Publications by authors named "A Lebovici"
Background And Aims: The conventional computed tomography (CT) appearance of ovarian cystic masses is often insufficient to adequately differentiate between benign and malignant entities. This study aims to investigate whether texture analysis of the fluid component can augment the CT diagnosis of ovarian cystic tumors.
Methods: Eighty-four patients with adnexal cystic lesions who underwent CT examinations were retrospectively included.
Background: The accurate discrimination of uterine leiomyosarcomas and leiomyomas in a pre-operative setting remains a current challenge. To date, the diagnosis is made by a pathologist on the excised tumor. The aim of this study was to develop a machine learning algorithm using radiomic data extracted from contrast-enhanced computed tomography (CECT) images that could accurately distinguish leiomyosarcomas from leiomyomas.
View Article and Find Full Text PDFArticle Synopsis
- * A systematic review of literature identified 26 studies focusing on MRI-based radiomics for bladder cancer, with applications in preoperative staging, predicting tumor grade, and assessing treatment response, primarily utilizing second-order features from filtered images.
- * While MRI-based radiomics shows potential as a quantitative method for bladder cancer characterization and prognosis, there is a critical need for standardization and validation of the techniques before clinical implementation.
Introduction: Prostate magnetic resonance imaging (MRI) has been recently integrated into the pathway of diagnosis of prostate cancer (PCa). However, the lack of an optimal contrast-to-noise ratio hinders automatic recognition of suspicious lesions, thus developing a solution for proper delimitation of the tumour and its separation from the healthy parenchyma, which is of primordial importance.
Method: As a solution to this unmet medical need, we aimed to develop a decision support system based on artificial intelligence, which automatically segments the prostate and any suspect area from the 3D MRI images.
Article Synopsis
- Bladder MRI is now used for diagnosing bladder cancer, but automatically identifying suspicious lesions is challenging, prompting the need for an AI-based solution to segment tumors and healthy tissue in 3D MRI images.
- The study assessed 33 patients, where radiologists manually segmented MRIs and compared it to an automated model based on a 3D U-Net architecture with various training setups and data augmentations.
- Results showed that the best model achieved a Dice coefficient of 0.902, indicating strong performance in segmenting the bladder wall, but improvements in tumor segmentation were less significant with larger training datasets.