Digital health is an area with many start-ups exploring possibilities to bring patients and healthcare professionals closer together. Many of these solutions explore how artificial intelligence (AI) can improve healthcare, especially for those monitoring their health at home.
There are many potential benefits of AI in digital health. AI can diagnose diseases earlier and more accurately than ever before. It can identify patterns in data that would be difficult for humans to spot. People with wearable devices can monitor their health and become familiar with what is normal and abnormal, so they can seek early treatment if needed.
By providing new tools for doctors and patients, AI has the potential to improve patient care and outcomes. As digital health continues to grow, I see AI playing an increasingly important role in its development.
Use cases for AI and digital health
Artificial intelligence use cases in healthcare have increased in recent years as more researchers and healthcare start-ups have sought new ways to improve healthcare for individuals and support delivery for healthcare professionals. As AI technology continues to evolve, we can expect even more transformative changes in how healthcare is delivered. Some critical emerging use cases of AI in digital healthcare include:
Blood pressure monitoring
Blood pressure is a key factor in cardiovascular health. High blood pressure (hypertension) can increase a person’s risk of developing deadly health problems like stroke and heart disease.
Monitoring a person’s blood pressure is a standard checkup point in the doctor’s office. You would be familiar with the cuff and device used to measure blood pressure and you might even have such a machine in your home if you need to monitor your blood pressure readings frequently. While these remain essential methods for measuring blood pressure, readings are not always accurate due to several factors, such as the person’s body movements or the cuff fitting.
An AI solution by the start-up CardioX leverages an ECG taken by a person’s wearable device to estimate their blood pressure reading and offers to send the result to the attending physician. While this method requires cuff-based devices to give definitive readings, it provides frequent data. It illuminates changes in the person’s blood pressure, leading to earlier detection and treatment of conditions such as hypertension.
Analysing blood sugar levels
Sustained high blood sugar levels can cause health problems and damage major organs, the nervous system and the eyes. As such, analysing blood sugar levels is critical for people with diabetes. Finger pricking has remained the key method for many years. But, finger pricking is not always convenient, or the person might forget to check their blood sugar levels on a particular day. New wearable devices, alongside AI, are set to change that for people living with the condition.
One example of a wearable AI device is the GluCare system. People can wear the device on the part of their body that is most convenient for them. The device continuously monitors the person’s blood sugar levels and uses AI to recognise when blood sugar levels drop or increase to alert the wearer. Doctors can then leverage the data collected to inform their decisions about the person’s treatment.
Diagnosing diseases from chest x-rays
Healthcare practitioners often leverage x-rays to find and diagnose lung cancer and other pulmonary diseases. They can leverage AI with radiology to recognise health problems on x-rays. AI analyses images much faster than humans and highlights areas of concern for a physician to observe more thoroughly. While doctors’ well-trained eyes can pick up on abnormalities, some may be hard to recognise on a scan, or the doctor might benefit from a program to assist them when time is short.
Another element of AI and radiology is taking the x-rays themselves. Like the blood pressure cuff, a few factors can impact the quality of an x-ray, such as positioning the patient and equipment properly. One company, Carestream, aims to improve imaging with intelligent solutions to ensure the correct positioning and equipment settings. Again, these are skills that radiologists possess but could significantly improve patient throughput and accuracy.
Barriers to AI adoption in digital health
So, what stops us from adopting AI solutions if beneficial use cases like these exist?
Data limitations
One barrier standing in the way of adoption is the lack of data access. We need data to train AI, but healthcare data is hard to access, and de-identifying someone’s information is not enough to call it completely anonymous. We must address these ethical and privacy concerns before widely adopting AI in healthcare.
Another barrier is the lack of standardisation. Healthcare data is often unstructured and varies from one institution to the next, different labs use different measurement equipment, which have different standard ranges. These make it difficult to develop models that can be generalised across different datasets and institutions. Interoperable healthcare solutions would go a long way to help address this problem.
Compliance with regulations
While AI holds great potential for transforming healthcare and improving patient outcomes, several regulatory hurdles must be addressed before it can be widely adopted. Healthcare data regulations demand strict compliance, making it difficult to introduce new technologies into the market, including AI. Again, this goes back to data privacy. Many of these regulations are about safeguarding patient data and ensuring that it can only be viewed by, and shared with, certain people.
Another challenge is the review and approval processes that healthcare technology must undergo before being released to the public. A new AI solution would need validation through clinical studies, which lengthens the time to market for such products. We need a solution that enables healthcare technology providers to bring their products to market easily without compromising patient safety.
Fluffy Spider is your consultant in AI and digital health
We help organisations move toward a future of connected digital healthcare, making existing systems interoperable and modernising infrastructure to unlock the potential of new technologies.
We can help you identify the relevant opportunities to incorporate modern web services and standards for health information exchange, such as HL7 and FHIR fast healthcare interoperability resources, enabling systems to interoperate with other modern health information exchange technologies from the medical software industry and those already implemented by large healthcare providers such as Government health departments.
Visit our Healthcare Integration Services page to learn more about our capabilities and solutions.
