Focal choroidal excavation associated with nonmelanocytic iris tumor

Purpose To describe a unique unilateral association between an iris stromal tumor and a macular focal choroidal excavation. Case Description A 40-year old patient presented with a small iris tumor associated with a unilateral macular lesion disclosed during a routine ophthalmologic examination. The patient was asymptomatic and visual function was not affected. After clinical and instrumental evaluation, a diagnosis of nonmelanocytic undefined stromal tumor of the iris associated with macular focal choroidal excavation was made. The size and shape of the two lesions remained stable during a 7-year follow-up and the patient did not develop other signs. Conclusion The concurrent presence of a stromal iris tumor associated with focal choroidal excavation has never been reported. Further reports of this association are required in order to understand its exact pathogenesis.

Deep Learning in Opthalmology: Iris Melanocytic Tumor Intelligent Diagnosis

Recently, Convolutional neural networks (CNN) have shown a growth due to their ability of learning different level image representations that helps in image classification in different fields. These networks have been trained on millions of images, so they gained a powerful ability of extracting the rightful features from input images, which results in accurate classification. In this research, we investigate the effects of transfer learning based convolutional neural networks for the iris tumor malignancy identification as it is notoriously hard to distinguish an iris nevus from an iris tumor. Features are transferred from a CNN trained on a source task, i.e. ImageNet, to a target task, i.e. iris tumor datasets. We transfer features learned from AlexNet and VGG-16 that are trained on ImageNet, to classify three different iris images types which are: iris nevus unaffected, iris cysts, and iris melanocytic tumors. The employed pre-trained models are modified by replacing their feedforward neural network classifier, Softmax, by a support vector machine (SVM) that is expected to slightly boost their performance (AlexNet-SVM and VGG16-SVM). All employed models are trained (fine-tuned) on a 60% of the available large dataset of iris images in order to investigate their power of generalization when trained using large amount of data. The networks are also tested on 40% of the data. The best performance was achieved by the VGG16-SVM which scored a high accuracy of 96.27% and strong features extraction capability as compared to the other models. Experimentally, it was seen that adding SVM contributed in improving the network performance compared to original models which use a feedforward neural network classifier.

Iris Melanoma in a Patient with Retinitis Pigmentosa

This report describes the histopathological findings of an iris melanoma that developed in a patient with retinitis pigmentosa (RP). A 66-year-old man was referred to our hospital with a diagnosis of iris tumor. He had a medical history of RP for over 25 years. His best corrected visual acuity at presentation was 20/20 (OD) and 20/200 (OS). Slit lamp examination revealed an iris melanocytoma in his right eye and a brown iris mass involving the pupil in his left eye. Ocular fundus examination of his right eye showed diffuse chorioretinal atrophy with attenuated retinal vessels and scattered pigment. A diagnosis of iris melanoma was made and enucleation of his left eyeball was performed. Histopathological examination of the eyeball showed an iris tumor with proliferation of spindle cells positive for Melan-A, HMB-45, and S-100 protein by immunohistochemistry. Diffuse destruction of retinal architecture was observed, with loss of outer segment and replacement by glial cells. Accumulation of melanin pigment around retinal vessels was found in peripheral retina. These histopathological findings were compatible with advanced stage of RP. This case demonstrates that iris melanoma can occur in eye with RP.

The Effectiveness of Ultrasound Biomicroscopic and Anterior Segment Optical Coherence Tomography in the Assessment of Anterior Segment Tumors: Long-Term Follow-Up

Background. Differential diagnosis and follow-up of small anterior segment tumors constitute a particular challenge because they determine further treatment procedures. The aim of this study was to evaluate the efficacy of the UBM (ultrasound biomicroscopy) and AS-OCT (anterior segment optical coherent tomography) in distinguishing different types of anterior segment lesions. Methods. It was a retrospective, noncomparative study of case series of 89 patients with the suspicion of anterior segment tumor referred to the Ophthalmology Clinic, Medical University of Białystok, Poland, between 2016 and 2020. UBM was used to assess tumor morphology including height, location, and internal and external features. In cases in which UBM did not provide enough data, the AS-OCT images were analyzed. The data on demographics, best corrected visual acuity (BCVA), intraocular pressure (IOP), and rate of complications were also collected. Patients were followed up from 1 to 48 months. Results. The mean observation period was 26.61 ± 16.13 months. Among the patients, there were 62 women and 27 men at a mean age of 55.59 ± 19.48 (range: from 20 to 89 years.) The types of tumors were cysts (41%), solid iris tumors (37.1%), ciliary body tumors (7.9%), peripheral anterior synechiae (PAS 3.4%), corneal tumors (4.5%), and others (5.6%). Patients with cysts were younger than patients with solid iris tumor (p=0.002). Women had a cyst as well as solid iris tumor more frequently than men, but less often a ciliary body tumor (p<0.05). The horizontal size of tumor was positively correlated with patients’ age (rs = 0.38 and p=0.003) and negatively correlated with visual acuity (rs = −0.42 and p=0.014). During the 4 years of diagnosis, only 2.2% of lesions exhibited growth (growth rate of 0.02 mm per year). Among 15 cases in which visualization with UBM was not satisfactory (mostly iris nevi), AS-OCT was helpful in diagnosis of 13 patients. Conclusions. Both UBM and AS-OCT are effective methods in detection and diagnosis of tumors of the anterior eye segment, but in some cases, AS-OCT adds additional value to the diagnosis. Many lesions can be managed conservatively because they did not demonstrate growth during 4 years of the follow-up period.