Automated diagnosis based at least in part on pulse waveforms

A diagnostic system, comprising:

• at least one processor circuit;
• at least one nontransitory memory communicatively coupled the at least one processor circuit and which stores at least one of processor executable instructions or data, execution of which causes the at least processor circuit to:
• for each of a number of visits by each of a plurality of subjects, receive respective pulse signal information representative of:
• i) a first pulse signal waveform captured from the respective subject at a first applied pressure at a first location, ii) at least a second pulse signal waveform captured from the respective subject a second applied pressure at the first location, the second applied pressure different than the first applied pressure, and iii) at least one value indicative of at least one of the first or the second applied pressures, and receive diagnosis information that includes at least a primary diagnosis associated with the respective visit by the respective subject;
• for each of the number of visits by each of the plurality of subjects, receive timing information which represents an event in a cardiac cycle that is sensed separately from the plurality of pulse waveforms;
• store the received respective pulse signal information, timing information, and diagnosis information to the at least one nontransitory memory;
• and
• perform machine learning on the pulse waveform, the timing information, and diagnosis information during a training time to determine one or more correlations between various ones of a number of defining characteristics of representations of the first and at least the second pulse signal waveforms and the diagnosis information.