Source: forbes.com.
Data science and artificial intelligence are no longer buzz words in the biomedical research community. Physicians and other caregivers are increasingly being encouraged by hospitals and health insurance companies to utilize low-resolution dense biometric data captured using wearable medical devices. However classical healthcare heavily relies on high accuracy sparse datasets, i.e. patients are expected to get a thorough medical checkup once in a while, as opposed to continuous monitoring of a handful of vital parameters. The most significant impact of data science will be in helping physicians extract clinically relevant information from such dense low-quality data sets.
Radiologists and cardiologists are increasingly relying on automated high dimensional image processing algorithms to detect the likelihood of coronary artery disease from non-contrast chest CT scans. Similarly, radiologists and pulmonologists are using similar artificial intelligence based technology to identify clinically relevant structural and functional parameters of the lungs from chest CT scans. Understanding the basics of artificial intelligence will empower physicians to go beyond using these tools as black-boxes and deliver maximum impact for care pathways.
In this article, we have listed five such reasons why physicians and caregivers should learn about emerging technology such as data science and artificial intelligence
1. Diagnose using large volumes of data generated from continuous monitoring
With the advent of wearable medical device companies such as CloudDX, Snap40 and QuasaR clinicians can now look at continuous daily biometric data collected over months. Both primary and advanced data science techniques can be used to derive medically relevant outcomes from these dense data. Basic descriptive statistical results like the average resting heart rate could give you a quick understanding of the overall cardiac health of the patient. More advanced indicators such as stress index or LF/HF ratio of RR distance could be used to predict chances of heart arrhythmias more accurately. Data science will allow physicians to analyze these data sets both at local (days or weeks) and global (months or years) timescales, using a combination of both early warning scores and visual inspection of the data.
2. Diagnose using multiparameter data
The most significant insight in health care is often obtained by combining multiple data sources. For example, combining heart rate and heart rate variability can be used to compute overall stress. Respiratory conditions such as COPD and asthma conditions could be triggered by both internal factors, as well as environmental factors such as pollution. Companies like Propeller is combining patient’s respiratory health data from collected using the Propeller spirometer with Propeller Air an open API that uses data from environmental sources to predict how asthma may be affected by local environmental conditions. Learning data science techniques such as data fusion can help physicians understand how data Cis merged in these systems, and therefore diagnose patients more efficiently.
In the case of geriatric emergency care, a quick analysis of the cause of fall can ensure that the emergency physician can deliver the best care pathways. Starkey Hearing Technologies’ new Livio AI hearing aids can already do fall detection using motion sensors built into hearing aids. Given that it can also measure biometric parameters like heart rate, it’s advanced AI engine should one day also tell the caregiver the exact reason of fall, i.e., differentiate between slippage and fall from a fall due to a heart attack. Understanding the underlying data science processes will help physicians design better care pathways for these novel devices.
3. Diagnose using data visualization
Radiologists analyze high dimensional medical images such as CT and MRI scans, to aid other specialists such as cardiologists and pulmonologists to deliver critical care. Radiologists are already using machine learning based software tools which automatically color codes the different features of an internal organ. Learning data science will help radiologists understand the strengths and limitations of these software, helping them to deliver even better diagnostic outcomes.
Some of these tools include Philips’ echocardiography which uses an AI called HeartModelᴬ⋅ᴵ⋅ to additionally build a 3D model of the patient’s heart from echocardiography images. Arterys’ AI-powered Cardiac MR Suite is FDA 510(k) approved and allows cardiologists to view the patient’s heart in 4D, by color coding the blood flow in the heart in real time from magnetic resonance imaging (MRI) images.
4. Understand AI workflow
With the advent of AI physicians and other caregivers will soon come across multiple health predictors such as early warning scores, that were designed using deep learning. For example, Cardiogram’s DeepHeart that works with Apple Watches is a semi-supervised AI learning for cardiovascular risk prediction. Understanding how these machine learning algorithms were designed and therefore their limitations will help caregivers to rely on these early warning scores just the right amount.
5. Understand the statistical significance of clinical studies
As a part of continuing, medical education clinicians are always learning about the latest and most exciting case studies and clinical trials in their fields of expertise. However often some of these results may not be reproducible due to lack of statistical significance of the patient population size on which they were carried out. Learning data science can help clinicians evaluate the relevance of such studies and choose which ones should be incorporated into their own practice. Learning data science will also be extremely useful in the era of personalized medicines, where clinicians will be not only be prescribing medication but will also point out the chances of success based on the patient’s genetic makeup.