AI tools for automating malaria detection, improving stroke care and predicting suicide risk are among the technologies Partners believes have the greatest potential to impact healthcare in the next year.
Boston-based Partners HealthCare has selected 12 emerging artificial intelligence technologies that it believes have the greatest potential to impact healthcare in the next year, according to a news release.
This is the fifth time Partners has revealed such a list, which it has dubbed the “Disruptive Dozen.”
The technologies that make the cut are chosen by Partners researchers, thought leaders and clinicians. Nominated technologies are narrowed down, and a vote determines a consensus on the final dozen. These are announced at the World Medical Innovation Forum, an event hosted by Partners.
Want to publish your own articles on DistilINFO Publications?
Send us an email, we will get in touch with you.
Here are the AI technologies that were chosen this year.
- Rethinking medical imaging. Researchers see artificial intelligence changing mammography so it becomes a more targeted tool for assessing risk. They also envision it altering ultrasounds to help with disease detection and rapid acquisition of clinical-grade images.
- Better predicting suicide risk. AI can help pinpoint patients at risk of suicide (based on EHR data) and can examine social media content to look for early warning signs. Partners notes that these technologies aren’t cleared for clinical use.
- Streamlining diagnosis. Clinicians can leverage AI to aid their diagnostic and treatment decisions. AI can be used to identify abnormal findings in medical imaging or screen pathology cases based on the presence of urgent findings.
- Automating malaria detection. Deep learning tools can help automate malaria diagnosis. In addition to offering more timely diagnoses, such solutions could enable better monitoring of treatment efficacy.
- Real-time monitoring and analysis of brain health. Automating the analysis of EEGs and other waveforms can allow clinicians to quickly detect electrical abnormalities. Additionally, deep learning algorithms can help automatically detect seizures in critically ill patients.
- AI for eye health and disease. Last year, the FDA approved an AI-based system for detecting diabetic retinopathy. According to Partners, tools like this could be useful in low-resource settings where patients may not have great access to ophthalmologic care.
- FHIR. Fast Healthcare Interoperability Resources, or FHIR, is a new data standard for sharing health information. It could improve patient-centered care and give patients better access to their own medical information.
- Decreasing the healthcare administration burden. AI can be applied to automate repetitive administrative functions such as medical coding and billing. Such technologies can reduce the complexity of the tasks and the number of mistakes.
- Changing stroke care. For patients who have ischemic strokes, AI can help determine whether there’s bleeding in the brain. Such algorithms can also review an individual’s head CT scan to identify a cerebral hemorrhage.
- Finding signs of intimate partner violence. Researchers are developing artificial intelligence-enabled solutions that can alert providers if a patient’s injuries likely stem from intimate partner violence. One potential application of these solutions is to better flag abuse-related cases in the ER.
- Voice-first technologies. Healthcare companies continue to develop applications that can run on consumer-grade technology platforms and comply with HIPAA. Voice assistants are being explored for reducing doctors’ documentation burden as well.
- Bridging the gaps in mental healthcare. In discussing how AI can help mental health patients, Partners highlights an app under development for patients with drug addiction and co-concurring mental illness. The app provides individuals with a virtual form of integrated group therapy, which teaches skills for managing recovery and preventing relapse.
Date: April 17, 2019
Source: MedCityNews
