BACKGROUND: The optimization of growth media is a crucial step in enhancing microbial growth and product formation in industrial bioprocesses. While Design of Experiments (DoE) is widely used for this purpose, it has inherent limitations, particularly in capturing complex interactions between media components and managing the high number of required experiments without overlooking important factors or interactions. This study evaluates batch Bayesian optimization (BBO) as an alternative approach by applying it in simulations and in wet‐lab experiments, using Sporosarcina pasteurii as a model organism, to demonstrate its effectiveness.RESULTS: In silico, BBO significantly outperformed DoE on average (+14%), although it was not superior in all test cases, highlighting the need for further investigation. In automated high‐throughput microbioreactor experiments with S. pasteurii, BBO led to a 28% increase in maximum backscatter value, indicating a higher biomass titer compared to a DoE‐optimized medium. BBO enabled a more targeted optimization of media components, allowing broader exploration when needed and focusing on specific adjustments where appropriate. The BBO‐optimized growth medium, designed in small‐scale experiments, also outperformed the DoE‐optimized medium at a 2 L bioreactor scale.CONCLUSION: In our use cases, BBO offers a more effective and targeted approach for growth media optimization than traditional DoE. Placing our results in the context of other studies, BBO shows promise for outperforming DoE in bioprocess optimization tasks. The ability of BBO to streamline the optimization process could enhance yields and reduce costs in industrial bioprocesses. © 2025 The Author(s). Journal of Chemical Technology and Biotechnology published by John Wiley & Sons Ltd on behalf of Society of Chemical Industry (SCI).

T cells play a pivotal role in cancer research, particularly in immunotherapy, which harnesses the immune system to target malignancies. However, conventional expansion methods face limitations such as high reagent consumption, contamination risks, and difficulties in maintaining suspension cells in dynamic culture environments. This study presents a microfluidic system for long-term culture of non-adherent cells, featuring automated perfusion and image acquisition. The system integrates deep learning-based image analysis, which quantifies cell coverage and estimates cell numbers, and efficiently processes large volumes of data. The performance of this deep learning approach was benchmarked against the widely used Trainable Weka Segmentation (TWS) plugin for Fiji. Additionally, two distinct lab-on-a-chip (LOC) devices were evaluated independently: the commercial ibidi^® LOC and a custom-made PDMS LOC. The setup supported the proliferation of Jurkat cells and primary human T cells without significant loss during medium exchange. Each platform proved suitable for long-term expansion while offering distinct advantages in terms of design, cell seeding and recovery, and reusability. This integrated approach enables extended experiments with minimal manual intervention, stable perfusion, and supports multi-reagent administration, offering a powerful platform for advancing suspension cell research in immunotherapy.

Current platforms for drug screening typically do not account for the tumor microenvironment. Microfluidic shear flow assays provide a highly sensitive tool for studying tumor cell communication at the single cell level with the microenvironment. Adhesion thereby serves as a functional readout that reflects cellular state, including loss of viability. However, previous platforms required extensive manual handling and time‐consuming post‐assay analysis. We developed a bright‐field microscopy‐enabled, semi‐automated shear flow platform that combines hardware operation with a machine‐learning‐based analysis pipeline. The algorithm delivers consistent, high‐quality results within minutes with a precision of 98.3% and a recall of 99.1%, indicative of high tracking specificity and object discrimination.

Staining-based assays are widely used for cell analysis but are invasive, alter physiology, and prevent longitudinal monitoring. Label-free, morphology-based approaches could enable real-time, non-invasive drug testing, yet detection of subtle and dynamic changes has remained difficult. We developed a deep learning framework for stain-free monitoring of leukemia cell cultures using automated bright-field microscopy in a semi-automated culture system (AICE3, LABMaiTE, Augsburg, Germany). YOLOv8 models were trained on images from K562, HL-60, and Kasumi-1 cells, using an NVIDIA DGX A100 GPU for training and tested on GPU and CPU environments for real-time performance. Comparative benchmarking with RT-DETR and interpretability analyses using Eigen-CAM and radiomics (RedTell) was performed. YOLOv8 achieved high accuracy (mAP@0.5 > 98%, precision/sensitivity > 97%), with reproducibility confirmed on an independent dataset from a second laboratory and an AICE3 setup. The model distinguished between morphologically similar leukemia lines and reliably classified untreated versus differentiated K562 cells (hemin-induced erythroid and PMA-induced megakaryocytic; >95% accuracy). Incorporation of decitabine-treated cells demonstrated applicability to drug testing, revealing treatment-specific and intermediate phenotypes. Longitudinal monitoring captured culture- and time-dependent drift, enabling separation of temporal from drug-induced changes. Radiomics highlighted interpretable features such as size, elongation, and texture, but with lower accuracy than the deep learning approach. To our knowledge, this is the first demonstration that deep learning resolves subtle, drug-induced, and time-dependent morphological changes in unstained leukemia cells in real time. This approach provides a robust, accessible framework for label-free longitudinal drug testing and establishes a foundation for future autonomous, feedback-driven platforms in precision oncology. Ultimately, this approach may also contribute to more precise and adaptive clinical decision-making, advancing the field of personalized medicine.