesional triamcinolone injection (six patients) and surgical excision (three patients); 18 patients were followed without treatment. Among the six patients who underwent intralesional triamcinolone injection, five showed complete resolution, and one showed partial resolution. Among the 18 patients who were followed without management, three showed spontaneous resolution over a 5-month follow-up period. Lipogranuloma can develop in the eyelid after autologous fat injection into the face. Both surgical excision and intralesional triamcinolone injection yield relatively good outcomes. Simple observation can be a good option because spontaneous resolution can occur in a subset of patients.

Background: We previously developed a machine learning (ML)-assisted system for predicting the clinical activity score (CAS) in thyroid-associated orbitopathy (TAO) using digital facial images taken by a digital single-lens reflex camera in a studio setting. In this study, we aimed to apply this system to smartphones and detect active TAO (CAS ≥3) using facial images captured by smartphone cameras. We evaluated the performance of our system on various smartphone models and compared it with the performance of ophthalmologists with varying clinical experience. Methods: We applied the preexisting ML architecture to classify photos taken with smartphones (Galaxy S21 Ultra, iPhone 12 pro, iPhone 11, iPhone SE 2020, Galaxy M20, and Galaxy A21S). The performance was evaluated with smartphone-captured images from 100 patients with TAO. Three ophthalmology residents, three general ophthalmologists with <5 years of clinical experience, and three oculoplastic specialists independently interpreted the same set of images taken under a studio environment and compared their results with those generated by the smartphone-based ML-assisted system. Reference CAS was determined by a consensus of three oculoplastic specialists. Results: Active TAO (CAS ≥3) was identified in 28 patients. Smartphone model used in capturing facial images influenced active TAO detection performance (F1 score 0.59–0.72). The smartphone-based system showed 74.5% sensitivity, 84.8% specificity, and F1 score 0.70 on top three smartphones. On images from all six smartphones, average sensitivity, specificity, and F1 score were 71.4%, 81.6%, and 0.66, respectively. Ophthalmology residents’ values were 69.1%, 55.1%, and 0.46. General ophthalmologists’ values were 61.9%, 79.6%, and 0.55. Oculoplastic specialists’ values were 73.8%, 90.7%, and 0.75. This smartphone-based ML-assisted system predicted CAS within 1 point of reference CAS in 90.7% using facial images from smartphones. Conclusions: Our smartphone-based ML-assisted system shows reasonable accuracy in detecting active TAO, comparable with oculoplastic specialists and outperforming residents and general ophthalmologists. It may enable reliable self-monitoring for disease activity, but confirmatory research is needed for clinical application.