If you haven't noticed, AI is getting big in Healthcare
Editor’s Letter
As next Sunday is Christmas day there will not be a newsletter. The week after will be our annual update on Unmasking Medicine as an organization as well as our plans for 2023.
Could chat GPT replace doctors?
On November 30, San Francisco-based startup OpenAI released AI chatbot ChatGPT. ChatGPT has already radically changed the tech industry with it's ability to answer questions and write essays.
In the short timespan since its release, medical professionals have already begun to utilize ChatGPT's surprising capabilities. One striking example of ChatGPT's impact on healthcare is its ability to assist doctors in diagnosing patients. An Australian doctor was able to use ChatGPT to turn hypothetical symptoms into a diagnosis. After inputting the patient's symptoms and medical history into ChatGPT, the AI program was able to provide a list of potential diagnoses, complete with relevant information and treatment options. Having experimented with the program, he remarked "ChatGPT might just take my job as a doctor."
But the benefits of ChatGPT and other AI programs go beyond simply assisting with diagnoses. These programs are also changing the way that healthcare professionals are educated. Medical students have already begun using ChatGPT to generate practice questions or case studies. The program could also potentially be used to generate explanations or summaries of medical concepts for students to learn from.
In addition to ChatGPT, there are several other AI programs that are already having a significant impact on healthcare. One such program is "DeepMind," which is used to analyze medical records and identify patterns that could lead to more accurate diagnoses and better treatment outcomes.
Another AI program called "Enlitic" is used to analyze medical images, such as x-rays and MRIs, to help doctors identify abnormalities and make more accurate diagnoses.
Overall, the use of ChatGPT and other AI programs in healthcare is leading to more accurate diagnoses, improved treatment outcomes, and more effective education for healthcare professionals. As these technologies continue to advance, it's likely that they will play an even greater role in the future of healthcare. So, it is important to keep an eye on these developments and see how they can be used to improve patient care. To demonstrate how effective ChatGPT can be, this entire article was written by the program, then carefully edited by our Unmasking Medicine staff.
In another AI: cardiac amyloidosis diagnoses
Pfizer is teaming up with Anumana to diagnose cardiac amyloidosis. Cardiac amyloidosis is a disease wherein misfolded proteins called amyloids start depositing in the heart muscle and stiffen the heart over time. This makes it harder for the heart to properly contract leading to eventual failure.
Cardiac amyloidosis is difficult to diagnose due to a lack of clear symptoms and the disease is often misdiagnosed for others since the symptoms tend to be unrelated to affect organs like the kidney and GI tract among others.
Now, Pfizer is collaborating with Anumana and using their specialization in AI applications for heart disease to develop better ways to diagnose this disease. The idea is based on analyzing the many features of electrocardiograms (EKGs). EKGs have a lot of features that are not visible to the plain eye and using AI to help find these minute deviations in heart signals could help diagnose the disease. Afterall, an EKG measures the voltage changes or membrane potential changes caused by the flow of potassium and some other ions in and out of heart cells. In theory, AI should be able to spot the minute changes in electrical activity due to amyloid buildup since it affects potassium levels in heart cells.
Developing such a diagnostic tool is in Pfizer's interest since they got approval for some drugs that treat cardiac amyloidosis and better tools to diagnose it can help them get their treatment to more people. Overall, this is another example of AI helping in advancing medicine and being used as a potentially very effective tool to diagnose diseases that are otherwise very hard to diagnose.
Read more:
Featured Fake News
In non AI related news, claims about the blood group that was discovered in September, Er, being called a result of “DNA injury” from the mRNA covid vaccines have been circulating this week.
This is completely false.
New blood groups have been discovered and named for decades. This newest group is composed of 5 antigens and we discovered the first one 40 years ago. We just did not know or have enough evidence to claim that it belonged to a new blood group at the time.
Here is a link to the original paper discovering Er.
When emailed about this claim by Reuters, Dr. Timothy Satchwell, one of the original authors replied “There is absolutely no evidence whatsoever to support the claim made, which has no basis in logic or scientific fact, particularly given that the first observations of these antigenic variants were made in the 1980s, prior to COVID and many of the samples in our work are from archived frozen material from many years ago.”
If you are enjoying the Unmasking Medicine Newsletter, why not share it? You can get 3 free stickers for every 3 people who subscribe using this form!