MV-MS-FETE: Multi-view multi-scale feature extractor and transformer encoder for stenosis recognition in echocardiograms.

BACKGROUND: aortic stenosis is a common heart valve disease that mainly affects older people in developed countries. Its early detection is crucial to prevent the irreversible disease progression and, eventually, death. A typical screening technique to detect stenosis uses echocardiograms; however, variations introduced by other tissues, camera movements, and uneven lighting can hamper the visual inspection, leading to misdiagnosis. To address these issues, effective solutions involve employing deep learning algorithms to assist clinicians in detecting and classifying stenosis by developing models that can predict this pathology from single heart views. Although promising, the visual information conveyed by a single image may not be sufficient for an accurate diagnosis, especially when using an automatic system; thus, this indicates that different solutions should be explored. METHODOLOGY: following this rationale, this paper proposes a novel deep learning architecture, composed of a multi-view, multi-scale feature extractor, and a transformer encoder (MV-MS-FETE) to predict stenosis from parasternal long and short-axis views. In particular, starting from the latter, the designed model extracts relevant features at multiple scales along its feature extractor component and takes advantage of a transformer encoder to perform the final classification. RESULTS: experiments were performed on the recently released Tufts medical echocardiogram public dataset, which comprises 27,788 images split into training, validation, and test sets. Due to the recent release of this collection, tests were also conducted on several state-of-the-art models to create multi-view and single-view benchmarks. For all models, standard classification metrics were computed (e.g., precision, F1-score). The obtained results show that the proposed approach outperforms other multi-view methods in terms of accuracy and F1-score and has more stable performance throughout the training procedure. Furthermore, the experiments also highlight that multi-view methods generally perform better than their single-view counterparts. CONCLUSION: this paper introduces a novel multi-view and multi-scale model for aortic stenosis recognition, as well as three benchmarks to evaluate it, effectively providing multi-view and single-view comparisons that fully highlight the model's effectiveness in aiding clinicians in performing diagnoses while also producing several baselines for the aortic stenosis recognition task.

MiR-200c sensitizes Olaparib-resistant ovarian cancer cells by targeting Neuropilin 1.

BACKGROUND: Ovarian cancer (OC) is the most lethal gynecological malignancy and the second leading cause of cancer-related death in women. Treatment with PARP inhibitors (PARPi), such as Olaparib, has been recently introduced for OC patients, but resistance may occur and underlying mechanisms are still poorly understood. The aim of this study is to identify target genes within the tumor cells that might cause resistance to Olaparib. We focused on Neuropilin 1 (NRP1), a transmembrane receptor expressed in OC and correlated with poor survival, which has been also proposed as a key molecule in OC multidrug resistance. METHODS: Using three OC cell lines (UWB, UWB-BRCA and SKOV3) as model systems, we evaluated the biological and molecular effects of Olaparib on OC cell growth, cell cycle, DNA damage and apoptosis/autophagy induction, through MTT and colony forming assays, flow cytometry, immunofluorescence and Western blot analyses. We evaluated NRP1 expression in OC specimens and cell lines by Western blot and qRT-PCR, and used RNA interference to selectively inhibit NRP1. To identify miR-200c as a regulator of NRP1, we used miRNA target prediction algorithms and Pearsons' correlation analysis in biopsies from OC patients. Then, we used a stable transfection approach to overexpress miR-200c in Olaparib-resistant cells. RESULTS: We observed that NRP1 is expressed at high levels in resistant cells (SKOV3) and is upmodulated in partially sensitive cells (UWB-BRCA) upon prolonged Olaparib treatment, leading to poor drug response. Our results show that the selective inhibition of NRP1 is able to overcome Olaparib resistance in SKOV3 cells. Moreover, we demonstrated that miR-200c can target NRP1 in OC cells, causing its downmodulation, and that miR-200c overexpression is a valid approach to restore Olaparib sensitivity in OC resistant cells. CONCLUSIONS: These data demonstrate that miR-200c significantly enhanced the anti-cancer efficacy of Olaparib in drug-resistant OC cells. Thus, the combination of Olaparib with miRNA-based therapy may represent a promising treatment for drug resistant OC, and our data may help in designing novel precision medicine trials for optimizing the clinical use of PARPi.