Artificial Intelligence: Coming Soon to a Medical Office Near You?

FIGURE 1: Virtual Colonoscopy (VC): 3D imaging of a colon superimposed over a human skeleton – VC software is an example of one of the successful early integrations of technology and imaging.


“I am human.” These three words seem simple, but their complexity is not to be overlooked. Being human signifies that we have the potential to understand our weaknesses, the intelligence to invent and the compassion to help each other. This way of thinking has allowed the worlds of radiology and medicine to advance so dynamically. Yet, our limitations as humans are continuously being tested. We live in an era where the amount of information available can prove to be a double-edged sword. With such vast data, the task of isolating a single disease can sometimes prove to be as overwhelming as locating a single drop of water in the ocean. So, where do we go from here? The answer may lie in a symbiotic partnership between the medical community and artificial intelligence (AI) – the collision of two worlds.



Only a decade ago, the concept of artificial intelligence was largely science fiction ... remember the movie Minority Report? iRobot? Terminator?

While Hollywood is having fun scaring everyone, scientists have been busy researching. In just the last four years1, AI has been demonstrated to have benefits in nearly every realm of medicine: cancer2, neurological disorders3, cardiovascular disease4, urogenital abnormalities5, pregnancy6, digestive trauma7, respiratory ailments8, skin anomalies9, endocrine defects10 and nutrition11. For example, algorithms are already helping radiologists determine if lung nodules are likely to be cancerous.

Until recently, the thought of using computers to analyze patterns in clinical imaging as part of the diagnostic care for an ill patient seemed futuristic. However, our passion to improve the quality of life for our patients has enabled us to turn science fiction into science fact.



Imagine how healthcare might be delivered on the iPhone 25. With increasing sensors and computing power, doctors may ask to speak with our smartphones!

It is human nature to question what we do not fully understand – and modern artificial intelligence is as complicated as any topic in computer science. With AI just around the corner, medical professionals will soon be faced with a choice: adapt to the early stages of a new paradigm, or continue with familiar, yet outdated, methods.

The leap may not be as massive as it seems, though. Radiologists already use software on a wide range of images from breast MRI to virtual colonoscopy (see Figure 1). Several of these tools did not exist just ten years ago, and are now standard of care.

Medical staff can and will adapt to the ‘futuristic’ concept of working alongside a program that studies in a similar fashion to another human, and AI software will require countless hours of ‘training’ to learn to recognize patterns in patient medical records (e.g. scanned images and medical history).



A combination of human knowledge and technological capability that will allow for true progress to take place!

One of the most profound fears among the medical community (as well as the population at large) is that humans will be replaced by machines. In the case of image analysis, for instance, complex ‘deep-learning’ algorithms could allow AI to sift through patterns of abnormality among thousands of images, locate related patient data in electronic medical records (EMR), and assist with diagnosis development in a fraction of the time required by a trained physician. Imagine the computer knowing that your appendix is likely to burst before you even talk to a doctor! Seems cold and clinical, right?

However, try to envision the landscapes of radiology and medicine only 10 or 20 years ago. The technology of that time was nothing like it is today – MRI only gained widespread adoption recently. And yet, humans have not been replaced because humans and technology have limitations that require us to depend on one another.

  • Humans are not outfitted with the capability to store, categorize, and extract thousands of data files from our system within seconds.
  • Computers are not designed to listen to the patient and develop a diagnosis based on ‘outside-the-box’ thinking.

It is the combination of human knowledge and technological capability that will allow for true progress to take place. So, it may not necessarily be a question of whether AI will replace humans, but rather, will humans make the right choices working with AI to significantly improve our lives. If history is a guide, the future is bright for man and machine. 



Jiang F, Jiang Y, Zhi H, et al. (2017) Artificial intelligence in healthcare: past, present and future. Stroke and Vascular Neurology, 0.

Somashekhar SP, Kumarc R, Rauthan A, et al. (2016) Validation study to assess performance of IBM cognitive computing system Watson for oncology with Manipal multidisciplinary tumour board for 1000 consecutive cases: An Indian experience. Annals of Oncology, 27 (Suppl_9):ix179-ix180.

Farina D, Vujaklija I, Sartori M, et al. (2017) Man / machine interface based on the discharge timings of spinal motor neurons after targeted muscle reinnervation. Nat Biomed Eng, 1:0025.

Dilsizian SE and Siegel EL. (2014) Artificial intelligence in medicine and cardiac imaging: harnessing big data and advanced computing to provide personalized medical diagnosis and treatment. Curr Cardiol Rep, 16:441.

Montironi R, Görtler J, Kayser K, Borkenfeld S, Carvalho R, and Kayser G. (2017) Cognitive Algorithms and Digitized Tissue-based Diagnosis. Diagnostic Pathology, 3:248.

Bartmann AK, Junior MF, Pulice de Barros G, Sampaio de Paula L, Bettini NR. (2016) The Number of Embryos Obtained can offset the Age Factor in IVF Results According to an Artificial Intelligence System. Women’s Health & Gynecology, 2 (5):035.

Kumar R, Sharma A, Siddiqui MH, Tiwari RK. (2017) Prediction of Human Intestinal Absorption of Compounds Using Artificial Intelligence Techniques. Curr Drug Discov Technol, 14(4):244-254.

Rodríguez JC, Arizmendi CJ, Forero CA, Lopez SK, Giraldo BF (2017) Analysis of the respiratory flow signal for the diagnosis of patients with chronic heart failure using artificial intelligence techniques. VII Latin American Congress on Biomedical Engineering CLAIB, 60:481-482.

Esteva A, Kuprel B, Novoa RA, Ko J, Swetter SM, Blau HM, and Thrun S. (2017) Dermatologist-level classification of skin cancer with deep neural networks. Nature, 542: 115-118.

Ersoy R, Aydin C, Topaloglu O, et al. (2017) Predictive value of CHAID Algorithm in the diagnosis of malignancy in thyroid nodules with Bethesda III (AUS/FLUS) cytology. Endocrine Abstracts, 49 (EP1377).

Bourguet JR, Thomopoulos R, Mugnier ML, Abecassis J. (2013) An artificial intelligence-based approach to deal with argumentation applied to food quality in a public health policy. Expert Systems with Applications, 40 (11):4539-4546.