Applications of AI in the Healthcare Industry

Adhya Dagar
Published in
9 min readOct 18, 2021

How AI is disrupting the Healthcare Industry

Photo by Chris Liverani on Unsplash

As the Artificial Intelligence revolution spawns across industries, AI and robotics are now increasingly becoming a part of the healthcare eco-system.

In this analysis, I have discussed how AI can be applied in healthcare across different sub-domains, prevalent health-tech companies bringing about this change and the business potential of this industry.

1. Monitoring well being

With the Internet of Medical Things (IoMT) market estimated to be worth $158.1 billion in 2022, medtech companies are aiming to get IoMT right from a business perspective and use this opportunity to deliver more value to health care.

Photo by Onur Binay on Unsplash

With the ability to generate, collect, analyze and transmit health data, IoMT tools are playing a central part in tracking and preventing chronic illnesses for patients and clinicians.

The IoMT infrastructure consists of sensor based tools including wearables and medical devices, software and health systems applications and services.

In real life, paired with smartphone applications, this technology allows patients to send their health information to doctors in order to better surveil diseases and track and prevent chronic illnesses.

The benefits of integration of AI and IoT to health increases the ability for healthcare professionals to better understand the day-to-day patterns and needs of the people they care for, and with that understanding they are able to provide better feedback, guidance and support for staying healthy.

System Architecture

Overview of solution showing IoMT device sending data to AWS Cloud to be processed
The building blocks of IoMT systems

Implementation Example

  • Twin Health: How it’s using AI in healthcare: Twin Health is pioneering a holistic method of addressing and potentially reversing chronic conditions like Type 2 Diabetes through a mixture of IoT tech, AI, data science, medical science and healthcare. With data from health sensors and blood reports, company’s signature product Whole Body Digital Twin™ learns your body’s unique response to daily life, creating precision treatment to heal metabolism and reverse diabetes.

Market Potential

Goldman Sachs estimates that IoMT will save the healthcare industry $300 billion annually in expenditures primarily through remote patient monitoring and improved medication adherence.

The global IoMT market was valued at $44.5 billion in 2018 and is expected to grow to $254.2 billion in 2026, according to AllTheResearch. The smart wearable device segment of IoMT, inclusive of smartwatches and sensor-laden smart shirts, made up for the largest share of the global market in 2018, at roughly 27 percent, the report finds.

2. Early detection of diseases

Photo by on Unsplash

There are some diseases that do not have the early warning signals.

Further, seeing medical specialists may not be something that many people do regularly. For some, the waiting time to see a medical professional may be a constraint too. This is where AI algorithms can help to do the first level screenings to pick up the subtle details that may point to some underlying issues and then refer them to the specialists.

Diseases and applications

  • Diabetic retinopathy (DR) is a disease which causes lesions on the retina and adversely affects vision. Many diabetic patients develop DR without displaying any symptoms during the initial stages. Unless these asymptomatic patients seek medical help from an eye specialist early, later stages of DR may set in over the course of time and leading to irreversible blindness. DR is one of the leading causes of blindness and prevention is possible with early detection and treatment.
  • Deep Learning can help to mitigate some of the challenges here by detecting DR conditions during the early and asymptomatic stages. This could mean that more diabetic patients can undergo screening and more importantly, early diagnosis is possible so that the patient identified with DR may be referred to the eye specialist for treatment. This is made possible with color fundus photographs (CFP), where deep learning is used to assess these raw CFPs to produce a target outcome prediction. The deep learning algorithm is trained on high-quality CFPs which are graded for DR severity by experts.
  • The deployment of such a predictive DR detection algorithm could help tackle the daunting issue of blindness for diabetic patients and help governments reduce the cost in trying to cope with this problem.
  • Heart conditions can be diagnosed by identifying individual components (segments) in the ECG waveform. This is done by looking for subtle patterns and repeating features to correctly identify each region of the ECG wave. Examples of the heart conditions which can be detected are Coronary Heart Disease which is an occlusion of the vessels that irrigate the heart, Heart Arrhythmias which occur when the electrical impulses that coordinate your heartbeats don’t work properly, and others.The ECG waveform is a time-series signal with periodicity, composing of P waves, QRS complexes, and T waves.
  • This combination of periodicity and feature characteristics of the signal is best analyzed by using a Conv-LSTM deep learning algorithm.

Implementation Example

  • Harvard University’s teaching hospital, Beth Israel Deaconess Medical Center, is using artificial intelligence to diagnose potentially deadly blood diseases at a very early stage
  • Doctors are using AI-enhanced microscopes to scan for harmful bacterias (like E. coli and staphylococcus) in blood samples at a faster rate than is possible using manual scanning. The scientists used 25,000 images of blood samples to teach the machines how to search for bacteria. The machines then learned how to identify and predict harmful bacteria in blood with 95% accuracy.

3. Diagnosis of Medical Conditions

Deep learning is increasingly being used to detect and diagnose diseases using biomedical imaging modalities like radiological images obtained by MRI machines, CT scanners, and X-rays.

A recent study published in the Journal of the National Cancer Institute shows that the AI system has achieved a breast cancer detection accuracy comparable to an average breast radiologist.

Google, in their latest research, proves that a neural network can be trained to detect signs of lung cancer earlier and faster than trained radiologists.

Photo by National Cancer Institute on Unsplash

These AI and deep learning solutions have provide radiologists with essential support as they help manage high imaging volumes, offer them the ability to streamline workflows, save time, increase capacity, diagnosis reliability and efficiency. This reduces the burden on the radiologist significantly.

Implementation Example

  • AI-Rad Companion Chest CT is an AI-powered healthcare solution from Siemens Healthineers that can read the chest CT images, perform automatic measurements, and prepare the medical report with valuable clinical images and quantifications.
  • Zebra Medical Vision provides radiologists with an AI-enabled assistant that receives imaging scans and automatically analyzes them for various clinical findings it has studied. The findings are passed onto radiologists, who take the assistant’s reports into consideration when making a diagnosis.
  •’s qCT-Lung empowers lung cancer screening programs and facilitates opportunistic screening using AI

Market Potential

The global AI in medical diagnostics market is projected to reach USD 3,868 million by 2025 from USD 505 million in 2020, at a CAGR of 50.2% during the forecast period.

Growth in this market is primarily driven by government initiatives to increase the adoption of AI-based technologies, increasing demand for AI tools in the medical field, growing focus on reducing the workload of radiologists, influx of large and complex datasets, growth in funding for AI-based start-ups, and the growing number of cross-industry partnerships and collaborations.

4. Decision Making

The increasing computational power of hardware and AI algorithms is also enabling the introduction of platforms that can link EHRs with other sources of data, such as biomedical research databases, genome sequencing databanks, pathology laboratories, insurance claims, and pharmacovigilance surveillance systems, as well as data collected from mobile Internet of Things (IoT) devices such as heart rate monitors.

AI-assisted analysis of all this big data can generate clinically relevant information in real time for health professionals, health systems administrators, and policymakers.

These Clinical Decision Support Systems (CDSS) are programmed with rule-based systems, fuzzy logic, artificial neural networks, Bayesian networks, as well as general machine-learning techniques (Wagholikar et al. 2012). CDSS with learning algorithms are currently under development to assist clinicians with their decision-making based on prior successful diagnoses, treatment, and prognostication

Using pattern recognition to identify patients at risk of developing a condition — or seeing it deteriorate due to lifestyle, environmental, genomic, or other factors — is another area where AI is beginning to take hold in healthcare.

Implementation Example

  • IBM’s Watson for Health is helping healthcare organizations apply cognitive technology to unlock vast amounts of health data and power diagnosis. Watson can review and store far more medical information — every medical journal, symptom, and case study of treatment and response around the world — exponentially faster than any human.
  • Google’s DeepMind Health is working in partnership with clinicians, researchers and patients to solve real-world healthcare problems. The technology combines machine learning and systems neuroscience to build powerful general-purpose learning algorithms into neural networks that mimic the human brain.
  • Deep Genomic’s AI platform helps researchers find candidates for developmental drugs related to neuromuscular and neurodegenerative disorders. Finding the right candidates during a drug’s development has statistically raised the chances of successfully passing clinical trials while also decreasing time and cost to market. Deep Genomics is also working on “Project Saturn,” which analyzes over 69 billion different cell compounds and provides researchers with feedback.
  • KenSci combines big data and artificial intelligence to predict clinical, financial and operational risk by taking data from existing sources to foretell everything from who might get sick to what’s driving up a hospital’s healthcare costs.

Market Potential

The global clinical decision support system market is expected to reach USD 1.8 billion by 2025 from an estimated USD 1.2 billion in 2020 at a CAGR of 9.1% during the forecast period.

5. Treatment

AI can help clinicians take a more comprehensive approach for disease management, better coordinate care plans and help patients to better manage and comply with their long-term treatment programmes.

Hospitals are using robots to help with everything from minimally-invasive procedures to open heart surgery. According to the Mayo Clinic, robots help doctors perform complex procedures with a precision, flexibility and control that goes beyond human capabilities.

Implementation Example

  • Vicarious Surgical combines virtual reality with AI-enabled robots so surgeons can perform minimally invasive operations. Using the company’s technology, surgeons can virtually shrink and explore the inside of a patient’s body in much more detail.
  • BERG is a clinical-stage, AI-based biotech platform that maps diseases to accelerate the discovery and development of breakthrough medicines. By combining its “Interrogative Biology” approach with traditional R&D, BERG can develop more robust product candidates that fight rare diseases. BERG recently presented its findings on Parkinson’s Disease treatment — they used AI to find links between chemicals in the human body that were previously unknown — at the Neuroscience 2018 conference.
Photo by Towfiqu barbhuiya on Unsplash

6. Research

The path from research lab to patient is a long and costly one. According to the California Biomedical Research Association, it takes an average of 12 years for a drug to travel from the research lab to the patient. Only five in 5,000 of the drugs that begin preclinical testing ever make it to human testing and just one of these five is ever approved for human usage. Furthermore, on average, it will cost a company US $359 million to develop a new drug from the research lab to the patient.

Drug research and discovery is one of the more recent applications for AI in healthcare. By directing the latest advances in AI to streamline the drug discovery and drug repurposing processes there is the potential to significantly cut both the time to market for new drugs and their costs.

Pfizer is using IBM Watson, a system that uses machine learning, to power its search for immuno-oncology drugs. Sanofi has signed a deal to use UK start-up Exscientia’s artificial-intelligence (AI) platform to hunt for metabolic-disease therapies, and Roche subsidiary Genentech is using an AI system from GNS Healthcare in Cambridge, Massachusetts, to help drive the multinational company’s search for cancer treatments. Most sizeable biopharma players have similar collaborations or internal programmes.

If the proponents of these techniques are right, AI and machine learning will usher in an era of quicker, cheaper and more-effective drug discovery.




Adhya Dagar

Computer Science Engineer| AI for Social Good| Social Entrepreneurship|