KEY POINTS: Twitch transdiaphragmatic pressure elicited by cervical magnetic stimulation of the phrenic nerves is a fully non-volitional method for assessing diaphragm contractility in humans, yet it requires invasive procedures such as oesophageal and gastric catheter balloons. Ultrafast ultrasound enables a very high frame rate allowing the capture of transient events, such as muscle contraction elicited by nerve stimulation (twitch). Whether indices derived from ultrafast ultrasound can be used as an alternative to the invasive measurement of twitch transdiaphragmatic pressure is unknown. Our findings demonstrate that maximal diaphragm tissue velocity assessed using ultrafast ultrasound following cervical magnetic stimulation is reliable, sensitive to change in cervical magnetic stimulation intensity, and correlates to twitch transdiaphragmatic pressure. This approach provides a novel fully non-invasive and non-volitional tool for the assessment of diaphragm contractility in humans. ABSTRACT: Measuring twitch transdiaphragmatic pressure (Pdi,tw ) elicited by cervical magnetic stimulation (CMS) is considered as a reference method for the standardized evaluation of diaphragm function. Yet, the measurement of Pdi requires invasive oesophageal and gastric catheter-balloons. Ultrafast ultrasound is a non-invasive imaging technique enabling frame rates high enough to capture transient events such as evoked muscle contractions. This study investigated relationships between indices derived from ultrafast ultrasound and Pdi,tw , and how these indices might be used to estimate Pdi,tw . CMS was performed in 13 healthy volunteers from 30% to 100% of maximal stimulator intensity in units of 10% in a randomized order. Pdi,tw was measured and the right hemidiaphragm was imaged using a custom ultrafast ultrasound sequence with 1 kHz framerate. Maximal diaphragm axial velocity (Vdi ,max ) and diaphragm thickening fraction (TFdi,tw ) were computed. Intra-session reliability was assessed. Repeated-measures correlation (R) and Spearman correlation coefficients (rho) were used to assess relationships between variables. Intra-session reliability was strong for Pdi,tw and Vdi,max and moderate for TFdi,tw . Vdi,max correlated with Pdi,tw in all subjects (0.64 < rho < 1.00, R = 0.75; all P < 0.05). TFdi,tw correlated with Pdi,tw in eight subjects only (0.85 < rho < 0.93, R = 0.69; all P < 0.05). Coupling ultrafast ultrasound and CMS shows promise for the non-invasive and fully non-volitional assessment of diaphragm contractility. This approach opens up the prospect of both diagnosis and follow-up of diaphragm contractility in clinical populations.