Online deployment of an O-PLS model for dielectric spectroscopy-based inline monitoring of viable cell concentrations in Chinese hamster ovary cell perfusion cultivations.

Viable cell concentration (VCC) is an essential parameter that is required to support the efficient cultivation of mammalian cells. Although commonly determined using at-line or off-line analytics, in-line capacitance measurements represent a suitable alternative method for the determination of VCC. In addition, these latter efforts are complimentary with the Food and Drug Administration's initiative for process analytical technologies (PATs). However, current applications for online determination of the VCC often rely on single frequency measurements and corresponding linear regression models. It has been reported that this may be insufficient for application at all stages of a mammalian cell culture processes due to changes in multiple cell parameters over time. Alternatively, dielectric spectroscopy, measuring capacitance at multiple frequencies, in combination with multivariate mathematical models, has proven to be more robust. However, this has only been applied for retrospective data analysis. Here, we present the implementation of an O-PLS model for the online processing of multifrequency capacitance signals and the on-the-fly integration of the models' VCC results into a supervisory control and data acquisition (SCADA) system commonly used for cultivation observation and control. This system was evaluated using a Chinese hamster ovary (CHO) cell perfusion process.

Towards an automated approach for smart sterility test examination.

As new technologies emerge, deep learning applications are often integral parts of new products as features and often as differentiating benefits. This is especially notable in commercial consumer products in everyday applications, such as voice assistants or streaming content recommendation systems. Due to the power and applicability of these deep learning technologies significant efforts are being directed to the development and integration of appropriate models into science and engineering applications to supplant analogue systems that may be highly prone to human error. Here we present an innovative, low-cost approach to advance sterility assessment workflows that are required and regulated within drug release/manufacturing processes. The model system leverages off-the-shelf hardware as well as deep learning models to detect and classify different microbial contaminations in test containers. The paired hardware and software tools were evaluated in experiments using common model organisms (C. sporogenes, P. aeruginosa, S. aureus). With this approach we were able to detect all three test organisms across 40 experiments, furthermore we were capable of classifying the present organisms with an average classification accuracy of over 87%.

Human-Device Interaction in the Life Science Laboratory.

The interaction of the human user with equipment and software is a central aspect of the work in the life science laboratory. The enhancement of the usability and intuition of software and hardware products, as well as holistic interaction solutions are a demand from all stakeholders in the scientific laboratory who desire more efficient workflows. Shorter training periods, parallelization of workflows, improved data integrity, and enhanced safety are only a few advantages innovative intuitive human-device-interfaces can bring. With recent advances in artificial intelligence (AI), the availability of smart devices, as well as unified communication protocols, holistic interaction solutions are on the rise. Future interaction in the laboratory will not be limited to pushing mechanical buttons on equipment. Instead, the interplay between voice, gestures, and innovative hard- and software components will drive interactions in the laboratory into a more streamlined future.

A Machine Vision Approach for Bioreactor Foam Sensing.

Machine vision is a powerful technology that has become increasingly popular and accurate during the last decade due to rapid advances in the field of machine learning. The majority of machine vision applications are currently found in consumer electronics, automotive applications, and quality control, yet the potential for bioprocessing applications is tremendous. For instance, detecting and controlling foam emergence is important for all upstream bioprocesses, but the lack of robust foam sensing often leads to batch failures from foam-outs or overaddition of antifoam agents. Here, we report a new low-cost, flexible, and reliable foam sensor concept for bioreactor applications. The concept applies convolutional neural networks (CNNs), a state-of-the-art machine learning system for image processing. The implemented method shows high accuracy for both binary foam detection (foam/no foam) and fine-grained classification of foam levels.