Others have demonstrated optical sensors for blood pressure and blood alcohol content. 24/7 monitoring of HR, Resp., glucose, BP, & alcohol will be interesting.
Yes, you can build a wearable device that mostly gets your blood pressure right, most of the time. What are you supposed to do with that? It's not a useful product (and the FDA won't let you sell it, for that reason).
If you can't depend on the results, you can't quantify in exactly what way the results are erroneous, and you don't have an enormous volume of clinical data and outcome studies, etc. with your device that allows you to draw clinical conclusions about your results regardless of their errors, you have nothing actually useful.