In the recent years, due to the technological advancements brought by Industry 4.0, flexible in-line quality control processes that analyze 100% of produced parts targeting Zero Defect Manufacturing have become a reality. The electronics industry in general and the manufacturing of PCBAs in particular have also adopted this paradigm, where quality inspection applies especially to the most critical and error-prone feature: the solder joints between PCBs and components. When the literature related to AI-based solder joint quality inspection is analyzed, we find a lack of applications in real production lines, since most of them are limited to lab scenarios. Therefore, we present a deep learning-based method for solder joint defect detection, which is deployed in a real manufacturing line. The mentioned system is trained using the database we captured using an ad hoc robotic system with onboard front and lateral cameras designed also in this work, which we also describe in detail. The deployment of the defect detector in the real production line is done using that same robotic system too, ensuring the coherence and compatibility between the training data and the real inference scenario.

In this work, a vision system oriented to the quality inspection of solder joints in electronic boards is presented. The proposed vision system is composed of two cameras (one frontal and one lateral), to achieve a general view of the joints; a light source, to ensure good and robust lighting conditions; and a mobile system (i.e. 3-axis cartesian robot), to automatically move to each image capture position and get the optimal focus. Moreover, a classifier based on Artificial Intelligence is fed with the captured images to perform an automatic inspection of the soldering joints. The output for each one of them is the belonging to either a correct or incorrect joint group. The tests carried out with real samples show the validity of the proposed system for its future deployment in the factory.

Pancreatic ductal adenocarcinoma (PDAC) is characterized by a dense desmoplastic extracellular matrix (ECM) that contributes to tumor progression, therapeutic resistance, and poor patient survival. However, the relationship between in vivo imaging-derived mechanical properties, ex vivo tissue biomechanics, ECM architecture, and cellularity remains incompletely understood. Here, we combined pre-operative in vivo clinical magnetic resonance elastography (MRE) with ex vivo biomechanical testing of fresh human PDAC tissue and histopathological analyses. Nine patients undergoing pancreatic resection were prospectively enrolled. Quantitative MRE was performed pre-operatively to assess tissue stiffness through shear wave speed (c) and relative viscosity or fluidity through the loss angle (φ). Fresh tumor and adjacent non-malignant tissue biopsies were subsequently analyzed ex vivo by unconfined uniaxial compression testing to determine elastic moduli and stress relaxation halftime. Histological analyses quantified collagen-rich fibrous tissue area, cell nuclei density, and nuclear morphology. Tumor tissue exhibited significantly increased stiffness and collagen fraction compared with adjacent non-malignant tissue, together with reduced cellularity, smaller nuclear area and more elongated nuclei. Ex vivo stiffness positively correlated with collagen content and negatively correlated with patient survival. Reduced stress relaxation halftime, indicative of increased tissue viscosity, was associated with lower cellularity and elongated nuclei. Importantly, pre-operative MRE parameters of the intact surrounding environment correlated significantly with ex vivo tumor mechanics, cellular organization, and survival. Specifically, a softer and less viscous surrounding environment was associated with stiffer and more viscous tumors, with lower cellularity and elongated nuclei, and poorer prognosis. These findings demonstrate that MRE-derived mechanical biomarkers reflect underlying ECM remodeling and tumor mechanobiology in PDAC. Integrating in vivo imaging with ex vivo tissue mechanics and histopathology may improve non-invasive disease characterization and support biomechanically-informed therapeutic strategies.

1. The use of robotic systems in industry, such as static robot arms in robotic cells or mobile manipulators, has increased in the last decade thanks to the improved precision, speed and robustness...