Mobile Technology for Adaptive Aging Proceedings of a Workshop (2020) / Chapter Skim
Currently Skimming:
5 Using Machine Learning to Forecast and Improve Clinical Outcomes and Healthy Aging Using Sensor Data - Alvin Rajkomar
5 Using Machine Learning to Forecast and Improve Clinical Outcomes and Healthy Aging Using Sensor Data - Alvin Rajkomar
Pages 85-104
The Chapter Skim interface presents what we've algorithmically identified as the most significant single chunk of text within every page in the chapter.
Select key terms on the right to highlight them within pages of the chapter.
From page 85... ...
By analyzing real time locations and speeds of cars, apps can automatically detect traffic and re-route you to your destination to arrive sooner. It seems natural that if a system could collect lifestyle habits of millions of people through ubiquitous sensors, such as those in cell phones, and follow what happened to them -- whether they developed diseases or disability -- then it could direct people how to live better to reduce the risk of diabetes or to inform how we can promote an aging parent to live safely at home, effectively re-routing their life to a longer, independent life.
|
From page 86... ...
The question is how can researchers thoughtfully apply best practices in machine learning (ML) and clinical research as they use data to forecast progression of aging and clinical trajectories and identify ways to improve patient outcomes.
|
From page 87... ...
Input Data Types of Sensors The ubiquity of low-cost, miniature, and novel sensors allows for the collection of data that were previously too expensive or inconvenient to collect at scale. There is inconsistent terminology to categorize these sensors; some authors use the term "wearable" to emphasize the form factor and ease of collection, others use mobile health to highlight connection to a sensor carried in a mobile phone.
|
From page 88... ...
heart rate, heart cardiovascular rate variability health Electrical activity Electrodes Electrocardiograms Heart rhythms, (EKG) , sleep states, electroencephalogr emotional state ams (EEG)
|
From page 89... ...
USING MACHINE LEARNING TO FORECAST AND IMPROVE CLINICAL OUTCOMES 89 Chemical analytes on skin Potentiometric and Glucose, lactate, amperometric sodium sensors measurements in sweat Temperature Thermistor Body temperature Elevated risk of infection (Abbasi, 2017) Location Global position Movement Location entropy satellite to indicate measurements depression Measurements from ambient sources Video Cameras Pixels Activity classification in the home, vital signs (Prakash and Tucker, 2018)
|
From page 90... ...
. Active versus Passive Data Collection Sensors commonly collect data passively, meaning a person is not a ctively engaging with the sensor as they go about their day (Sim, 2019)
|
From page 91... ...
A model could be used to detect if the recorded electrical pattern is consistent with atrial fibrillation. If the user is prompted to indicate their emotional state of anxiety at the time of an elevated heart rate, a model could use the same sensor to classify emotional state.
|
From page 92... ...
Typically, a sequence input with a variety of measured Common and types for sensor models can associate this of data points are labels. If the sensor itself measures the outcome, such as heart rate monitor detecting abnormal measured over timepossible atrial fibrillation, thenthis input with a varietyas a detection label.
|
From page 93... ...
Surrogate outcomes are laboratory or sensor measurements that are thought to be correlated with hard endpoints, such as detection of atrial fibrillation, which is strongly associated with stroke. However, it is well known in clinical research that successful prediction of surrogate endpoints is not guaranteed to lead to better hard outcomes, and in many cases, it can lead to worse or unintended consequences (Mandl and Manrai, 2019; Prasad et al., 2015; Weintraub, Lüscher, and Pocock, 2015)
|
From page 94... ...
Applying ML to clinical data gathered by sensors requires consented, discrete, measurable, and reproducible labels that may not always be possible or easy to obtain in widespread populations. Hard endpoints like cognitive decline or death may take decades to occur, and clinical outcomes, like diagnosis, require regular clinical assessments that are not uniformly rigorous or applied across a population.
|
From page 95... ...
. The results of the completed trials shocked the medical community: treating the abnormal rhythms was associated with increased mortality, forcing a rapid change in the standard of care and highlighting the dangers of using surrogate measures rather than clinical outcomes to assess the utility and safety of interventions (Pfeffer and McMurray, 2016)
|
From page 96... ...
. In the early 2000s, these physiological effects together with observational and clinical trial data which suggested that patients with higher blood sugar had worse outcomes led to widespread adoption of tight blood sugar control in intensive care units.
|
From page 97... ...
For example, a newly physically active individual may develop a slower heart rate due to improved cardio ascular v health, or the same finding may reflect that he is newly employed and now has health insurance to pay for a prescribed beta-blocker for migraine prevention. Traditional clinical studies have protocols to try to discern plausible causal factors that account for changes in outcomes.
|
From page 98... ...
What Are the Effects of Healthcare Disparities in Data and Machine Learning? Collecting and using consented data from groups that have experienced discrimination or human and structural biases brings the attendant risk of worsening healthcare disparities (Rajkomar et al., 2018)
|
From page 99... ...
Privileged bias refers to the phenomenon of aging populations not having a voice in the types of technologies being developed that they can use or afford. As a result of privileged bias, systems may not be designed to solve the problems facing aging populations, such as limited internet connectivity or e-literacy that limits adoption of even interested elderly patients (Van Winkle, Carpenter, and Moscucci, 2017)
|
From page 100... ...
Although using hard outcomes in large-scale studies is preferable, thoughtfully using surrogate outcomes in smaller-scale but high-risk cohorts can accelerate knowledge generation and direct limited resources to run larger, expensive trials with hard outcomes. The rapid development of new wearables means that the ability to rapidly evaluate sensors for clinical promise is increasingly important if researchers are to design studies that take advantage of new technologies (Kim et al., 2019)
|
From page 101... ...
. Wearable digital thermometer improves fever detection.
|
From page 102... ...
. Machine learning for comprehensive forecasting of Alzheimer's Disease progres sion.
|
From page 103... ...
. Personal sensing: Understanding mental health using ubiquitous sensors and machine learning.
|
From page 104... ...
. Accuracy of wrist-worn heart rate monitors.
|
Key Terms
This material may be derived from roughly machine-read images, and so is provided only to facilitate research.
More
information on Chapter Skim is available.