Dr. Big Data: “How are you feeling today?”
“Hello Doctor, I haven’t been feeling well for the past days. I think I am sick.”
We have all said this sentence in one form or another far too often. If we don’t feel well and this feeling doesn’t pass or gets worse we often seek a doctor for medical help. But lets see about how Big Data and more generally speaking Technology will change the way we seek medical help and get or stay healthy.
How does a doctor diagnose a disease?
When you go see a doctor in most of the cases your answers to several questions as well as your current temperature and/or some other very basic tests will lead to a diagnosis. This ability to diagnose a disease comes from years of hard work at university and a lot of experience with patients. Your answer to a question will determine which question the doctor will ask next or which test will be done on you and so on. After a certain number of steps the possibility of diseases has decreased to a small enough number to make a fair diagnosis based on the probability of a diagnosis. If you are feeling nauseous and have a fever it is quiet likely that you have a basic flue and not a super rare disease that shows similar symptoms. So in the end you are eliminating possible diseases based on data and using probability to determine the most likely diagnosis. A doctor usually works reactionary and not preventive.
What is Big Data?
Big Data is a very large accumulation of digital data. By itself it might be worthless, but with smart data analytics it can lead to very interesting insights. In the past year data creation has increased enormously. The amount of data that we nowadays are creating in a matter of days we created in a year two decades ago. By using those analytical tools we can break down gigantic bunches of information to their core meanings and try to hear beyond the noise to gather information [2]. This allows us to see trends and patterns that could not be see before within fields such as weather prediction, banking or insurance and a lot more.
So what does this mean for us?
I am by no means trying to talk down the great amount of work that goes into becoming and being a doctor. I am simply laying out the facts that medical diagnosis is a rather repetitive and purely analytical task that is based on experience and previous case-matching. A task that in many cases could be done by a somewhat smart computer with a lot of data. A doctor might have a couple of thousand cases in a career that can be used as a reference for new patients and a doctor will get a formal education once and go on training courses every now and than to keep up to date and also read some scientific paper. But a digital programme that uses big data could read millions and millions of patient records and get the latest research on a rolling basis and consider those for a diagnosis. So the potential is gigantic and partly realised in machines such as IBM’s Watson[8]. If you now consider, that IoT devices on your body could feed the computer with even more data it gets even more exiting. Your digital doctor will not only support you reactionary but will try to prevent you from getting sick by reading any changes in your health very soon. I personally believe that it will take a long time until doctors will be replaced, since the human contact and the soft skills are a very important subconscious dimension to healing. But computers will hopefully play a assisting role in diagnosis to increase humans life spans and quality of life in the future.
Sources
- https://www.forbes.com/sites/peterlipson/2017/01/16/will-big-data-replace-your-doctor/#29df23572f72
- http://www.mckinsey.com/business-functions/digital-mckinsey/our-insights/big-data-the-next-frontier-for-innovation
- https://quartsoft.com/blog/201307/technology-improving-healthcare-industry
- http://www.mckinsey.com/industries/pharmaceuticals-and-medical-products/our-insights/the-role-of-big-data-in-medicine
- https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5331970/
- http://catalyst.nejm.org/big-data-machine-learning-clinical-medicine/
- http://www.nature.com/news/the-power-of-big-data-must-be-harnessed-for-medical-progress-1.21026
- https://www.ibm.com/watson/
6 comments on “Dr. Big Data: “How are you feeling today?””
Comments are closed.
Prior to my Undergraduate studies I lived in China where I worked at a consultancy and learned more about Chinese Culture and Language. In Berlin I worked at a Startup in analytics, Customer Relationship Management and Marketing.
It is one of my goals to become fluent in Chinese with the aim to strengthen the economic partnership between China and Europe, especially within technology.
Great article, thank you for your perspective on this issue!
The rapid growth of AI makes it inevitable for new technologies to take over customary functions. Especially after Julia Hu’s presentation on Lark Technologies, I found your article more relevant and accurate. I believe that the applications of AI in healthcare will have a great impact in the future. You pointed out that algorithms trained on large datasets will be able to make more precise diagnosis since they have more experience and cases to compare to. Recent research from companies like Google also prove that deep learning will be used in groundbreaking applications for diagnosis and treatment.
Interesting article on the topic:
https://www.forbes.com/sites/bernardmarr/2017/05/16/how-ai-and-deep-learning-is-now-used-to-diagnose-cancer/#5ee3c843c783
You can read more about Google’s research on:
https://research.googleblog.com/2017/03/assisting-pathologists-in-detecting.html
I like your post a lot and I agree with many statements that you made. I agree we shouldn’t underestimate the value of the doctors and medical education, but on the other hand the quality in medical services is very difficult to maintain and control, and the decisions made by doctors are so crucially important. Mistakes or incorrect diagnosing can lead to very bad outcome. Doctors are people, and people do mistakes. I have two kids and long history if interacting with doctors, I personally experienced few cases when you get completely different diagnosis from two different doctors, based on the same symptoms, also mistakes in the medication prescriptions, the dosage etc, many frightening cases that made me worrying about my kids. Personally, I am looking forward to the new opportunities that big data will bring to the health sector and medical services.
Thank you, Dean. Great blog subject selection!
I was intrigued by the point Julia Hu stressed:
A: Adding a band-aid or extending the capabilities of medical staff. VS.
B: Replacing medical staff with AI.
I agree that they should be treated and developed separately but think that advancement in both spheres would significantly benefit humanity.
I am a patient at one of a concierge medicine service office. Supposedly the office has access to the latest research and the best technology. I have been with them for a few years and am happy. Something interesting happened last 2 times I went for my annual check in: My doctor suggested some things to help me with my hands/wrists issues, then turned around, and in front of me searched a database to see if there are other things she should recommend. She relied on her experience but then sourced more info from a database. I found that odd but ,after thinking about it more, was glad that doctors are now openly starting to take advantage of advanced technology and data.
Thank you for a very interesting post! It is definitely true that AI will have more applications in the healthcare sector in the future. When comes to medical diagnosis, there already exists a couple of apps which perform medical diagnosis based on the symptoms described by the patient, such as iTriage [1], Isabel Symptom Checker [2] and WebMD [3]. However, these are mainly targeted for personal use and are not very reliable. It will be interesting to see whether such apps will also be developed for professional use in the future!
[1] https://www.itriagehealth.com
[2] https://play.google.com/store/apps/details?id=com.isabelhealthcare.isabelSC
[3] http://symptoms.webmd.com/#introView
This is an interesting field thinking how big data will change our lives. From my perspective, the hard point is not how to determine what kind of disease the patient has, but how should we collect health data of people in their daily lives, and prevent people from having disease before they feel sick. And at that time, the number of hospitals can decrease substantially, and most work of medical care can be done at the everyone’s home.
Great post! I have a number of Doctors in my family, and the feedback I received from them is that Big Data will help in a few ways. It’s a resource to help make a diagnosis and treatment, and should be looked upon as that. What big data, AI, and ML won’t have is a) the experience, or “gut feeling ” about a certain case, and b) the bedside manner to help assist with the diagnosis. Case in point, when my father was diagnosed with stage four lung cancer the scans and tests could tell us this, and the treatment was narrowed down to give him the best chance at maximizing his time left with us. What Big Data didn’t give us was the hope that the Doctor at the Mayo Clinic did, or when my daughter was born six weeks early the confidence that the OB gave us that she would be ok. The human element is so important in medicine and treatment. The best Doctors will utilize big data to find cures for cancer and other rare diseases, and will be able to tell you that they will do everything they can to help. We can’t forget this aspect of medicine.