Man vs Machine
Machines were always designed to help accomplish what humans could not physiologically do by themselves. In the medical world, image technology enables physicians to "see" within the body without cutting it open. The arms of a surgical robot extend through tiny incisions to work tirelessly with impeccable precision and greater degrees of movement whereas a surgeon's hands are bulky and prone to fatigue. With Rosales' invention, titled "Guiding Differential Diagnosis through Information Maximization", will we begin to see a future where machines, rather than humans, begin taking the role as a patient's primary medical provider?
Over the past 15 years, medicine has evolved quite rapidly along with the technology supporting it. One of the greatest stigmas attached to a physician is his/her notoriously bad handwriting. According to a study done by the National Institutes of Health, 38% of handwritten, paper-based prescriptions contained errata. After the adoption of e-prescribing, a system that allows providers to write and send prescriptions electronically to a pharmacy, prescription errors dropped dramatically to 6.6%. During this time, surgery also transformed with leaps and bounds after the FDA approval of the Da Vinci Surgical Robot in 2000. As the surgeon sits at the control console, the Da Vinci Robot provides surgeons with a three dimensional "landscape" inside the patient's body while eliminating hand tremors and fatigue. Surgeries are becoming less invasive as medical equipment continues to advance - reducing recovery time, complications, and scarring.
As diseases become more complex in presentation and more difficult to treat/manage, an increased burden on the healthcare system occurs when excessive "exploratory" tests are performed to obtain a diagnosis. Through innovation, Rosales strives to improve patient care while seeking to minimize the financial stresses imposed on the system and its patients.
Rosales proposes a networked database system that will 1) analyze/compare patient data (i.e. vitals, blood work, etc) with a database of patient information, 2) provide the physician with several "probable" diseases based on such comparisons, 3) offer a set of tests that will improve the "confidence" of such diagnosis, and lastly 4) offer a selection of treatments while notifying the physician on the "risk, cost, effectiveness, discomfort, etc" of such treatments. Once the patient seeks treatment, the updated results/outcomes would then be entered into the database to further enhance the scope of the system.
The advantages to Rosales' invention are quite clear. Not only will this diagnostic system impact the practice of medicine and the cost of healthcare, but this system could also affect health care on a global scale. By accessing a centralized medical database, physicians could provide "tailored" treatments to patients based on their unique set of symptoms. Furthermore, this diagnostic system could offer accurate diagnoses and location-specific treatment options in remote areas where the number of available physicians are extremely limited. Shortages of physicians are even anticipated right outside our doorstep. According to a statistic by the American Medical Association, there is an estimated shortage of 40,000 primary care physicians by 2020. Rosales' system could allow other health professionals, such as nurse practitioners and physician's assistants, to make diagnoses without a physician present at each clinic.
There are also daunting questions that must be addressed when considering the global implementation of Rosales' invention. When considering any mechanical device, we must discuss the rate of error of the device and its implications. If the system suggests an incomplete diagnosis or an ineffective plan of treatment, which leads to the death of a patient, who would be liable? Additionally, numerous studies value human touch and interaction. While a machine may provide an excellent level of medical care, many patients may question whether the level of personal, "quality" care is the same when working with a machine instead of another human being.
Technology similar to Rosales' system currently exists, but is not approved by the FDA for use in real-world clinical settings. IBM is seeking to implement its Watson, a supercomputer designed to "understand natural language, generate hypotheses based on evidence, and learn/adapt", into oncological settings. Watson is able to sift through thousands of medical literature to aid the physician to an accurate diagnosis and effective treatment. Below is a demonstration of Watson's step by step approach in diagnosing a cancer patient and offering her various treatment options.
Science fiction has always hinted at a future where robotic machines "practice" medicine completely autonomously. These machines would make the "best" decisions based on probability and implement treatment according to preset policies/procedures. However, it is important to remember that physicians and patients do not always act "logically". A physician's "gut feeling" and a patient's "personal preference" are two fundamentally "human" aspects that machines may never comprehend. While the cold, metallic touch of machines will never replace the warm touch of another human being, we believe that technology will continue to become more integrated to assist humans in diagnosing and treating patients.