Beyond the Rectangle

This is the authors' abstract. We don't add key points for this paper.

Objective. Deep brain stimulators, cardiac pacemakers, cochlear implants and transcutaneous nerve stimulators all deliver the same rectangular industrial pulse across target tissues that differ by several orders of magnitude in electrical conductivity. This shape was inherited from 1970s electronics rather than being chosen on physiological grounds. This paper establishes a quantitative electrical benchmark for comparing candidate waveforms on spectral, energetic and hardware axes, which is a prerequisite to any fibre-level analysis. Approach. Eight waveforms, including the industrial rectangle and its clinical asymmetric variant, as well as a smooth curve, are driven through equivalent circuit models of eight target tissues, which cover the clinical conductivity range. Each tissue is represented by a linear passive network fitted to its measured dielectric spectrum, and every waveform-tissue pair is solved in the time domain under voltage- and current-driven modes. Main results. The smooth curve outperforms the industrial rectangle by two to three orders of magnitude on spectral leakage, and requires approximately ninety times less peak current per unit of delivered charge. On peak tissue voltage, tissue impedance dominates the ordering. The asymmetric rectangle sits at the opposite end of the ordering. Spectral rankings are invariant to drive mode, while peak tissue voltage varies in magnitude across tissues at a fixed clinical dose. Significance. Smooth-waveform alternatives to the industrial rectangle could reduce the peak current, thermal and battery demands of neural implants by up to two orders of magnitude, while maintaining the delivered charge. This would open up a design margin that is currently unavailable to clinicians and device makers. For patients receiving a chronic implant, this translates directly into longer intervals between battery-replacement surgeries at equal clinical dose.