What is Artificial Intelligence in medicine?

Artificial Intelligence (AI) is computer science that uses algorithms, heuristics, pattern matching, rules, deep learning and cognitive computing to approximate conclusions without direct human input. By using AI, researchers can take on complex problems that would be difficult – or almost impossible – for humans to solve. Because AI can identify meaningful relationships in raw data, it can be used to support diagnosing, treating and predicting outcomes in many medical situations. AI has the potential to be applied in almost every field of medicine including drug development, patient monitoring and personalized patient treatment plans. 

Artificial Intelligence is patterned after the brain’s neural networks. It uses multiple layers of non-linear processing units to “teach” itself how to understand data – classifying the record or making predictions. AI can synthesize electronic health record (EHR) data and unstructured data to make predictions about patient health. For instance, AI software can quickly read a retinal image or flag cases for follow up when multiple manual reviews would be too cumbersome. Doctors benefit from having more time and concise data to make better patient decisions.

AI can be used in variety of ways in medicine. Here are four examples:

  • Annotator for clinical data. Around 80 percent of healthcare data is unstructured, and AI can read and understand unstructured data. AI’s ability to process natural language allows it to read clinical text from any source and identify, categorize and code medical and social concepts.
  • Insights for patient data. Artificial intelligence can identify the problems contained in patients’ historical medical records – both in the structured and unstructured text. It summarizes the history of their care around those problems and can provide a cognitive summary of a patient records.
  • Patient similarity. AI can identify a measure of clinical similarity between patients. This allows researchers to create dynamic patient cohorts, rather than static patient cohorts. It also enables an understanding which care path works better for a given group of patients.
  • Medical insights. With AI technologies, researchers can find information in unstructured medical literature to support hypotheses – helping in the discovery of new insights. AI can read through a complete set of medical literature, such as Medline, and identify the documents that are semantically related to any combination of medical concepts.

Evolution of Artificial Intelligence in medicine

Artificial Intelligence in medicine goes as far back as 1972 with Stanford University’s MYCIN – an AI prototype program used for treating blood infections. Early AI research continued at largely US institutions including MIT-Tufts, Pittsburgh, Stanford and Rutgers. In the 1980s Stanford continued its medical AI work with the Stanford University Medical Experimental Computer – Artificial Intelligence in Medicine (SUMEX-AIM) project. 

While AI has been touted as being “the next big thing” for decades, widespread practical uses only began to take off in the 2000s. AI has drawn more than $17 billion in investments since 2009 and will likely grow to $36.8 billion by 2025.

Before the widespread use of AI in medicine, predictive models in healthcare could only consider limited variables in well-cleansed health data. With AI, neural networks can process masses of raw data and learn how to organize that data using the most important variables in predicting health outcomes.

Today, AI technologies such as IBM Watson are being used at Memorial Sloan Kettering Cancer Center to support diagnosis and create management plans for oncology patients. Watson is accomplishing these plans by effectively synthesizing millions of medical reports, patient records, clinical trials and medical journals. Watson’s results are routinely “out-diagnosing” medical residents in certain situations. IBM has also partnered with CVS Health for chronic disease treatment using AI technology. Johnson & Johnson and IBM are using AI to analyze scientific papers to find new connections for drug development.

Other examples of AI currently being used in medicine include patient care in radiology. AI can search and quickly interpret billions of data points – both text and image data – within the patient’s electronic medical record. It can do this using other patient similar cases and across the most up-to-date medical research.

In genomics, AI can extract unstructured data from peer-reviewed literature to continually grow its knowledge base. It provides variant information and clinical content that is up to date – based on the latest approved therapeutic options including targeted and immunotherapy options, professional guidelines, biomarker-based clinical trial options, genomic databases and relevant publications.

Why is Artificial Intelligence in medicine important?

Artificial Intelligence in medicine is important because it can potentially optimize the care trajectory of chronic disease patients, suggest precision therapies for complex illnesses and improve subject enrollment into clinical trials.

Other reasons why AI in medicine is important include:

  • Keeping abreast of mountains of data. Medical data is expected to double every 73 days by 2020. AI can make sense of the overwhelming amount of clinical data, genomic data and social determinants of health data to find the best path for each patient.
  • Providing contextual relevance. AI can empower physicians to see expansively by quickly interpreting billions of data points – both text and image data – to identify contextually relevant information.
  • Improving clinical reliability. AI helps physicians reliably recognize medical solutions by aggregating and displaying information that may otherwise be easily overlooked. In 2016, IBM Watson AI technology was able to cross-reference 20 million oncology records quickly and correctly diagnose a rare leukemia condition in a patient.
  • Helping physicians communicate objectively. AI can assist by analyzing structured and unstructured patient data and presenting insights for physicians’ consideration.
  • Reducing errors related to human fatigue. Human error is costly and human fatigue can cause errors. Artificial intelligence tools don’t suffer fatigue, distractions, or moods. They can process vast amounts of data at incredible speed and out-perform humans in terms of accuracy.
  • Decreasing mortality rates. AI can help reduce death rates by prioritizing patients in more urgent need. It can also help by recommending individualized treatments.
  • Diminishing medical costs. Frost & Sullivan reports that AI has the potential to improve outcomes by 30 - 40 percent and reduce the cost of treatment by up to 50 percent. Also, new drug development and vaccines are time-consuming and costly. AI can be used to process the estimated 30 million lab and data reports.
  • Identifying diseases more readily. AI can quickly and more accurately spot signs of disease in medical images such as MRIs, CT scans, ultrasounds and x-rays. Patients can be diagnosed faster and can begin treatment sooner.
  • Increasing doctor/patient engagement. Physicians spend more time on data entry and desk work than engaging with patients. AI can automate paperwork and free up a physician’s time to see patients.

Using software for Artificial Intelligence in medicine

Chen, Argentinis and Weber point out that life sciences researchers are under pressure to innovate faster than ever. Big data offer the promise of unlocking novel insights and accelerating break-throughs. Ironically, although more data are available than ever, only a fraction is being integrated, understood and analyzed. The challenge lies in harnessing volumes of data, integrating the data from hundreds of sources and understanding their various formats.

New technologies such as Artificial Intelligence offer promise for addressing this challenge because cognitive solutions are specifically designed to integrate and analyze big data sets. AI software can understand different types of data such as lab values in a structured database or the text of a scientific publication. These software solutions are trained to understand technical, industry-specific content and use advanced reasoning, predictive modeling and machine learning techniques to advance research faster.

IBM is one of the pioneers that has developed AI software for specifically for medicine. More than 230 healthcare organizations worldwide use IBM AI (Watson) technology.


Here are some of the key software offerings:

IBM Watson for Oncology combines leading oncologists’ deep expertise in cancer care with the speed of IBM Watson to help clinicians as they consider individualized cancer treatments for their patients.

IBM Watson for Genomics enables molecular pathology laboratories to scale their precision oncology programs to meet the existing and growing needs of personalized cancer care.

Watson for Clinical Trial Matching helps clinicians quickly find a list of clinical trials for an eligible patient – and it assists clinical trial coordinators in finding qualified patients.

IBM Watson for Drug Discovery reveals connections and relationships among genes, drugs, diseases and other entities by analyzing multiple sets of life sciences knowledge. 

Watson Care Manager helps organizations unlock and integrate the breadth of information from multiple systems and care providers – while automating care management workflows to meet the demands of growing populations. 

Watson Imaging Clinical Review is a retrospective AI-enabled data review tool that highlights both primary diagnoses and incidental findings – which may help limit the need to retest patients.

IBM Watson Imaging Patient Synopsis is a radiologist-trained AI tool that can extract patient information from the electronic health record and project it through a single-view summary in sync with PACS (picture archiving and communication system).

Future direction of Artificial Intelligence in medicine

Artificial Intelligence can analyze large amounts of data and turn that information into functional tools that can assist both doctors and patients. The increased integration of AI into everyday medical applications might improve the efficiency of treatments and lower costs in various ways.

Health IT Analytics lists several potential directions for AI for medicine in the near future:


  • Integrate mind and machine. Brain-computer interfaces (BCIs) backed by AI could restore or augment motor functions in some patients.
  • Create new radiology tools. AI-enhanced radiology tools could provide enough accuracy to replace tissue samples.
  • Provide care access to the underserved. AI could take over diagnostic functions in areas where there are too few clinicians.
  • Make EHRs more effective. AI can help automating the filling out of electronic health records (EHR) and making functions more intuitive. EHRs could also be turned into a reliable risk predictor, identifying hidden connections between data sets. 
  • Decrease the risks of antibiotic resistance. EHR data could be used to spot infection patterns and warn patients at risk – even before they experience symptoms.
  • Offer more accurate analytics for pathology images. Because AI can scan images down to individual pixels, researchers may be able to to identify nuances that may escape the human eye.
  • More effectively use immunotherapy for cancer treatment. AI can analyze complex data sets, allowing it to possibly target therapies for a person’s unique genetic makeup.
  • Leverage wearables, personal devices and smartphones for data and diagnostics. AI could play a major role in extracting the large amount of data contained in hand-held devices. Smartphones can produce images that are viable for analysis by artificial intelligence algorithms.


Case studies for Artificial Intelligence in medicine


Learn how Pfizer is accelerating immune-oncology research with Watson for Drug Discovery.

Barrow Neurological Institute

See why the institution uses IBM Watson for Drug Discovery to search research documents in its ALS studies.

Manipal Hospitals

Learn how Manipal is using Watson technologies to help oncologists provide cancer patients with individualized healthcare plans.

Quest Diagnostics

See how Watson is helping physicians bring precision cancer treatments to patients nationwide.

Mayo Clinic

Learn how Mayo Clinic’s clinical trial matching sees higher enrollment in breast cancer trials.

Montgomery County Juvenile Court

Learn how Watson Health is helping a juvenile treatment court improve efficiency and outcomes.

Call us at 1-855-878-8361 | Priority code: Watson Health

Visit us

Visit us