Why AI as an ‘intelligent assistant’ has the potential to improve care
The technology is beginning to mature in ways that improve clinician bandwidth and can address health equity issues.
While the future of healthcare is being shaped by advances in artificial intelligence, there are many impacts that this technology is having on patient care when paired with clinicians in the right way.
AI is tackling an overwhelming challenge that has made it difficult for our healthcare system to maximize patient outcomes – processing the oceans of available data in short amounts of time. Many clinicians are actively considering leaving the healthcare industry, with a survey of 13,000 radiologists finding that half reported burnout; 60 percent of those citing excessive bureaucratic tasks as the leading cause.
AI can help address some of these challenges. It can absorb a lifetime of medical history in seconds, enabling a time-constrained physician to leverage it to glean and optimize clinical insights much faster than ever before. This is helping to address long-standing inequities in the healthcare system by giving underserved patients more accurate, same-day diagnoses while their conditions are in earlier, more treatable stages.
While AI can provide the computational horsepower, clinicians remain indispensable in applying the wisdom and judgment earned from their years of training and experience. AI can't hold a sick patient's hand or calm a worried parent's nerves. Clinicians specialize in that kind of empathy and effectiveness.
AI as the intelligent assistant
Once skeptical, doctors now welcome AI as an intelligent assistant. The technology turbocharges expertise rather than competes against it. AI speeds up diagnostics, spots early warnings signs and analyzes data to reveal the most promising treatments. It automates mundane, manual tasks.
Connecting longitudinal data and layering it with analytics and AI is helping clinicians get the whole view of a patient across a care pathway so they can make personalized decisions more accurately at the point of care, using up-to-date information from routine monitors, medical record data, laboratory results and the fusion of multiple imaging modalities.
Just as different body parts work together harmoniously, AI’s deep learning networks integrate various facets of machine learning to create a comprehensive network of information. This layered approach produces a detailed picture of vast volumes of data, analogous to the intricate workings of the human anatomy. AI has vast potential to improve diagnostic accuracy and treatment outcomes across a variety of modalities.
Improving imaging diagnostics
Currently, AI is revolutionizing medical imaging across all forms of diagnostic imaging. From addressing image quality challenges related to noise and patient movement during scans to optimizing workflow inefficiencies resulting from increasing patient backlogs, staff shortfalls and clinical expertise shortages, AI is redefining what is possible.
AI-powered imaging technologies enable physicians to see a detailed snapshot of vast volumes of data, layering images together to fill in gaps that are less defined because of patient movement, breathing or coughing, which can create missing information or blurred pictures. AI also enables radiologists to more comprehensively analyze medical images to detect subtle anomalies that might be impossible to discern with the human eye.
AI enables providers to find information more quickly, helping them get to the right diagnosis faster. This benefits the patient by personalizing their care, and it assists the enterprise by finding life-saving treatments more quickly at a lower cost. For overtaxed physicians, AI is helping them rediscover their love of medicine by giving them more time to listen to patients and fine-tune treatment decisions.
Radiology is a crucial component of modern healthcare, enabling clinicians to visualize and diagnose a wide range of medical conditions. Without the right technology to support fair and efficient workload distribution, radiology teams can be hampered by lost time and productivity, anxiety, reactive staffing, unbalanced workload distribution and cherry-picking, all of which can have a negative impact on overall practice performance and radiologist morale.
Radiology departments are using AI-based technologies in X-rays, magnetic resonance imaging, computed tomography and molecular imaging to consistently produce high-quality images and improve the patient experience during the exam.
In cardiac imaging, AI has facilitated the acquisition of magnetic resonance-based images significantly faster than conventional methods, enabling cardiac imaging within a single heartbeat. The ability to reduce these cardiac scan times also helps enhance productivity in radiology departments, streamlining workflows, alleviating backlogs and reducing the burden on staff. New magnetic resonance technology also enables cardiac patients with arrhythmias and breath-holding challenges to complete exams more quickly and comfortably.
Addressing disparities in breast cancer
In the United States, a person’s ZIP code can be a better predictor of life expectancy than genetics, according to the Robert Wood Johnson Foundation. One way that these disparities in breast cancer care can be addressed is by using AI.
Women in Black and LatinX populations are diagnosed with breast cancer at more advanced stages, when the treatment options can be limited and costly, often leading to poor prognoses. According to a recent study, women of color also often experience delayed time to treatment. Black women have a 4 percent lower incidence rate of breast cancer than white women but a 40 percent higher breast cancer death rate. In addition, because of childcare responsibilities and economic, logistical and transportation challenges, women from underrepresented communities may have more difficulty making scheduled medical visits.
Mammograms are important standard health screenings that help detect breast cancer as early as possible, when treatment is typically most effective. Radiologists have an opportunity to improve access to more equitable breast cancer care. Many breast imaging centers now use computer-aided detection (CAD), an AI technology, alongside traditional mammograms. CAD helps radiologists by speeding up the reading process and providing additional clinical insights.
Studies show that mammogram screenings miss about one out of every eight cases of breast cancer, and CAD helps address this gap. While regular mammograms remain the gold standard for early detection, AI algorithms can help identify subtle signs that might be overlooked by human observers. In one clinical trial, adding an AI algorithm to mammography screenings spotted 20 percent more breast cancer cases.
These kinds of technologies give healthcare teams important information to properly diagnose cancers and determine treatment options. AI can help clinicians provide more clinical insights during a patient’s first appointment, so they leave with answers the same day. It can both help expedite the treatment path and eliminate the needless anxiety of patients who don’t require treatment, helping them learn more quickly that they need not return for additional follow-up scans.
While there is more to be done to improve health equity and access across the system, breast cancer screening is one way AI can strengthen health equity by enabling clinicians work smarter and faster, reach more patients and utilize data to correct biases in clinical decision-making.
Enabling precision care
The U.S. Food and Drug Administration has authorized more than 850 AI-enabled medical devices, demonstrating how the power of AI can be harnessed to create intuitive, intelligent tools to help time-constrained clinicians unlock the full potential of their expertise, spend more time with patients and deliver personalized care to a wider swath of people than ever before.
The last few years have turned healthcare on its head. Staff shortages, pandemic burnout, health inequality and runaway costs remain major issues. AI is enabling providers to take a holistic approach to precision care that integrates advanced medical imaging, machine learning and molecular diagnostics to tailor treatment, ongoing monitoring and management for optimal patient outcomes.
AI can help lighten the load for overburdened staff trying to square the circle. With the right strategic deployment of AI to assist experienced physicians, the technology has the potential to turn reams of clinical data into meaningful actions to improve both clinicians’ and patients’ lives.
Abu Mirza, MS, is the general manager and global senior vice president of digital products and engineering at GE HealthCare. He has overseen the development of multiple healthcare-enhancing AI platforms.