The Fight Tumour Blindness Registry: Efficient capture of high-quality real-world data in uveal melanoma.

OBJECTIVE: To describe the development of a web-based data collection tool to track the management and outcomes of uveal melanoma patients. DESIGN: Description of a clinical registry. PARTICIPANTS: Patients with uveal melanoma. METHODS: A panel of expert ocular oncologists, with input from other relevant specialties and individuals with expertise in registry development, collaborated to formulate a minimum data set to be collected to track patient centred, real-world outcomes in uveal melanoma. This data set was used to create the Fight Tumour Blindness! (FTB!) registry within Save Sight Registries. RESULTS: The data set to be collected includes patient demographics and medical history, baseline visit, follow-up visit including tumour treatment, metastatic staging and surveillance, pathology, and patient-reported questionnaires. The inbuilt mechanisms to ensure efficient and complete data collection are described. CONCLUSIONS: The FTB! registry can be used to monitor outcomes for patients with uveal melanoma. It allows benchmarking of outcomes and comparisons between different clinics and countries.

Morganella morganii and Enterococcus faecalis endophthalmitis following intravitreal injection.

BACKGROUND: Endophthalmitis following intravitreal injection is a potentially devastating complication of anti-VEGF injections. Post-injection endophthalmitis due to Enterococcus faecalis is rare, and no previous case of Morganella morganii endophthalmitis after intravitreal injection has been reported. CASE PRESENTATION: We present the first reported case of Morganella morganii and Enterococcus faecalis endophthalmitis after intravitreal injection in an immunocompetent patient in the absence of recent ocular surgery. Our patient presented with hand movement visual acuity one day after anti-VEGF injection and demonstrated no clinical improvement despite repeated intravitreal ceftazidime and vancomycin injections. A decision was made to proceed with early vitrectomy given failure of intravitreal antibiotics. Visual acuity improved to 6/90 at 12 weeks after vitrectomy without any evidence of disease recurrence. CONCLUSIONS: Post-injection endophthalmitis due to concurrent Morganella morganii and Enterococcus faecalis infections can have visually devastating consequences despite repeated empirical and targeted intravitreal antibiotics. Lack of clinical improvement following intravitreal antibiotics should warrant consideration of early vitrectomy. Our experience is a pertinent reminder of the ever-growing threat of uncommon and multi-resistant bacteria that must be considered when treating infections such as post-injection endophthalmitis.

Optimising vitrectomy operation note coding with machine learning.

BACKGROUND: The accurate encoding of operation notes is essential for activity-based funding and workforce planning. The aim of this project was to evaluate the procedural coding accuracy of vitrectomy and to develop machine learning, natural language processing (NLP) models that may assist with this task. METHODS: This retrospective cohort study involved vitrectomy operation notes between a 21-month period at the Royal Adelaide Hospital. Coding of procedures were based on the Medicare Benefits Schedule (MBS)-the Australian equivalent to the Current Procedural Terminology (CPT®) codes used in the United States. Manual encoding was conducted for all procedures and reviewed by two vitreoretinal consultants. XGBoost, random forest and logistic regression models were developed for classification experiments. A cost-based analysis was subsequently conducted. RESULTS: There were a total of 1724 procedures with individual codes performed within 617 vitrectomy operation notes totalling $1 528 086.60 after manual review. A total of 1147 (66.5%) codes were missed in the original coding that amounted to $736 539.20 (48.2%). Our XGBoost model had the highest classification accuracy (94.6%) in the multi-label classification for the five most common procedures. The XGBoost model was the most successful model in identifying operation notes with two or more missing codes with an AUC of 0.87 (95% CI 0.80-0.92). CONCLUSIONS: Machine learning has been successful in the classification of vitrectomy operation note encoding. We recommend a combined human and machine learning approach to clinical coding as automation may facilitate more accurate reimbursement and enable surgeons to prioritise higher quality clinical care.

Calciphylaxis, beware the ophthalmic mimic: A case series.

PURPOSE: We present two atypical cases of calciphylaxis presenting with ocular ischemic pathology - both without the hallmark cutaneous manifestations - to raise awareness of this rare yet highly disabling condition. OBSERVATIONS: We report two cases of ophthalmic calciphylaxis presenting as (1) anterior ischemic optic neuropathy (AION) and cilioretinal artery occlusion in a 76-year-old woman with pre-dialysis kidney failure, and (2) AION with contralateral central retinal artery occlusion (CRAO) in a 44-year-old man on hemodialysis. CONCLUSION AND IMPORTANCE: These cases highlight the need for judicious clinical suspicion of calciphylaxis in patients with kidney failure, presenting with microvascular ischemic ophthalmic pathology such as AION or CRAO. Confirmation with temporal artery biopsy is essential to direct targeted individualized multi-disciplinary treatment of calciphylaxis and avoid unnecessary steroid exposure in cases masquerading as giant cell arteritis (GCA).

Improved Productivity Using Deep Learning-assisted Reporting for Lumbar Spine MRI.

Background Lumbar spine MRI studies are widely used for back pain assessment. Interpretation involves grading lumbar spinal stenosis, which is repetitive and time consuming. Deep learning (DL) could provide faster and more consistent interpretation. Purpose To assess the speed and interobserver agreement of radiologists for reporting lumbar spinal stenosis with and without DL assistance. Materials and Methods In this retrospective study, a DL model designed to assist radiologists in the interpretation of spinal canal, lateral recess, and neural foraminal stenoses on lumbar spine MRI scans was used. Randomly selected lumbar spine MRI studies obtained in patients with back pain who were 18 years and older over a 3-year period, from September 2015 to September 2018, were included in an internal test data set. Studies with instrumentation and scoliosis were excluded. Eight radiologists, each with 2-13 years of experience in spine MRI interpretation, reviewed studies with and without DL model assistance with a 1-month washout period. Time to diagnosis (in seconds) and interobserver agreement (using Gwet κ) were assessed for stenosis grading for each radiologist with and without the DL model and compared with test data set labels provided by an external musculoskeletal radiologist (with 32 years of experience) as the reference standard. Results Overall, 444 images in 25 patients (mean age, 51 years ± 20 [SD]; 14 women) were evaluated in a test data set. DL-assisted radiologists had a reduced interpretation time per spine MRI study, from a mean of 124-274 seconds (SD, 25-88 seconds) to 47-71 seconds (SD, 24-29 seconds) (P < .001). DL-assisted radiologists had either superior or equivalent interobserver agreement for all stenosis gradings compared with unassisted radiologists. DL-assisted general and in-training radiologists improved their interobserver agreement for four-class neural foraminal stenosis, with κ values of 0.71 and 0.70 (with DL) versus 0.39 and 0.39 (without DL), respectively (both P < .001). Conclusion Radiologists who were assisted by deep learning for interpretation of lumbar spinal stenosis on MRI scans showed a marked reduction in reporting time and superior or equivalent interobserver agreement for all stenosis gradings compared with radiologists who were unassisted by deep learning. © RSNA, 2022 Online supplemental material is available for this article. See also the editorial by Hayashi in this issue.

Open book technique for conjunctival tumors: a surgical approach to the fornix.

Surgical management of large tumors involving the conjunctival fornix can be challenging, as exposure and clear margins may be difficult to achieve. In this case series, we report our initial experience with the open book technique in 4 patients. Through a canthotomy and cantholysis, this surgical approach provides a wide surgical field and facilitates fornix reconstruction post large tumor excision. In our series, one patient had a lateral canthus dehiscence.

Cilioretinal Artery Occlusion in Posterior Scleritis Secondary to Ankylosing Spondylitis.

The authors describe the case of a 25-year-old male who presented with a cilioretinal artery occlusion secondary to posterior scleritis. The patient had a history of juvenile spondyloarthritis that evolved into ankylosing spondylitis. Cilioretinal artery occlusion is a rare complication of posterior scleritis, having only been described once previously in the literature. This is the first reported case of a cilioretinal artery occlusion in posterior scleritis that was associated with an underlying systemic disease. [Ophthalmic Surg Lasers Imaging Retina. 2021;52:102-106.].

Circumscribed choroidal haemangioma: Indocyanine green angiography features on scanning laser ophthalmoscopy versus traditional digital fundus photography.

BACKGROUND AND OBJECTIVE: Circumscribed choroidal haemangioma (CCH) has several characteristic clinical and angiographic features. We aimed to compare indocyanine green angiography (ICG) findings of CCH captured on a traditional digital camera system (DCS) to newer scanning laser ophthalmoscopy (SLO) platforms. STUDY DESIGN/MATERIALS AND METHODS: A total of 35 patients over a 10-year period diagnosed with CCH using ICG were included (18 imaged with DCS and 17 with SLO). RESULTS: On early ICG frames, intrinsic vessels were apparent in two-thirds (12/18; 67%) of the DCS group compared with all of eyes in the SLO group (p = 0.020). In addition, at maximal hyperfluorescence, most eyes imaged with DCS had a feathery appearance (16/18; 89%) compared with those in the SLO group which all (17/17; 100%) displayed a granular appearance (p < 0.001). The presence of hot spots at maximal hyperfluorescence was also more common in the SLO group (12/17; 71%) versus the DCS group (0/18; 0%) (p < 0.001). Finally, intrinsic vessels and vascular loops could be identified throughout the entire duration of the ICG in 100% of the SLO cases (17/17) versus only 11% (2/18) of DCS cases (p < 0.001). CONCLUSIONS: The visualization of intrinsic vessels, vascular loops, and "hot spots" in CCH is significantly enhanced with SLO compared with DCS. Many characteristic mid-late angiographic findings of CCH are more optimally visualized on SLO which may negate the need for late frames (>30 min) without compromising diagnostic accuracy.

PHACOVITRECTOMY FOR PRIMARY RHEGMATOGENOUS RETINAL DETACHMENT REPAIR: A Retrospective Review.

PURPOSE: To analyze the single surgery success rate and anterior segment complications related to phacoemulsification and intraocular lens implantation in a series of patients undergoing phacovitrectomy for all types of primary rhegmatogenous retinal detachment. METHODS: We performed a retrospective interventional case series on 302 eyes undergoing phacovitrectomy for primary rhegmatogenous retinal detachment repair between November 1, 2016, and February 2, 2019, in Edmonton, Canada. Primary outcomes included single surgery retinal reattachment rate and anterior segment complications. Secondary outcomes included the effects of proliferative vitreoretinopathy and macula and/or peripheral internal limiting membrane peeling on the rate of surgical success. RESULTS: The single surgery success rate of phacovitrectomy for all types of primary rhegmatogenous retinal detachment was 85.1%. The presence of proliferative vitreoretinopathy was associated with lower surgical success (odds ratio, 0.33; P = 0.01). Macular internal limiting membrane peeling was associated with higher surgical success (odds ratio, 2.4; P = 0.05). Anterior segment complications included posterior capsular opacification (28.8%), posterior synechiae (10.9%), and posterior capsular rupture (2.3%). CONCLUSION: Phacovitrectomy is a safe and effective treatment option for the primary repair of rhegmatogenous retinal detachments. This study provides evidence to support the safe incorporation of phacoemulsification and intraocular lens implantation with retinal surgery.

Two Cases of Giant Melanocytoma (Hyperpigmented Magnocellular Nevus).

Melanocytoma or hyperpigmented magnocellular nevus is a variant of melanocytic nevus that is most commonly seen in the optic nerve, but has also been reported to occur in the iris, ciliary body, choroid, sclera, and conjunctiva. We present two cases of giant uveal melanocytoma with histopathology. The first case occurred in a 10-year-old girl who presented with decreased vision in the right eye and a mushroom-shaped pigmented choroidal lesion measuring 15.5 mm in apical height. The lesion was abutting the lens but not causing a cataract. This was diagnosed as a choroidal melanocytoma on open scleral window biopsy. The second case was in a 68-year-old lady, referred for a left nasal pigmented choroidal lesion measuring 8 mm in apical height and having a mushroom configuration. The lesion grew to 8.6 mm in height and was complicated by a vitreous hemorrhage and rhegmatogenous retinal detachment and was treated with iodine-125 plaque brachytherapy. Subsequently, the treated eye became a painful phthisical eye and was enucleated. Histopathology confirmed melanocytoma with extrascleral extension but without malignant transformation. Features of melanocytoma and other very large cases reported in the literature are discussed.

Review of Junior Resident Plain Film Reporting and Audit in Singapore.

BACKGROUND: Graduate medical education in Singapore recently underwent significant restructuring, leading to the accreditation of residency programs by the Accreditation Council for Graduate Medical Education-International (ACGME-I). In radiology, this involved a change in teaching and quality assurance of plain film (PF) reporting. PF reported by junior residents (postgraduate year 1-3) are subject to a 50% random audit. To date, national data on junior resident performance in PF reporting have not been published. OBJECTIVE: We reviewed performance in PF reporting under the current teaching and audit framework. METHODS: Retrospective review of junior resident reported PF audit data from all 3 radiology residency programs in Singapore. The number of residents audited, number of PF reported and audited, and major discrepancy rates were analyzed. RESULTS: On average, 86 440 PF were audited annually nationwide from an estimated 184 288 junior resident-reported PF. Each program trained between 4 to 24 junior residents annually (mean 15), averaging about 44 each year nationwide. A mean of 28 813 PF were audited annually in each program (range 4355-50 880). An estimated mean of 4148 PF (range 1452-9752) were reported per junior resident per year, about 346 PF per month. The major discrepancy rate ranged from 0.04% to 1.13% (mean 0.34%). One resident required remediation in the study period. CONCLUSIONS: Structured residency training in Singapore has produced a high level of junior resident competency in PF interpretation.

Convolutional Neural Networks for Automated Fracture Detection and Localization on Wrist Radiographs.

PURPOSE: To demonstrate the feasibility and performance of an object detection convolutional neural network (CNN) for fracture detection and localization on wrist radiographs. MATERIALS AND METHODS: Institutional review board approval was obtained with waiver of consent for this retrospective study. A total of 7356 wrist radiographic studies were extracted from a hospital picture archiving and communication system. Radiologists annotated all radius and ulna fractures with bounding boxes. The dataset was split into training (90%) and validation (10%) sets and used to train fracture localization models for frontal and lateral images. Inception-ResNet Faster R-CNN architecture was implemented as a deep learning model. The models were tested on an unseen test set of 524 consecutive emergency department wrist radiographic studies with two radiologists in consensus as the reference standard. Per-fracture, per-image (ie, per-view), and per-study sensitivity and specificity were determined. Area under the receiver operating characteristic curve (AUC) analysis was performed. RESULTS: The model detected and correctly localized 310 (91.2%) of 340 and 236 (96.3%) of 245 of all radius and ulna fractures on the frontal and lateral views, respectively. The per-image sensitivity, specificity, and AUC were 95.7% (95% confidence interval [CI]: 92.4%, 97.8%), 82.5% (95% CI: 77.4%, 86.8%), and 0.918 (95% CI: 0.894, 0.941), respectively, for the frontal view and 96.7% (95% CI: 93.6%, 98.6%), 86.4% (95% CI: 81.9%, 90.2%), and 0.933 (95% CI: 0.912, 0.954), respectively, for the lateral view. The per-study sensitivity, specificity, and AUC were 98.1% (95% CI: 95.6%, 99.4%), 72.9% (95% CI: 67.1%, 78.2%), and 0.895 (95% CI: 0.870, 0.920), respectively. CONCLUSION: The ability of an object detection CNN to detect and localize radius and ulna fractures on wrist radiographs with high sensitivity and specificity was demonstrated.© RSNA, 2019.

Sequelae of bilateral luxatio erecta in the acute post-reduction period demonstrated by MRI: a case report and literature review.

Luxatio erecta humeri (LEH), also known as inferior shoulder dislocation, is uncommon, comprising about 0.5% of all cases of shoulder dislocation. Synchronous bilateral LEH is exceedingly rare and, to our knowledge, there are no descriptions of axillary nerve injury on magnetic resonance imaging (MRI) following LEH. We present a case of traumatic bilateral LEH in a 59-year-old woman who fell from a fast-moving mobility scooter and sustained direct axial loading forces on the fully abducted shoulders. Both shoulders were successfully reduced using the traction-countertraction technique in the emergency department. In this article, we describe the characteristic features of LEH on plain radiography and the pattern of acute soft-tissue injuries on MRI. We emphasize the importance of reviewing the axillary neurovascular bundle, which by virtue of its location beneath the shoulder joint, is prone to injury in inferior shoulder dislocation and thus has a substantial impact on functional recovery. This important complication is unfortunately not routinely examined by radiologists, partly because of the paucity of literature highlighting its clinical significance.