A Snapshot of How Medicine is Harnessing the Power of AI Today
We talk to our phones. We rely on our customized music playlists. But how do we feel about the fact that so many aspects of our lives are touched by artificial intelligence (AI)? A Pew Research Center survey released earlier this year finds that one-half to two-thirds of Americans understand that the power of AI drives such necessities as wearable fitness trackers and customer service chatbots. However, only 15% say they are more excited than concerned about the increasing use of AI in daily life.
Artificial intelligence skepticism fuels another statistic: 60% of US adults say they would feel uncomfortable if their health care provider relied on AI to do things like diagnose disease and recommend treatments. Yet those are some of the very ways medicine is putting AI to work. In fact, patients who fear artificial intelligence might not even be aware when it’s being used in their care.
How is AI Being Used in Medicine Today?
Artificial intelligence is gaining relevance in diagnostics, personalized medicine, drug development and telehealth. Broadening applications and AI adoption rates mean physician staffing will play a key role in ensuring that practices have the skillsets to continue to apply these technologies in the ways listed below, and beyond.
Diagnostic Precision – Radiology and Imaging
Radiology’s heavy reliance on the interpretation of medical images such as X-rays, CT scans, and MRIs has made it a prime candidate for the types of advances AI offers.
Example: A remarkable application of AI in imaging comes from PathAI, an AI-powered diagnostic platform. Their algorithms assist pathologists in analyzing histopathology slides, detecting cancer, and providing quantifiable metrics for tissue analysis. Such innovations enhance diagnostic accuracy, potentially reducing misdiagnoses.
Personalized Treatment
AI is playing an instrumental role in tailoring healthcare interventions to an individual’s unique characteristics and making personalized treatment plans a reality. AI systems can forecast disease progression, enabling early interventions and tailored treatment plans.
Genomic Medicine
Example: Companies like Tempus Labs use machine learning to identify patterns in cancer patients’ genetic data and clinical information. Its AI-driven platforms analyze genetic data and provide oncologists with actionable insights. These insights help physicians select targeted therapies that are more likely to be effective, sparing patients from potentially ineffective treatments and their associated side effects.
Predictive Analytics
Example: Propeller Health, in partnership with Novartis, developed an AI-driven platform to monitor patients with chronic obstructive pulmonary disease (COPD). The system collects data from inhaler sensors, analyzing usage patterns and environmental factors. This information helps healthcare providers personalize treatment plans and predict exacerbations, ultimately improving patient outcomes.
Drug Discovery and Development
The drug discovery process is notorious for its protracted timelines and high costs. AI-driven platforms are changing the game by expediting drug discovery and streamlining development processes. Moreover, AI’s proficiency in sifting through large datasets enables the discovery of novel biomarkers, critical indicators of disease presence, progression, or treatment response.
Drug Repurposing
Example: BenevolentAI, an AI-powered drug discovery platform, employs deep learning to analyze vast datasets of scientific literature and biological information. BenevolentAI’s platform identified baricitinib, an FDA-approved rheumatoid arthritis drug, as a potential treatment for COVID-19. This discovery, backed by clinical trials, showcases AI’s ability to accelerate drug repurposing efforts, especially in times of urgent medical need.
Biomarker Discovery
Example: Oncology provides a compelling example for biomarker discovery. Tempus Labs uses AI to analyze clinical and molecular data from cancer patients, identifying unique biomarkers that guide targeted therapy selection. By identifying specific mutations or protein expressions, AI helps oncologists make informed decisions about the most suitable treatments for their patients.
Telemedicine and Remote Monitoring
The emergence of telemedicine and remote patient monitoring has been accelerated by AI technologies, especially in the wake of the COVID-19 pandemic.
Telemedicine Platforms
Example: Telemedicine platforms like Teladoc Health leverage AI-driven virtual visits, connecting patients with healthcare providers remotely. These platforms use natural language processing (NLP) to understand patients’ symptoms and provide medical advice. Teladoc’s AI-powered chatbots assist with initial triage, offering guidance to patients while minimizing the burden on healthcare facilities.
Remote Patient Monitoring
Example: AI-powered devices, such as wearables and home-based sensors, allow continuous data collection and analysis. The Apple Watch Series 4 and later versions incorporate an ECG feature and AI algorithms that can detect atrial fibrillation (AFib). By alerting users to potential heart rhythm irregularities, these devices empower individuals to seek timely medical attention.
Challenges and Ethical Considerations
No review of medical applications of the power of AI would be complete without a discussion of the nexus between AI and medical ethics. Considerations fall into four areas.
Data Privacy and Security
The use of AI in healthcare generates vast amounts of sensitive patient data. Ensuring robust data privacy and security measures is imperative to protect patient confidentiality and prevent unauthorized access or breaches.
Bias and Fairness
AI algorithms can inadvertently perpetuate biases present in the data they are trained on. Efforts are required to address algorithmic bias and ensure that AI applications in healthcare provide equitable care to all patients.
Regulatory Frameworks
The rapid pace of AI innovation poses challenges for regulatory bodies in keeping up with evolving technologies. Establishing clear and adaptable regulatory frameworks is essential to ensure safety, efficacy, and ethical standards.
Physician-Computer Interaction
Defining the roles and responsibilities of healthcare providers in an AI-driven landscape is crucial. Physicians must collaborate effectively with AI systems to maximize patient benefits while maintaining a human touch in care delivery.
The Journey Ahead
AI is at the forefront of contemporary healthcare innovation. Real-world examples demonstrate that it’s not just a theoretical concept but a practical reality. In June, the director of Cambridge University’s Centre for AI in Medicine published an op-ed in The Guardian stating that “Artificial intelligence can’t replace human professionals but it could transform the way they treat diseases such as cancer, and save lives”. This points to the ongoing reliance on physician staffing and the skillset training needed to leverage current and emerging AI applications.
The journey is still unfolding and will continue to generate ethical considerations. But the future of healthcare is undeniably intertwined with AI’s transformative potential to improve patient outcomes, reduce healthcare costs and enhance access to medical services. As its applications continue to spread, the power of AI will be used in the care of more and more patients – whether they’re aware of it or not.