Some technologies dramatically enhance the accuracy and speed of clinicians' efforts. Identifying a bacterial or viral infection, and the best drugs to treat it, can mean long waits for blood cultures. But scientists have developed biochips that can do a complete microbial scan in a couple of hours, without culturing -- and in the process may identify mutations that make some microbes antibiotic resistant.

Thanks to artificial intelligence and machine learning, diagnostic tools can be trained to read tissue samples and radiologic scans. Google researchers fed more than a quarter-million patients' retinal scans into algorithms that recognize patterns -- and the technology "learned" to spot which patterns predict a patient has high blood pressure or is at increased risk for heart attack or stroke. In some comparisons, digital tools produced more accurate analyses than did human pathologists, dermatologists, or radiologists.