The concept of Skin Model Shapes has been proposed as a method to generate digital twins of manufactured parts and is a new paradigm in the design and manufacturing industry. Skin Model Shapes use discrete surface representation schemes, such as meshes and point clouds, to represent surfaces, which makes them enablers to perform an accurate tolerance analysis and surface inspection. However, online inspection of manufactured parts through use of Skin Model Shapes has not been extensively studied. Moreover, the existing geometric variation inspection techniques do not detect unfamiliar changes within tolerance, which could be the precursors to the onset of the manufacturing of out of tolerance part. To detect the unfamiliar changes, as anomalies, and categorize them as systematic and random variations, some unique surface characteristics can be extracted and studied. Random surface deviations exhibit narrow normal distributions, and systematic deviations, on the other hand, exhibit wide, skewed, and multimodal distributions. Using those surface characteristics as key traits, machine learning classifiers can be used to classify deviations into systematic and random variations. To illustrate the method, multiple samples from a truck component manufacturing line were scanned and the collected 3D point cloud data was used to extract features. A prediction score of 97-100% can be achieved by decision tree, k-nearest neighbor, support vector machines, and ensemble classifiers. The purposed approach is expected to extend the existing online inspection approaches and applications of Skin Model Shapes in quality control.

The concept of Skin Model Shape has been introduced as a method for a close representation of manufactured parts using a discrete geometry representation scheme. However, discretized surfaces make irregular polyhedra, which are computationally demanding to model and process using the traditional implicit surface and boundary representation techniques. Moreover, there are still some research challenges related to the geometrical variation modelling of manufactured products; specifically, methods for geometrical data processing, the mapping of manufacturing variation sources to a geometric model, and the improvement of variation visualization techniques. To provide steps towards addressing these challenges this work uses Octree, a 3D space partitioning technique, as an aid for geometrical data processing, variation visualization, variation modelling and propagation, and tolerance analysis. Further, Skin Model Shapes are generated either by manufacturing a simulation using a non-ideal toolpath on solid models of Skin Model Shapes that are assembled to non-ideal fixtures or from measurement data. Octrees are then used in a variation envelope extraction from the simulated or measurement data, which becomes a basis for further simulation and tolerance analysis. To illustrate the method, an industrial two-stage truck component manufacturing line was studied. Simulation results show that the predicted Skin Model Shapes closely match to the measurement data from the manufacturing line, which could also be used to map to manufacturing error sources. This approach contributes towards the application of Octrees in many Skin Model Shape related operations and processes. of 21However, the computational cost scales up with mesh density and the number of sampled points per part. The meshes and the reconstructed point clouds form irregular polyhedra, whose representation, operation, and manipulation, based on implicit surfaces and boundary representation techniques, is computationally slow and memory intensive [5][6][7]. Since the prime aim of utilizing SMSs is to get a detailed digital representation of parts, computational efficacy of SMS modelling and operations is crucial. As an alternative, an approach based on a 3D space partitioning technique, using Octrees, has been proven to significantly improve computation time and memory in manipulation and processing of irregular polyhedra [8][9][10]. This work utilizes the computational efficacy of Octrees, in one hand, and their capability to localize regions of form errors, in the other hand, in the generation and variation analysis of SMSs.Moreover, despite many contributions in SMS generation methods and associated operations, there are still some challenges that need to be addressed; specifically, the mapping of manufacturing variation sources to geometric models, the development of geometrical data processing methods applicable in different stages of variation modelling, and the improvement of variation visualization techniques [11]. To address these challenges, in the context...

The high incidence rates of basal cell carcinoma (BCC) cause a significant burden at pathology laboratories. The standard diagnostic process is time-consuming and prone to inter-pathologist variability. Despite the application of deep learning approaches in grading of other cancer types, there is limited literature on the application of vision transformers to BCC on whole slide images (WSIs). A total of 1831 WSIs from 479 BCCs, divided into training and validation (1434 WSIs from 369 BCCs) and testing (397 WSIs from 110 BCCs) sets, were weakly annotated into four aggressivity subtypes. We used a combination of a graph neural network and vision transformer to 1) detect the presence of tumor (two classes), 2) classify the tumor into low and high-risk subtypes (three classes), and 3) classify four aggressivity subtypes (five classes). Using an ensemble model comprised of the models from cross-validation, accuracies of 93.5%, 86.4%, and 72% were achieved on two, three, and five class classifications, respectively. These results show high accuracy in both tumor detection and grading of BCCs. The use of automated WSI analysis could increase workflow efficiency and possibly overcome inter-pathologist variability.

Variation propagation models play an important role in part quality prediction, variation source identification, and variation compensation in multistage manufacturing processes. These models often use homogenous transformation matrix, differential motion vector, and/or Jacobian matrix to represent and transform the part, tool and fixture coordinate systems and associated variations. However, the models end up with large matrices as the number features and functional element pairs increase. This work proposes a novel strategy for modelling of variation propagation in multistage machining processes using dual quaternions. The strategy includes representation of the fixture, part, and toolpath by dual quaternions, followed by projection locator points onto the features, which leads to a simplified model of a part-fixture assembly and machining. The proposed approach was validated against stream of variation models and experimental results reported in the literature. This paper aims to provide a new direction of research on variation propagation modelling of multistage manufacturing processes.

No abstract