Best examples of artificial intelligence (AI) transforming healthcare


Healthcare is one of the most significant success stories of our time. Medical science has advanced rapidly, increasing global life expectancy. Still, as people live longer, healthcare systems face increased demand for their services, ever-rising costs, and a workforce struggling to meet their patients’ needs.

Artificial intelligence (AI), driven by automation, has the potential to revolutionize healthcare and solve the majority of problems. AI has the potential to improve care outcomes, increase productivity and efficiency in care delivery, and transform how care is delivered.

It can also improve healthcare practitioners’ daily lives by allowing them to spend more time caring for patients, raising staff morale, and improving retention. It can even expedite the release of life-saving treatments.

Finally, it can help with faster care delivery by shortening diagnosis times. It can assist healthcare systems in managing population health more proactively, allocating resources where they will have the greatest impact.

This post presents the best examples of transforming healthcare with AI.

Personal Electrocardiogram (ECG)

AliveCor, a company based in the United States, created KardiaMobile. This personal ECG device can monitor heart rhythm and instantly detect and alert clinical teams to atrial fibrillation, bradycardia, or tachycardia. In Europe, 7.6 million people over 65 suffer from atrial fibrillation, a figure expected to rise to 14.4 million by 2060.

How does it function? Patients can remotely record their ECGs with KardiaMobile and the Kardia app. A single or six-lead ECG can be obtained without additional leads and cables, depending on the patient’s monitoring requirements. The Kardia app allows patients to track data over time and directly share ECG recordings with their doctor.

What does this mean for medical professionals and organizations? On the KardiaPro platform, physicians can remotely monitor patient data. They can collect additional patient data points between visits to create a complete picture of the patient’s heart rhythm, which can then inform clinical decision-making. ECGs may reveal intermittent atrial fibrillation missed during periodic measurements at the doctor’s office, prompting a treatment change.

Online symptom checkers/e-triage tools

These tools improve healthcare access by allowing patients to check common pathologies typically addressed in primary care and providing information on related symptoms, potential treatments, and outcomes. Some companies provide follow-up via online chat or video consultations with doctors.

Today’s symptom checkers are broadly similar, but some significant differences exist (e.g., algorithm training, data access, or scale).

Key examples include:

  • Babylon Health’s AI-powered chatbot understands symptoms defined in the patient’s words and provides relevant health and triage information using algorithms trained on NHS data. It provides an initial diagnosis, possible scenarios, and a percentage-based estimate of how likely each will be correct. Babylon is also developing a technique inspired by quantum cryptography to search medical databases for causal links.
  • Mediktor employs an e-triage approach validated in a prospective observational study in a tertiary hospital emergency department.
  • K Health was created in collaboration with Maccabi Health Services in Israel, which provided the company with anonymized electronic medical record data from over 2 million people over the past 20 years. Using NLP and advanced modeling, the algorithm was trained to understand symptoms and the likelihood of an underlying diagnosis being associated with a patient’s data. A built-in feedback loop allows the system to learn from each case and continuously improve.
  • Ada Health (Germany) has a system that connects its medical knowledge with artificial intelligence. The app compares answers to symptom questions to similar cases drawn from extensive clinical literature.
  • The symptom checker by Ping An Good Doctor (China) is an integral part of a closed-loop ecosystem that connects patients with physicians, online or offline, after initial assessment. It is also physically deployed as an AI-enabled virtual doctor in Ping An’s One-MinuteClinics. Contracts for nearly 1,000 units in large and medium-sized enterprises, community centers, chain pharmacies, and other high-traffic areas have been signed-in eight Chinese provinces.

Symptom checkers can increase productivity by reducing the time practitioners spend collecting data and forming an early impression of patients, as well as helping to reduce the risk of misdiagnosis. They may also relieve pressure on primary-care providers and organizations, resulting in fewer people visiting emergency rooms and reducing overall caseloads. This allows organizations to devote more time to patients with the greatest needs (though patients who eventually present at the hospital may have more complex needs on average).

Sight Diagnostics

Sight Diagnostics, an Israeli company, has created OLO, a point-of-care blood testing device that can perform a full blood count (FBC) using AI machine-vision technology. The AI powering the blood diagnostics system was trained using nearly half a petabyte of anonymized blood image data. The device allows a healthcare professional to perform an accurate test from a finger prick in 10 minutes and requires little training. As a result, it is appropriate for use in primary care settings, emergency departments or outpatient settings, and settings without a lab.

How does it function? The AI technology interprets multiple images of a small blood sample to produce an FBC test comparable to a traditional laboratory test. Point-of-care blood testing allows the healthcare professional to receive results much faster, potentially allowing the patient to be diagnosed immediately rather than waiting hours or days. It also eliminates the need to transport, track, and test samples in a lab. Using a finger prick reduces the need for a larger blood sample to be drawn from a vein, which may necessitate phlebotomy services. The AI approach also eliminates the need for device calibration, typically performed by or under the supervision of lab technicians.

Amelia – Virtual Health Agent Platform

Amelia, an IPSoft cognitive virtual agent platform, demonstrates learning abilities and elements of emotional intelligence. It can manage some operational and administrative hospital processes and perform autonomic task management using conversational AI.

Amelia can act as a care protocol “whisper agent” (for example, reminding practitioners of steps that need to be taken) as well as a care operations agent, assisting in the documentation of a patient visit, admitting patients, retrieving medical history before a conversation, checking the availability of hospital beds, retrieving lab results, and scheduling specialist appointments. Amelia Health agents, powered by AI technology, continuously learn with each completed task and can communicate via voice, mobile, web, and chat.

Bionic Pancreas

The bionic pancreas (iLet, developed by the US company Beta Bionics) mimics the function of the pancreas by constantly monitoring and independently managing blood sugar levels in insulin-dependent type 1 diabetes patients. This could provide critical support for patients, particularly adolescent patients, who find strict monitoring and insulin-management regimens extremely restrictive. The iLet device is worn on the skin and wirelessly connects to a smartphone-sized portable unit containing the hormone (s). Through an algorithm, it specifies the timing and dosage to be administered.


Sensely, a company based in the United States, provides a virtual nurse assistant with chronic disease modules that can be used for personalized monitoring and follow-up care. Patients can use the virtual assistant on a tablet at home to help them manage their care and communicate with healthcare providers.

Sensely’s avatar-based chronic-care platform uses text-to-speech and speech-recognition technologies to provide personalized conversational content. It assists the patient in navigating daily monitoring requirements and can assess symptoms to determine whether a healthcare professional should be contacted. The assistant walks the patient through the process step by step (e.g., “Now it’s time to take your blood pressure; please make sure the cuff is on by pressing the orange button”) and provides immediate feedback (“Your blood pressure is a little high today”). The solution, which employs speech recognition, also enables the patient to communicate with the virtual assistant, for example, to report symptoms.


Karantis360, a UK-based company, has created an automated, personal monitoring and alerting system that allows elderly people to live independently while keeping caregivers and families informed.

Karantis360 has collaborated with IBM Watson and EnOcean to deliver a comprehensive solution that combines AI and Internet of Things capabilities with intelligent sensors connected to a mobile device. The device transmits data to caregivers and family members via a web and mobile dashboard and can send reports and alerts. The sensors can provide data about the patient’s daily activities, such as when they get up or go to bed, use the bathroom, or leave the house. Using AI, the system detects deviations from normal behavior that may indicate a problem, such as a fall. It notifies caregivers and families of these abnormalities to inform the action plan.