Healthcare organizations are going through a major shift on the heels of a digital revolution. Care providers must now manage large amounts of structured and unstructured patient data – and they need help keeping it organized. Applications of artificial intelligence (AI), such as machine learning, can provide this assistance, and healthcare organizations are beginning to invest in it.
Funding targets for AI healthcare projects
The number of AI-powered healthcare startups has grown significantly since early 2017, and the amount of funding requested for healthcare AI projects is expected to reach $6 billion by 2021. The sheer size and complexity of health care processes has created a prime market for these new technologies.
Here are some of the targeted functions for these AI-driven startups:
Medical imaging and diagnostics
- Using a technique called computer vision, machines can achieve a higher level of understanding about patient conditions. They can also tag or monitor patients at a higher rate of accuracy than human monitoring.
- This is particularly helpful with radiology files and cancer-related diagnoses. Some indicators might be difficult for humans to identify, but prove relatively easy for machines.
- AI-driven technology can hasten the process of identifying and synthesizing new drugs for treatment. Machines can more effectively understand the relationship between a patient’s molecular makeup and drug reactions making it easier prescribe optimal drug formulations for a specific patient’s needs.
- Consumer health and fitness tracking devices, like those from FitBit or Apple, can be part of a holistic patient monitoring system. In the past, these devices were used primarily to identify when something went wrong, but new advances allow doctors to be more predictive.
- These smart wearable devices are creating a tremendous increase number of data points physicians can use to track patient activity, sleep patterns, and the even serve as a proactive health monitor by identifying potential heart arrhythmias.
- Large organizations increasingly rely on these devices to help with overall lifestyle management for their employees. By encouraging healthier living, they hope to use the data to keep people out of the hospital (and save on medical costs).
Questions linger on differentiation
AI won’t solve the world’s healthcare issues by itself. Healthcare organizations need to think about how they can utilize AI technology to improve care experiences and achieve better outcomes. It starts by reassessing how they reach and engage their patients.
Here are some points to follow when assessing AI-driven potential in a health care setting.
Find the space where you have the largest amounts of data.
- In healthcare, this will likely involve patient health records, either electronic (EHR) or hard copy. Value-based service models can benefit from AI’s ability to mine health records and identify at risk patients and reduce unnecessary tests.
- Leveraging AI-driven platforms, providers can reduce the use of the archaic fax machine as a method for storing and sharing patient data, converting unstructured documents into machine-readable files that can added to the digital health record.
Consider an expansion of telemedicine.
- Chatbots, telecommunications, virtual visits and electronic home sensors are some of the technologies that increasingly bridge the physical gap between physician and patient.
- Many institutions are using lessons from the world of digital marketing with machine learning based models to improve patient engagement. Through personalized and analytics-based messaging, institutions are helping patients achieve desired outcomes through like completing health assessments, scheduling flu shots, and managing health conditions.
- While an AI platform can’t prescribe medication itself, it can access previous visit data (and other data from similar situations) and offer drug recommendations to the physician. Expect more use cases to follow that augment physician’s ability to spot trends and atypical risk factors for patients as well as recommending next best actions for additional testing or treatment.
Always keep security and transparency in mind.
- While the healthcare industry itself is heavily regulated, AI-related development is surprisingly light on regulatory oversight. There hasn’t been much guidance from governing bodies on regulated use of these technologies, so long as the data itself is protected and managed appropriately.
- However, all standard regulatory and legal frameworks must be followed when managing patient data, to avoid any security and transparency violations.
The current outlook for AI and healthcare
AI is still a relatively new concept for healthcare organizations. Yet many of them are now realizing the potential for AI-powered products to help lower costs and improve care. Right now, tech startups are focusing their efforts on a few key areas, with funding provided in part through insurers looking to reduce unnecessary patient trips to the hospital. Hospitals, meanwhile, look to AI applications to enhance the overall patient experience at every stage of treatment.
David Berglund serves as senior vice president and artificial intelligence lead on the U.S. Bank Innovation team. He focuses on driving technology innovation and product development, accelerating business growth, building strong team cultures and finding exponential advantages with technology.
©2018 U.S. Bank. Member FDIC.