The use of different cardiac imaging modalities such as MRI, CT or ultrasound enables the visualization and interpretation of altered morphological structures and function of the heart. In recent years, there has been an increasing interest in AI and deep learning that take into account spatial and temporal information in medical image analysis. In particular, deep learning tools using temporal information in image processing have not yet found their way into daily clinical practice, despite its presumed high diagnostic and prognostic value. This review aims to synthesize the most relevant deep learning methods and discuss their clinical usability in dynamic cardiac imaging using for example the complete spatiotemporal image information of the heart cycle. Selected articles were categorized according to the following indicators: clinical applications, quality of datasets, preprocessing and annotation, learning methods and training strategy, and test performance. Clinical usability was evaluated based on these criteria by classifying the selected papers into (i) clinical level, (ii) robust candidate and (iii) proof of concept applications. Interestingly, not a single one of the reviewed papers was classified as a "clinical level" study. Almost 39% of the articles achieved a "robust candidate" and as many as 61% a "proof of concept" status. In summary, deep learning in spatiotemporal cardiac imaging is still strongly research-oriented and its implementation in clinical application still requires considerable efforts. Challenges that need to be addressed are the quality of datasets together with clinical verification and validation of the performance achieved by the used method.

Registration of dynamic CT image sequences is a crucial preprocessing step for clinical evaluation of multiple physiological determinants in the heart such as global and regional myocardial perfusion. In this work, we present a deformable deep learning-based image registration method for quantitative myocardial perfusion CT examinations, which in contrast to previous approaches, takes into account some unique challenges such as low image quality with less accurate anatomical landmarks, dynamic changes of contrast agent concentration in the heart chambers and tissue, and misalignment caused by cardiac stress, respiration, and patient motion. The introduced method uses a recursive cascade network with a ventricle segmentation module, and a novel loss function that accounts for local contrast changes over time. It was trained and validated on a dataset of n = 118 patients with known or suspected coronary artery disease and/or aortic valve insufficiency. Our results demonstrate that the proposed method is capable of registering dynamic cardiac perfusion sequences by reducing local tissue displacements of the left ventricle (LV), whereas contrast changes do not affect the registration and image quality, in particular the absolute CT (HU) values of the entire CT sequence. In addition, the deep learning-based approach presented reveals a short processing time of a few seconds compared to conventional image registration methods, demonstrating its application potential for quantitative CT myocardial perfusion measurements in daily clinical routine.

Background: Traumatic brain injury (TBI) is a leading cause of disabilities resulting from cognitive and neurological deficits, as well as psychological disorders. Only recently, preclinical research on electrical stimulation methods as a potential treatment of TBI sequelae has gained more traction. However, the underlying mechanisms of the anticipated improvements induced by these methods are still not fully understood. It remains unclear in which stage after TBI they are best applied to optimize the therapeutic outcome, preferably with persisting effects. Studies with animal models address these questions and investigate beneficial long- and short-term changes mediated by these novel modalities. Methods: In this review, we present the state-of-the-art in preclinical research on electrical stimulation methods used to treat TBI sequelae. We analyze publications on the most commonly used electrical stimulation methods, namely transcranial magnetic stimulation (TMS), transcranial direct current stimulation (tDCS), deep brain stimulation (DBS) and vagus nerve stimulation (VNS), that aim to treat disabilities caused by TBI. We discuss applied stimulation parameters, such as the amplitude, frequency, and length of stimulation, as well as stimulation time frames, specifically the onset of stimulation, how often stimulation sessions were repeated and the total length of the treatment. These parameters are then analyzed in the context of injury severity, the disability under investigation and the stimulated location, and the resulting therapeutic effects are compared. We provide a comprehensive and critical review and discuss directions for future research. Results and conclusion: We find that the parameters used in studies on each of these stimulation methods vary widely, making it difficult to draw direct comparisons between stimulation protocols and therapeutic outcome. Persisting beneficial effects and adverse consequences of electrical simulation are rarely investigated, leaving many questions about their suitability for clinical applications. Nevertheless, we conclude that the stimulation methods discussed here show promising results that could be further supported by additional research in this field.

Nongenetic optical control of neurons is a powerful technique to study and manipulate the function of the nervous system. This research has benchmarked the performance of organic electrolytic photocapacitor (OEPC) optoelectronic stimulators at the level of single mammalian cells: human embryonic kidney (HEK) cells with heterologously expressed voltage‐gated K⁺ channels and hippocampal primary neurons. OEPCs act as extracellular stimulation electrodes driven by deep red light. The electrophysiological recordings show that millisecond light stimulation of OEPC shifts conductance‐voltage plots of voltage‐gated K⁺ channels by ≈30 mV. Models are described both for understanding the experimental findings at the level of K⁺ channel kinetics in HEK cells, as well as elucidating interpretation of membrane electrophysiology obtained during stimulation with an electrically floating extracellular photoelectrode. A time‐dependent increase in voltage‐gated channel conductivity in response to OEPC stimulation is demonstrated. These findings are then carried on to cultured primary hippocampal neurons. It is found that millisecond time‐scale optical stimuli trigger repetitive action potentials in these neurons. The findings demonstrate that OEPC devices enable the manipulation of neuronal signaling activities with millisecond precision. OEPCs can therefore be integrated into novel in vitro electrophysiology protocols, and the findings can inspire in vivo applications.