High-Content Analysis of Breast Cancer Using Single-Cell Deep Transfer Learning

High-content analysis has revolutionized cancer drug discovery by identifying substances that alter the phenotype of a cell, which prevents tumor growth and metastasis. The high-resolution biofluorescence images from assays allow precise quantitative measures enabling the distinction of small molecules of a host cell from a tumor. In this work, we are particularly interested in the application of deep neural networks (DNNs), a cutting-edge machine learning method, to the classification of compounds in chemical mechanisms of action (MOAs). Compound classification has been performed using image-based profiling methods sometimes combined with feature reduction methods such as principal component analysis or factor analysis. In this article, we map the input features of each cell to a particular MOA class without using any treatment-level profiles or feature reduction methods. To the best of our knowledge, this is the first application of DNN in this domain, leveraging single-cell information. Furthermore, we use deep transfer learning (DTL) to alleviate the intensive and computational demanding effort of searching the huge parameter's space of a DNN. Results show that using this approach, we obtain a 30% speedup and a 2% accuracy improvement.

Strength and damage analysis of composite-aluminium adhesively-bonded single-lap joints

With the need to find an alternative way to mechanical and welding joints, and at the same time to overcome some limitations linked to these traditional techniques, adhesive bonds can be used. Adhesive bonding is a permanent joining process that uses an adhesive to bond the components of a structure. Composite materials reinforced with fibres are becoming increasingly popular in many applications as a result of a number of competitive advantages. In the manufacture of composite structures, although the fabrication techniques reduce to the minimum by means of advanced manufacturing techniques, the use of connections is still required due to the typical size limitations and design, technological and logistical aspects. Moreover, it is known that in many high performance structures, unions between composite materials with other light metals such as aluminium are required, for purposes of structural optimization. This work deals with the experimental and numerical study of single lap joints (SLJ), bonded with a brittle (Nagase Chemtex Denatite XNRH6823) and a ductile adhesive (Nagase Chemtex Denatite XNR6852). These are applied to hybrid joints between aluminium (AL6082-T651) and carbon fibre reinforced plastic (CFRP; Texipreg HS 160 RM) adherends in joints with different overlap lengths (LO) under a tensile loading. The Finite Element (FE) Method is used to perform detailed stress and damage analyses allowing to explain the joints’ behaviour and the use of cohesive zone models (CZM) enables predicting the joint strength and creating a simple and rapid design methodology. The use of numerical methods to simulate the behaviour of the joints can lead to savings of time and resources by optimizing the geometry and material parameters of the joints. The joints’ strength and failure modes were highly dependent on the adhesive, and this behaviour was successfully modelled numerically. Using a brittle adhesive resulted in a negligible maximum load (Pm) improvement with LO. The joints bonded with the ductile adhesive showed a nearly linear improvement of Pm with LO.