Skip to main content
Dr. David Holland

“Artificial Intelligence – Just What The Doctor Ordered” presented by Dr. David Holland

Caitlyn Vlasschaert, PhD student (Translational Medicine) and Queen’s Internal Medicine Resident

In 1997, “Man Loses to Computer” was splayed across newspaper headlines worldwide after Garry Kasparov– the reigning grandmaster of chess– was defeated by the IBM Deep Blue. The breadth of innovation achievable through artificial intelligence (AI) has since captivated and frightened us in its multiple implementations. During his Medical Grand Rounds lecture on February 18th, Dr. David Holland walked us through the history of AI and shared what can (and cannot) be done with AI in 2021. This framework allowed us to explore the question of the hour: will AI replace physicians?

 

The roots of artificial intelligence (AI) can be traced to Alan Turing, who derived the framework for machine-based computation with his invention of the Turing Machine in 1936. Technological advances have since broadened the realm of feasibility in digital computing. Computing power has doubled every year or so, following a rapid expansion known as Moore’s law [1]. Many of us now hold powerful compact computers in the palms of our hands. AI-based pattern recognition software installed on our cellphones (and other advanced computers) enable digitization of the living world. These perceptron-based algorithms convert faces, drawings, speech to digital data– analogous to human sensory perception. How does the computer then discriminate what exactly this input is? Artificial neural networks– a type of machine learning named after its functional similarity to the interconnected neurons of the brain– can be trained to classify data. As the classification task increases in complexity, multi-layered neural networks are required, which is referred to as deep learning. Deep learning pattern recognition algorithms are being adapted to assist with visual diagnostics in medicine, from interpretating ECGs and radiographic images to identifying skin lesions [2–4].

 

Deep learning is also being used to predict medical outcomes using electronic health record (EHR) data. DeepMind, a Google-owned AI company that rose to prominence in 2016 for its successful AlphaGo program [5], recently developed an application called Stream that can forecast acute kidney injury (AKI) from laboratory data and warn care providers of imminent renal threats [6]. These deep learning algorithms are trained using large amounts of data with labelled outcomes. In some cases, data labelling is relatively simple: AKI, for example, is identified by numerical changes in serum creatinine. But in other areas of medicine, data labelling can be a challenge, as Dr. David Maslove shared during the Q&A period. Dr. Maslove directs a machine learning-focused critical care research program at Queen’s (http://www.conduitlab.org/). He discussed that, for instance, in order to develop a neural network that accurately identifies complex heart rhythms from ICU tracings, one needs to train the algorithm using multiple reference points, which requires significant human resources.

 

After his presentation, Dr. Holland sat down with the graduate students in the Translational Medicine (TMED) Program. We discussed several ethical implications of AI in healthcare, including data privacy and algorithm biases. As mentioned, these algorithms are trained using user-defined data; if these data represent a biased sample, then this can perpetuate bias in practice [7]. Conversely, if carefully implemented, AI algorithms can help us address racial disparities and unconscious bias in healthcare as well as assist in the delivery of care to underserviced regions globally [8].

 

As of 2021, AI capabilities are restricted to task-oriented functions. As such, Dr. Holland is confident that the complex roles of many healthcare workers including physicians will not be replaced by AI anytime soon. With the advent of AI-assisted visual diagnostics and flagging of EMR trends, healthcare providers may however benefit from formal training in AI-assisted medicine in order to responsibly harness its benefits in their practice.

References

  1. Moore GE. Cramming more components onto integrated circuits. Electronics. 1965 Apr 19; 38(8). Available at: https://newsroom.intel.com/wp-content/uploads/sites/11/2018/05/moores-l….
  2. Ribeiro AH, et al. Automatic diagnosis of the 12-lead ECG using a deep neural network. Nat Commun. 2020 Apr 9;11(1):1760.
  3. Montagnon E, et al. Deep learning workflow in radiology: a primer. Insights Imaging. 2020 Feb 10;11(1):22. PMID: 32040647.
  4. Esteva A, et al. Dermatologist-level classification of skin cancer with deep neural networks. Nature. 2017 Feb 2;542(7639):115-118.
  5. DeepMind. AlphaGo: the story so far. [Website] Available at: https://deepmind.com/research/case-studies/alphago-the-story-so-far.
  6. Tomašev N, et al. A clinically applicable approach to continuous prediction of future acute kidney injury. Nature. 2019 Aug;572(7767):116-119. PMID: 31367026.
  7. To, SB. Humans are the cause of bias in AI, but we’re also the solution. Forbes. 2020 Mar 3. [Website] Available at:  https://www.forbes.com/sites/forbestechcouncil/2020/03/03/humans-are-the-cause-of-bias-in-ai-but-were-also-the-solution/
  8. Pearl, R. How AI can remedy racial disparities in healthcare. Forbes. 2021, Feb 16. [Website] Available at: https://www.forbes.com/sites/robertpearl/2021/02/16/how-ai-can-remedy-r…

Comments

Name
Max Moloney

Mon, 02/22/2021 - 13:03

Thank you, Caitlyn, for your excellent summary of Dr. Holland’s presentation on Artificial Intelligence in Medicine during Medical Grand Rounds last Thursday.

Dr. Holland’s comprehensive presentation on artificial intelligence, from its beginnings with IBM Deep Blue to its current practical applications in medicine allowed the TMED students to gain an appreciation for how AI will impact the world in the future. An important takeaway I had from Dr. Holland’s presentation was the importance of ensuring AI assists medical experts as opposed to experts assisting the AI in the future. This point relates to another focal point of Dr. Holland’s presentation, the potential benefits and threats that general artificial intelligence poses to humanity in the future, a possibility that is critical to understand for all researchers as AI continues to improve.

Additionally, I would like to thank Dr. Holland for participating in a candid discussion on his career path which led him to Queen’s. Dr. Holland’s insight on the benefits and challenges of being a physician educator provided students with a unique perspective on balancing two demanding roles as both a physician in the clinic and an educator at Queen’s University.

Name
Max Moloney

Name
Melinda Chelva

Mon, 02/22/2021 - 14:11

Excellent post Caitlyn! You have effectively highlighted all of the topics that were covered during the MGR last week.

Thank you for also eloquently leading the facilitated discussion- I truly enjoyed our conversation with Dr. Holland. Similar to Max, the key point that AI should only be used as an “assistant” to physicians instead of experts assisting AI, continues to resonate with me.

As discussed, physicians consider social determinants of health and risk factors in order to make well-informed decisions regarding patient health. However, I wonder whether deep learning machines are being fed with this information? For example, genetic composition often plays a very critical role in cancer diagnosis. Without deep learning machines also being provided with such information, in addition to the images they are given to screen and detect for cancer, how can we be confident that the decisions made will be accurate? In other words, I worry that the decisions that these machines make will be less holistic, as a result. Overall, I am curious to learn more about whether algorithms for AI accommodate for information related to social determinants of health and risk factors.

Ultimately, by the end of our discussion with Dr. Holland, I was left feeling inspired to continue to learn more about AI and its potential in healthcare over the years!

Name
Melinda Chelva

Name
Charmi Shah

Tue, 02/23/2021 - 00:04

Thank you Caitlyn for effectively summarizing Dr. Holland’s presentation from last week’s MGR. I greatly appreciate Dr. Holland’s enthusiasm for learning, and this is displayed in his interest to present topics that not only inspire his students but also himself.

The MGR discussion reminded me of the healthcare phrase, “When you hear hoofs, think horse, not zebra.” This refers to the odds that patients are likely to have a more common diagnosis than a rare one. It was mentioned that during clinical practice, this type of thinking is beneficial in making diagnoses but may also cause an unconscious bias, so this may be an area where a seemingly more-objective AI system would be beneficial in providing accurate diagnoses.

However, it is crucial to remember that these AI algorithmic approaches can only be as accurate and reliable as the data they are provided. Relating this back to the experience of the insurance company, Optum, that used a healthcare algorithm to assess the cost of each patient’s past treatments with the goal to help re-distribute medical resources to those who would benefit most from added care. When researchers revisited the patients by their illnesses they found that the percentage of Black patients who should have been enrolled in specialized care programs jumped from 18% to 47%, as stated here: https://www.forbes.com/sites/robertpearl/2021/02/16/how-ai-can-remedy-r…. This seems like racial bias within the algorithm, but the issue was with the physicians who failed to provide adequate care to Black patients in the first place, so Black patients were recorded as needing less medical resources. As Caitlyn discussed, we need carefully applied AI healthcare algorithms to help move our healthcare system towards a path of antiracism.

I would like to thank Dr. Holland again, for the thought-provoking presentation.

Name
Charmi Shah

Name
Michaela Spence

Tue, 02/23/2021 - 12:54

Thank you Caitlyn for the excellent summary and to Dr. Holland for the wonderful and informative presentation at Grand Rounds.

One of the things that resonated with me during our discussion period was the concept of patient’s “owning their own data”. As I had very little idea of how AI was currently being implemented in healthcare settings and the amount of data that the algorithms require to be successful, it led me to wonder if the general public also had very little knowledge of this topic as well. I wonder if patient’s were in control of their own data got to decide who and what had access to it, if it would facilitate more research from the patient’s side as to what exactly medical AI systems are and how they help or sometimes hinder the medical system. This could cause people to shy away from how medical AI is presented in the media as a “machine that does the doctors work for them” and more to the perspective that AI is the “assistant” to medical professionals. It would also potentially mitigate some privacy concerns that go hand in hand with the large data pools required for AI algorithms to function in the first place. I’m hopeful that as the AI algorithms become more developed and useful to health care professionals, that the general public will take it upon themselves to become more informed as to what exactly AI algorithms do in a healthcare setting.

Overall, this talk left me hopeful that the use of AI in future healthcare could help lighten the load on medical professionals.

Name
Michaela Spence

Name
Jordan Harry

Wed, 02/24/2021 - 11:09

Thank you, Caitlyn, for an impressive and detailed overview of Dr. Hollands presentation and the following discussion section.
As a person aspiring to become a physician, the question “will AI replace physicians” is of vital interest. As we phase into an era of integrating advanced technologies such as AI into everyday patient care, it calls into question what its benefits and disadvantages are in comparison to current clinical practice. Dr Holland as well as my classmates have accurately highlighted some of these, but it is also important to note that in such an emerging field that there are additional positives and negatives that we have yet to be exposed to.
I think that the concept of AI also further highlights the complexity of the role of healthcare workers. Humans can establish a human connection, develop compassion, and can take all patient information and contextualize it. Resultingly, human healthcare workers will remain superior to AI. Dr. Holland’s suggestion to not allow AI to become the master, but rather serve as an assistant is a strong suggestion regarding how to combine technology and humanity to develop an overall improvement in the current approach to patient care.

Name
Jordan Harry

Add new comment

Plain text

  • No HTML tags allowed.
  • Lines and paragraphs break automatically.
  • Web page addresses and email addresses turn into links automatically.