A new report shows that doctors and hospitals are increasingly utilizing robots and computer programs to diagnose diseases and prescribe treatments. Is this a good or a bad thing? Do you want to be diagnosed and treated by ‘Doctor Robot’ the next time you go to the clinic?
While many of us might wonder about the logic of such a strategy, the conventional medical industry is applauding the increasing use of computers to diagnose and treat disease.
Responding to this, BM has introduced a program that utilizes the Watson robotic artificial intelligence system to diagnose cancers and other diseases. The program is called Watson Health.
Research supports doctor robot
The Watson Health project was boosted by a 2013 study of 500 individual patients, randomized from 6,700 clinical depression patients. The researchers, from Indiana University, tested computer diagnosis and treatment for related conditions against the diagnoses and treatments of medical doctors in each case. The research found the computer’s diagnoses and prescriptions were almost 42 percent more effective than the diagnoses and treatments of the doctors.
Not surprisingly, they also found that the computer, set up with Markov Decision Process and Dynamic Decision Networks, was more cost-effective than the doctors. The computer cost averaged $189 per outcome unit compared to nearly $500 for every unit outcome for the doctors.
Lead researcher Dr. Casey Bennett discussed the advantages of the program:
“We’re using modern computational approaches to learn from clinical data and develop complex plans through the simulation of numerous, alternative sequential decision paths. The framework here easily out-performs the current treatment-as-usual, case-rate/fee-for-service models of health care.”
Computerized cancer diagnosis and treatment
A team of doctors at Memorial Sloan Kettering Cancer Center, lead by Dr. Mark Kris, is now helping train the computer to analyze diagnostic information and design treatment protocols.
The program is being implemented at varying degrees in hospitals and clinics. In a recent interview with PBS Newshour, Dr. Kris explained that the computer system is set up to provide all the diagnostic information and possible treatment protocols that a doctor would normally make.
In other words, the computer will name the disease. Then it will prescribe those medications that have been programmed into the computer for that particular disease. Dr. Kris explains:
“So, the first thing this system does is, it shows all the different treatments that are recommended. And then now I ask what kind of chemo to give, and it points to a chemo regimen, two different drugs.”
What’s the problem with naming and prescribing?
Each person is not only different in terms of the progression and development of any condition: Each person also metabolizes medications, as well as supplements, differently.
This goes to the core root of one of the major issues that natural medicine experts have with conventional medicine: Conventional medicine wants to simplify the process of diagnosis and treatment. By design conventional medicine seeks to categorize everyone with certain types of symptoms or diagnostic charts. Then it seeks to give every person in that category basically the same drug or treatment plan.
But as most private practice doctors and practically every natural medical professional knows, the same drug or treatment does not work for everyone. And a standard dose of any prescriptive recommendation can be dangerous to one person while under-productive for another person.
This was illustrated in a study I reported on not so long ago from Tufts University School of Medicine. The study showed that women metabolize drugs differently than men. This is after decades of indiscriminate pharmaceutical dosing between men and women.
This is one of the issues that traditional healers have with naming diseases as well. People not only have different metabolisms: We also have different medical histories. We have different sensitivities. We have different lifestyles. We have different levels of awareness. We have different personalities.
The ten minute consultation
All of these and other differences are typically not addressed by conventional medicine. Conventional medicine is set up in such a way that doctors are only allowed 10-15 minutes with each patient. This allows for a quick diagnosis – a condition name – and a standard drug prescription and that’s about it.
And the drug prescription is typically derived from the Physician’s Desk Reference and similar references, which overwhelmingly represent pharmaceutical company drugs.
No wonder computers outperformed doctors in the study above, given this state of 10-minute conventional medicine.
Systematized medicine for machines
So what conventional medicine has been trying to systematize medicine. This is effectively attempting to turn people into machines.
Some of the pitfalls to such a progression was addressed in the PBS report by Dr. Robert Wachter from the University of California, San Francisco. Dr. Wachter responded:
“A lot of medicine kind of lives in that middle ground, where it’s really messy. And someone comes in to see me and they have a set of complaints and physical exam findings all that. And it could be — if you look it up in a computer, it could be some weird — it could be the Bubonic plague, but it probably is the flu.”
Taking the human element out of the process, or even relying too much upon the computer to diagnose and treat can also be dangerous. Dr. Wachter discussed some examples of how relying on computers can cause harm:
“So, in two different cases, the computers threw up alerts on the computer screen that said, this is an overdose. But the alert for a 39-fold overdose and the alert for a 1 percent overdose looked exactly the same. And the doctors clicked out of it. The pharmacists clicked out of it. Why? Because they get thousands of alerts a day, and they have learned to just pay no attention to the alerts.”
Personalized medicine for persons
Dr. Wachter’s point about much of medicine being “really messy” is reflecting upon the key element of the human condition: We are not machines. There is someone inside this body machine and this person – this personality – requires personal attention if we want to encourage the body’s natural healing response.
Yes, machines can be diagnosed quite well with computers. When we take our car to the repair shop and they plug it into the computer, the diagnosis is often right: Not always, but often.
But we are not machines. Yes, the body is like an extremely complex machine, but within this machine is a person. Just as a person drives an automobile, there is a person within each body who drives it: A person composed not of matter, but of spirit. And the condition of the spirit interacts with and often influences the health of the body and mind (e.g. “body-mind-spirit”).
This has been confirmed by research increasingly finding that diseased conditions of the body are often associated with conditions of the spirit. This is why matters of love, relationships and stress are often at the root of many ill conditions within the body.
Certainly there is a place for computers in medicine. But not to replace the healer.
Bennett CC, Hauser K. Artificial intelligence framework for simulating clinical decision-making: a Markov decision process approach. Artif Intell Med. 2013 Jan;57(1):9-19. doi: 10.1016/j.artmed.2012.12.003.
Sledge GW, Hudis CA, Swain SM, Yu PM, Mann JT, Hauser RS, Lichter AS. ASCO’s approach to a learning health care system in oncology. J Oncol Pract. 2013 May;9(3):145-8. doi: 10.1200/JOP.2013.000957.
Malin JL. Envisioning Watson as a rapid-learning system for oncology. J Oncol Pract. 2013 May;9(3):155-7. doi: 10.1200/JOP.2013.001021.
Indiana University Release: Can computers save health care? IU research shows lower costs, better outcomes. Feb. 11, 2013