I long ago designed exercise bike controllers.
Figures below are from memory and 'less than exact' but should give ballpark ideas.
Use a (conceptual) bike!
Rather than trying to harness blood flow and heart-pump which are not optimised for energy output, use a portion of the body which is - ie leg muscles with suitable electrical control signals for leg activation driving an alternator or equivalent. The "user" could be in any suitable position - on their back with legs bent to drive pedal mechanism would work well enough (bicycle position rotated backwards about 120 degrees. Or a "recumbent" posture. Many ideas here.
Various ideas of how maximal energy can be extracted over longer periods.
Note attitude of rear rider in tandem configuration.
Start with an energy input of about 2000 kilo calories per day for a semi sedentary person.
1 cal ~= 4.2 watt-second = 4.2 Joule.
2000 kCal = 2000 x 1000 x 4.2 J ~= 8.4 MJ/day.
1 kWh = 1000 W x 3600 seconds = 3.6 MJ.
So TOTAL food energy for average person ~~~= 2.333 kWh .
At (unobtainable) 100% efficiency that's about $0.60 of electricity at 25c/unit.
If you can extract 10% of total energy from user that's ~=1/4 kWh or 6c of electricity equivalent./day. More food = more energy and probably a greater % out but still small.
I could pedal an exercycle for say 15 hours/day at 50 Watts if my life depended on it. 50 Watts is a very easy pedalling rate against a light but noticeable load. After an hour os so I'd find it annoying (some wouldn't). After 15 hours I'd be exhausted but it should be maintainable given enough food and fitness.
That's 50W x 15 hours x 3600 sec/hour = 2.7 MJ.
My food requirement would rise - I'd expect to 3000-4000 kCal/day or say 12-16 Mj/day range.
I could maybe do 100 Watts for an hour and be "rather tired".
Once I could do 500 Watts for about 10 seconds and then needed a good lie down :-).
To fly the English Channel under your own power takes 300-500 Watts for as long as it takes.
That makes the mechanical output about say 15-20% efficient.
On the level on a bicycle I'd expect 50 Watts continuous to give me say 4 to 5 kilometres per hour so that would take me about 60 to 75 km in a day :-)
Any hills would greatly reduce that.
Leg muscles, heart etc could probably be run semi indefinitely (given enough food input and suitable plumbing) at say 100 Watts, so about triple my 100 Watts x 15 hours results.
Despite what people say here, I've long been impressed at how much work you can get from a human body on a kg or so of semi random carbohydrates. If you used solar heated ponds to grow "** Cyano-bacteria aka blue- green algae aka 'pond scum"** you may [tm] be able to run the energy production system on the output. Each person-unit is only running at 100 Watts say but 24 hours / day so that's equivalent to eg a solar panel's 4hrs mean sun/day year round in most areas. So a 100W person gives the same daily energy as a 600 Watt PV panel which has an area of about 4 square meters at current module delivered efficiencies of say 15%. A panel does not need algal fuel - but a person unit can probably fit in one square meter and can be stacked N high, unlike panels which must "see" the sun directly. Persons have to be plumbed (algal matter and water in, waste products out). The algal ponds take extra area - probably in excess of the PV panel area needs.
Overall it may depend on available resources - but PV panels sound like an easier task :-).