We present an agglomerative in-silico study that explores the feasibility of leveraging nanosecond pulsed electric field on cervical cell samples toward improving cervical cell neoplasia and cancer detection by enhancing the visual fidelity of the screening tests. This comprehensive, multimodal study explores the effect of these pulsed electric fields on membrane electroporation and deformation in cell clusters extracted from real cervical sample images. Apart from performing a quantitative assessment of dye uptake change under the predefined electric stress, we also perform a thermal analysis to check whether the thermal envelope is breached during the application of a pulsed electric field. For the sake of completeness, we also present the hardware details of a Marx generator-based pulsar that provides the necessary stimulus for various multiphysics studies.
Devices based on electrical stimulation are in common use for a variety of therapeutic bio-applications, including, but not limited to, neuro-prosthetics, pain management or even in situ local anesthetic modalities. Many require the use of multielectrode systems to selectively activate a group of nerves. In this context, a modeling study is carried out to probe some of the details of nerve activation resulting from multielectrode excitation. In particular, aspects such as threshold stimulus currents, their variation with the number of electrodes used, dependence on nerve radii, and the possibility of blocking an action potential (AP) have been quantitatively analyzed. The injection currents needed to initiate an AP are shown to decrease as the number of stimulating electrodes increases. It is also demonstrated that blocking AP propagation in a nerve segment could be achieved more efficiently at lower magnitudes of the interruption signal if more electrodes in an electrical excitation array were to be used. This result is important and would have practical relevance since the lower intensity external signals indicate a safer and more reliable approach to both AP launches and possible AP blockages.
Generative modeling of physiological signals, particularly 12-lead electrocardiograms (ECG), faces significant challenges in preserving local morphological fidelity while adhering to strict global temporal constraints, such as R-peak positioning. Traditional conditional Variational Autoencoders (cVAEs) often suffer from posterior collapse or temporal misalignment when conditioning information is restricted to the network bottleneck. In this paper, we propose a Spatially-Conditioned VAE (SC-VAE) that injects temporal constraints (with Gaussian smoothing to reduce conditioning shock) at multiple resolutions within the decoder architecture. Furthermore, we introduce a Latent Manifold Harvesting strategy that iteratively curates a “gold-standard” latent bank during training, enabling robust post-hoc sampling via jittered replay. Our methodology significantly outperforms Diffusion Probabilistic Models in inference speed while resolving the topological drift commonly observed in standard VAEs.