Organ-on-chip systems are promising new in vitro research tools in medical, pharmaceutical, and biological research. Their main benefit, compared to standard cell culture platforms, lies in the improved in vivo resemblance of the cell culture environment. A critical aspect of these systems is the ability to monitor both the cell culture conditions and biological responses of the cultured cells, such as proliferation and differentiation rates, release of signaling molecules, and metabolic activity. Today, this is mostly done using microscopy techniques and off-chip analytical techniques and assays. Integrating in situ analysis methods on-chip enables improved time resolution, continuous measurements, and a faster read-out; hence, more information can be obtained from the developed organ and disease models. Integrated electrical, electrochemical, and optical sensors have been developed and used for chemical analysis in lab-on-a-chip systems for many years, and recently some of these sensing principles have started to find use in organ-on-chip systems as well. This perspective review describes the basic sensing principles, sensor fabrication, and sensor integration in organ-on-chip systems. The review also presents the current state of the art of integrated sensors and discusses future potential. We bring a technological perspective, with the aim of introducing in-line sensing and its promise to advance organ-on-chip systems and the challenges that lie in the integration to researchers without expertise in sensor technology.

Microfluidic devices are widely used for biomedical applications but there is still a lack of affordable, reliable and user-friendly systems for transferring microfluidic chips from an incubator to a microscope while maintaining physiological conditions when performing microscopy. The presented carrier represents a cost-effective option for sustaining environmental conditions of microfluidic chips in combination with minimizing the device manipulation required for reagent injection, media exchange or sample collection. The carrier, which has the outer dimension of a standard well plate size, contains an integrated perfusion system that can recirculate the media using piezo pumps, operated in either continuous or intermittent modes (50-1000 ml/min). Furthermore, a film resistive heater made from 37 mm-thick copper wires, including temperature feedback control, was used to maintain the microfluidic chip temperature at 37 °C when outside the incubator. The heater characterisation showed a uniform temperature distribution along the chip channel for perfusion flow rates up to 10 ml/min. To demonstrate the feasibility of our platform for long term cell culture monitoring, mouse brain endothelial cells (bEnd.3) were repeatedly mon-

IntroductionMicrofluidic devices offer a precise and rapid assay platform with fully controlled microenvironment conditions with regards to pH and chemical composition. Some assays require particle manipulation for sorting, incubation or perfusion and here acoustophoresis offers a non-contact and label-free method to achieve this. Acoustic trapping is a method to retain microparticles in a microfluidic channel against a flow and its use has already been demonstrated for sample enrichment [1], and multistep bead-based assays [2]. One challenge when using acoustic trapping for biological samples is the heating effects from the piezoelectric transducers (PZT) used to actuate the system. The common method to control and measure the temperature is by measuring the external temperature of the PZT in the assumption the internal temperatures of the chip will be lower than that of the heat source [3]. Here, we present an acoustic trap with integrated temperature sensors, measuring the in situ temperature of the trap comparing it to five external measuring points. With the system we were able to measure temperature variations in the range of ±0.01 °C. Our goal was to test the hypothesis, whether the PZT would always be the hottest position during operation. Our results confirmed this assumption for the system without external cooling, but refuted it with applied external convective cooling. This demonstrates the need for integrated temperature sensors to monitor all microenvironmental aspects in an acoustic trap.