Event models for tumor classification with SAGE gene expression data

Serial Analysis of Gene Expression (SAGE) is a relatively new method for monitoring gene expression levels and is expected to contribute significantly to the progress in cancer treatment by enabling a precise and early diagnosis. A promising application of SAGE gene expression data is classification of tumors. In this paper, we build three event models (the multivariate Bernoulli model, the multinomial model and the normalized multinomial model) for SAGE data classification. Both binary classification and multicategory classification are investigated. Experiments on two SAGE datasets show that the multivariate Bernoulli model performs well with small feature sizes, but the multinomial performs better at large feature sizes, while the normalized multinomial performs well with medium feature sizes. The multinomial achieves the highest overall accuracy.

Development of an in-situ, non-destructive ultrasonic monitoring technique for solder pastes

This paper concerns the use of a non-destructive ultrasonic technique for characterising the rheological properties of solder paste and specifically, the use of through-mode microsecond ultrasonic pulses for evaluation of viscoelastic properties of paste materials at the molecular level. Ultrasonic techniques are a widely used and a reliable form of non-destructive testing of materials. This is because techniques such as ultrasounds while used for testing or monitoring material properties, has offered immense benefits in applications where access to the sample is restricted or when handling the sample for testing could interfere with the monitoring or analysis process. Very often, this would mean that the measurements taken are not a true representation of the behaviour of the material (due to externally incorporated changes into the material's physical state during the removal or testing process). Ultrasonic based techniques are being increasingly used for quality control and production monitoring functions which requires evaluation of the changes in material properties over wide range of industrial applications such as cement paste quality, plastic/polymer extrusion process, dough, and even sugar content in beverage drinks. In addition, ultrasound techniques are of great interest for their capacity to take rapid measurements in systems which are optically opaque. The viscometer and rheometer are two of the most widely used rheological instruments used in industry for monitoring the quality of solder pastes, during the production and packaging stage. One of the potential limitations of viscometer and rheometer based measurements is that the collection and preparation of the solder paste samples can irreversibly alter the structure and flow behaviour of the sample. Hence the measurement may not represent the actual quality of the whole production batch. Secondly, rheological measurements and the interpretation of rheological data is a very technical and time consuming process, which requires professionally trained R&D personnel. It is for these reasons that materials suppliers (who formulate and produce solder pastes) and solder paste consumers (especially, contract electronics manufacturers) are keen to see the development of simple, easy to use and accurate techniques for the theological characterisation of solder pastes. The results from the work show that the technique can be used by R&D personnel involved in paste formulation and manufacture to monitor the batch-to-batch quality and consistency.

In-situ non-destructive ultrasonic rheology technique for monitoring different lead-free solder pastes for surface mount applications

This paper investigates the application of a non-destructive ultrasonic technique for characterising the rheological properties of solder paste through the use of through-mode microsecond ultrasonic pulses for evaluation of viscoelastic properties of lead-free solder paste containing different types of flux. Ultrasonic techniques offer a robust and reliable form of non-destructive testing of materials where access to the sample is restricted or when sample handling can interfere with the monitoring or analysis process due to externally incorporated changes to the material’s physical state or accidental contamination during the removal or testing process. Ultrasonic based techniques are increasingly used for quality control and production monitoring functions which requires evaluation of changes in material properties for a wide range of industrial applications such as cement paste quality, plastic/polymer extrusion process, dough and even sugar content in beverage drinks. In addition, ultrasound techniques are of great interest for their capability to take rapid measurements in systems which are optically opaque. The conventional industry approach for characterising the rheological properties of suspensions during processing/packaging stage is mainly through the use of viscometer and some through the use of rheometer. One of the potential limitations of viscometer and rheometer based measurements is that the collection and preparation of the solder paste samples can irreversibly alter the structure and flow behaviour of the sample. Hence the measurement may not represent the actual quality of the whole production batch. Secondly, rheological measurements and the interpretation of rheological data is a very technical and time consuming process, which requires professionally trained R&D personnel. The ultrasound technique being proposed provides simple, yet accurate and easy to use solution for the in-situ rheological characterisation of solder pastes which will benefit the materials suppliers (who formulate and produce solder pastes) and solder paste consumers (especially, contract electronics manufacturers). The results from the work show that the technique can be used by R&D personnel involved in paste formulation and manufacture to monitor the batch-to-batch quality and consistency.