Monitoring in vivo mitochondrial oxygen tension (mitoPO₂) enables the measurement of mitochondrial oxygen consumption (mitoVO₂), providing deeper insights into the skin’s mitochondrial environment. However, current mitoVO₂ analysis often relies on manual identification of start and end points, which introduces substantial inter-user variability. Addressing this limitation is crucial for broader adoption, comparability, and reproducibility across research groups. Therefore, the aim of this study was to develop a neural network–based software that automatically analyzes mitoVO₂. A Bi-directional Long Short-Term Memory neural network was trained on 125 mitoPO₂ measurement sequences and optimized through Bayesian optimization. It identifies start points and measurement periods, then applies a modified Michaelis-Menten fit to calculate mitoVO₂. This framework, embedded in automated software, was validated against the consensus of 3 raters. Bayesian optimization yielded an overall network performance of 94.2% on the test set. The neural network identified 91% of mitoVO₂ start points within a ± 5-sample range of the manual consensus. Mean mitoVO₂ values for the consensus and software were 6.56 and 6.63 mmHg s^{− 1}, respectively, corresponding to a bias of -0.057 mmHg s^{− 1}. Multiple runs of the network on the same dataset produced identical results, confirming consistency and eliminating inter-user variability. The developed neural network–based software automatically and consistently analyzes mitoVO₂ measurements, substantially reducing reliance on subjective judgments. By enabling a standardized approach to mitoVO₂ analysis, this tool improves data comparability and reproducibility across research settings. Future work will focus on further refining precision and extending functionality through multi-center collaborations.

Objective. Electrical impedance tomography (EIT) produces clinical useful visualization of the distribution of ventilation inside the lungs. The accuracy of EIT-derived parameters can be compromised by the cardiovascular signal. Removal of these artefacts is challenging due to spectral overlapping of the ventilatory and cardiovascular signal components and their time-varying frequencies. We designed and evaluated advanced filtering techniques and hypothesized that these would outperform traditional low-pass filters. Approach. Three filter techniques were developed and compared against traditional low-pass filtering: multiple digital notch filtering (MDN), empirical mode decomposition (EMD) and the maximal overlap discrete wavelet transform (MODWT). The performance of the filtering techniques was evaluated (1) in the time domain (2) in the frequency domain (3) by visual inspection. We evaluated the performance using simulated contaminated EIT data and data from 15 adult and neonatal intensive care unit patients. Main result. Each filter technique exhibited varying degrees of effectiveness and limitations. Quality measures in the time domain showed the best performance for MDN filtering. The signal to noise ratio was best for DLP, but at the cost of a high relative and removal error. MDN outbalanced the performance resulting in a good SNR with a low relative and removal error. MDN, EMD and MODWT performed similar in the frequency domain and were successful in removing the high frequency components of the data. Significance. Advanced filtering techniques have benefits compared to traditional filters but are not always better. MDN filtering outperformed EMD and MODWT regarding quality measures in the time domain. This study emphasizes the need for careful consideration when choosing a filtering approach, depending on the dataset and the clinical/research question.