Over the last years Deep Learning has shown to yield remarkable results when compared to traditional computer vision algorithms, in a large variety of computer vision applications. The deeplearning models outperformed in both accuracy and processing time. Thus, once a deeplearning models won the Image Net Large Scale Visual Recognition Contest, it proved that this area of research is of great potential. Furthermore, these increases in recognition performance resulted in more applied research and thus, more applications where deeplearning is useful: one of which is defect detection (or visual defect detection). In the last few years, deeplearning models achieved higher and higher accuracy on the complex testing datasets used for benchmarking. This surge in accuracy and usage is also supported (besides swarms of researchers pouring into the race), by incremental breakthroughs in computing hardware: such as more powerful GPUs(Graphical processing units), CPUs(central processing units) and better computing procedures (libraries and frameworks). To offer a structured and analytical overview(stating both advantages and disadvantages) of the existing popular object detection models that can be re-purposed for defect detection: such as Region based CNNs(Convolutional neural networks), YOLO(You only look once), SSD(single shot detectors) and cascaded architectures. A further brief summary on model compression and acceleration techniques that enabled the portability of deeplearning detection models is included. It is of great use for future developments in the manufacturing industry that many of the popular, above mentioned models are easy to re-purpose for defect detection and, thus could really contribute to the overall increase in productivity of this sector. Moreover, in the experiment performed the YOLOv4 model was trained and re-purposed for industrial cable detection in several hours. The computing needs could be fulfilled by a general purpose computer or by a high-performance desktop setup, depending on the specificity of the application. Hence, the barrier of computing shall be somewhat easier to climb for all types of businesses.

Organ-on-a-chip (OoC), also known as micro physiological systems or “tissue chips” have attracted substantial interest in recent years due to their numerous applications, especially in precision medicine, drug development and screening. Organ-on-a-chip devices can replicate key aspects of human physiology, providing insights into the studied organ function and disease pathophysiology. Moreover, these can accurately be used in drug discovery for personalized medicine. These devices present useful substitutes to traditional preclinical cell culture methods and can reduce the use of in vivo animal studies. In the last few years OoC design technology has seen dramatic advances, leading to a wide range of biomedical applications. These advances have also revealed not only new challenges but also new opportunities. There is a need for multidisciplinary knowledge from the biomedical and engineering fields to understand and realize OoCs. The present review provides a snapshot of this fast-evolving technology, discusses current applications and highlights advantages and disadvantages for biomedical approaches.

The use of agricultural and food by-products is an economical solution to industrial biotechnology. The apricot press residues are abounding by-products from juice industry which can be used as substrates in solid state fermentation process (SSF), thus allowing a liberation and increase of content from various biomolecules with high added value. The evolutions of phenolic levels (by colorimetric assays and high performance liquid chromatography, HPLC–MS) and antioxidant activities (by DPPH assay) during SSF of apricot pomaces with Aspergillus niger and Rhizopus oligosporus were investigated. The changes in fatty acid compositions of oils in apricot kernels during SSFs were also analyzed by gas chromatography (GC–MS). The results showed that the levels of total phenolics increased by over 70% for SSF with R. oligosporus and by more than 30% for SSF with A. niger. A similar trend was observed in the amounts of total flavonoids (increases of 38, and 12% were recorded for SSF by R. oligosporus and A. niger, respectively). Free radical scavenging capacities of methanolic extracts were also significantly enhanced. The main phenolic compounds identified through HPLC–MS in fermented apricot press residues were chlorogenic acid, neochlorogenic acid, rutin, and quercetin 3-acetyl- glucoside. This work also demonstrated that the SSF with filamentous fungal strains not only helped in higher lipid recovery from apricot kernels, but also resulted in oils with better quality attributes (high linoleic acid content). The utilization of apricot by-products resulting from the juice industry as waste could provide an extra income and at the same time can help in solving solid waste management problems

This paper presents two tuning algorithms for fractional-order internal model control (IMC) controllers for time delay processes. The two tuning algorithms are based on two specific closed-loop control configurations: the IMC control structure and the Smith predictor structure. In the latter, the equivalency between IMC and Smith predictor control structures is used to tune a fractional-order IMC controller as the primary controller of the Smith predictor structure. Fractional-order IMC controllers are designed in both cases in order to enhance the closed-loop performance and robustness of classical integer order IMC controllers. The tuning procedures are exemplified for both single-input-single-output as well as multivariable processes, described by first-order and second-order transfer functions with time delays. Different numerical examples are provided, including a general multivariable time delay process. Integer order IMC controllers are designed in each case, as well as fractional-order IMC controllers. The simulation results show that the proposed fractional-order IMC controller ensures an increased robustness to modelling uncertainties. Experimental results are also provided, for the design of a multivariable fractional-order IMC controller in a Smith predictor structure for a quadruple-tank system.