Purpose: This retrospective study aimed to evaluate the safety and efficacy of MIGS (Xen® and Cypass®) compared to trabeculectomy and Baerveldt® implants. Patients and Methods: This single-center study included patients from the Ophthalmic Hospital of Turin between January 2015 and 2018. Efficacy was assessed based on the intraocular pressure and the number of medications necessary to control IOP at 1, 3, 6, 12, 24, and 36 months and the potential complications. Complete surgical success was defined as IOP ≤ 21 mmHg, with no medication or additional glaucoma surgery. Qualified success was defined as IOP ≤ 21 mmHg with the reintroduction of topical medications. Our third measure of success was a delta IOP reduction greater than 20% compared to baseline. Results: The delta IOP compared to the baseline was significant at every instance, except in the Xen® implant group. The average IOP distribution at different time points was significant, except at 1 and 12 months. The delta of the medication number compared to the baseline was significant at every instant. All three methods of success were analyzed using a Kaplan-Meier survival curve that identified a significant difference on the success rates of Xen®-trabeculectomy, Xen®-Baerveldt®, and Cypass®-Baerveldt®. Conclusion: MIGS can reduce the number of medication classes, thus increasing patient compliance and reducing topical ocular toxicity and general costs. Trabeculectomy remains the most effective option but still presents a percentage of failures that might increase during follow-up.
Background. The preoperative differential diagnosis of myometrial lesions remains a significant challenge when using conventional imaging techniques, such as ultrasound (US) and magnetic resonance imaging (MRI). Radiomics and machine learning, which leverage quantitative features beyond human visual perception, are increasingly recognized as promising tools for improving differential diagnosis in gynecology. Methods. This retrospective study included patients who underwent surgery for uterine masses and had preoperative MR. A machine learning model was developed to analyze radiomic features extracted from T2-weighted and diffusion-weighted MR images. Results. 44 subjects were included: 19 (43.2%) classified as "sarcoma" and 25 (56.8%) as "fibroid" based on histology after surgery. This dataset was used for training and cross-validation of different models. Three models, comprising ensembles of machine learning classifiers (random forests, support vector machines, and k-nearest neighbors), were developed for binary classification using histological diagnosis as reference standard. The best-performing model achieved the following results: AUC 90%, accuracy 82%, sensitivity 95%, specificity 72%, PPV 72%, and NPV 95%. Conclusions. Our model demonstrated high sensitivity and moderate accuracy, suggesting its potential as a valuable tool for assisting clinicians in the preliminary assessment of myometrial lesions and guiding decision-making toward conservative management in cases of non-suspicious masses.