Automatic classification of HEp-2 specimens by explainable deep learning and Jensen-Shannon reliability index.

The Anti-Nuclear Antibodies (ANA) test using Human Epithelial type 2 (HEp-2) cells in the Indirect Immuno-Fluorescence (IIF) assay protocol is considered the gold standard for detecting Connective Tissue Diseases. Computer-assisted systems for HEp-2 image analysis represent a growing field that harnesses the potential offered by novel machine learning techniques to address the classification of HEp-2 images and ANA patterns. In this study, we introduce an innovative platform based on transfer learning with pre-trained deep learning models.

State-Space Model for Arrival Time Simulations and Methodology for Offline Blade Tip-Timing Software Characterization.

Blade tip-timing is an extensively used technique for measuring blade vibrations in turbine and compressor stages; it is one of the preferred techniques used for characterizing their dynamic behaviors using non-contact probes. Typically, arrival time signals are acquired and processed by a dedicated measurement system. Performing a sensitivity analysis on the data processing parameters is essential for the proper design of tip-timing test campaigns.

Experimental Investigation on Hardware and Triggering Effect in Tip-Timing Measurement Uncertainty.

Non-destructive testing for structural health monitoring is becoming progressively important for gas turbine manufacturers. As several techniques for diagnostics and condition-based maintenance have been developed over the years, the tip-timing approach is one of the preferred approaches for characterizing the dynamic behavior of turbine blades using non-contact probes. This experimental work investigates the uncertainty of the time-of-arrival of a Blade Tip-Timing measurement system, a fundamental requirement for numerical and aeromechanical modeling validation.