e.g., does 20 minutes before a heart attack have a reliable signature? No. It depends on what causes the heart attack, but it's generally either a clot that d

e.g., does 20 minutes before a heart attack have a reliable signature? No. It depends on what causes the heart attack, but it's generally either a clot that dislodges from somewhere else and lodges in a coronary artery, or a new clot that forms suddenly. Neither of those events could be predicted (other than the general prediction of risk factors). For that matter, it also depends on what event you consider the 'heart attack'. Is it the occlusion of blood flow, or the resulting ischemia? There will be a time lag between those as well. Detecting a heart attack shortly _after_ it occurred would also be useful, and likely much easier to do (detect cardiac enzymes, etc).