Global malaria eradication is the consensus long-term goal for malaria, which will be achieved by eliminating malaria from countries and regions. Strategic planning for malaria eradication and successful implementation of elimination campaigns can be improved by gathering, integrating and utilizing big and small data.
What is Personal Big Data? How is it captured? How is it measured? What does it mean? What's possible now? What's possible in the future? Why should DARPA care? By harnessing the power of Personal Big Data, we unlock the ability to:
Behavioral assessment and measurement today are typically invasive and human intensive (for both patient and clinician). Moreover, by their nature, they focus on retrospective analysis by the patient (or the patient’s loved ones) about emotionally charged situations — a process rife with biases, not repeatable, and expensive. We examine all the data in the &lquot;white space&rquot; between interactions with the healthcare system (social media data, wearables, activities, nutrition, mood, etc.), and have shown quantified signals relevant to mental health that can be extracted from them. These methods to gather and analyze disparate data unobtrusively and in real time enable a range of new scientific questions, diagnostic capabilities, assessment of novel treatments, and quantified key performance measures for behavioral health. These techniques hold special promise for suicide risk, given the dearth of unbiased accounts of a person’s behavior leading up to a suicide attempt. We are beginning to see the promise of using these disparate data for revolution in mental health.
Forecasting the onset of epidemics is important for hospital emergency departments and related services. We are exploring the use of social media for providing early warning of respiratory disease epidemics by monitoring tweets originating in our hospital’s catchment community. By applying machine learning and time series analysis to georeferenced tweets, it may be possible to provide lead-time sufficient for hospitals to prepare for seasonal events such as influenza and surprise events such as enterovirus. We describe recent and ongoing work in Cincinnati.
In the future, intelligent systems in homes, offices, public spaces, and embedded on and in humans will generate too much health-related data to be processed by humans. Some of this data will be used by individuals (patients), but much of it will end up in analytical systems and then be used by advertisers, health care providers, the government, and even data criminals. Since cutting edge applications typically pay attention to utility before they address privacy and security, data breaches will likely take place on a massive scale, sometimes negatively affecting physical and mental health. Many of these problems can be addressed up front by specialists in medical cyber security and privacy. We here present initial thoughts on developing these experts, based on our experience since 2002 leading a successful cyber security education program.
Biomedical researchers cannot adequately leverage large-scale genomic, transcriptomic and proteomic data that is now collected. Current informatics tools do not capitalize on the great advancements made in Machine Learning (ML) that can enable them to generate more, and more rapid, breakthroughs. Big Data technologies can facilitate the complete integration of heterogeneous sets of experimental data to identify key metabolic pathways and drug targets to enable precision medicine.
An open hypothesis remains regarding the potential correlation between genotype and social/behavioral interactions. The latter is an important element in a clinician’s diagnosis of a patient’s mental health, while traditionally being difficult to quantify with any reliability. We are at an opportune time to bridge the traditional genotypic/phenotypic gap with social media’s ubiquity, the surge in wearables, and gene sequencing becoming more mainstream and affordable. This wide swath of data will allow for the emergence of quantified behavioral profiles spanning genotypes and phenotypes. This might provide new fundamental classes for scientific examination, with the hope of explaining some of the widely observed variance and accuracy of traditional modeling approaches that use only one facet. The integrated analysis of these sets of heterogeneous data covering the spectrum of human experience will lead to new insight, better predictive models, more accurate clinical diagnoses, and providing the foundation for pragmatic precision medicine — in totality, there is great potential for revolution in both behavioral genetics and mental healthcare.
Understanding users’ self-reported symptoms and descriptions of their state is an important task with many applications including pharmacovigilence and health detection. We will describe methods for doing so in the context of a larger system for monitoring a user’s “state of being.”
Vaccine refusal is an increasingly important public health concern. Public health agencies must understand how rationales for this dangerous behavior vary by group to target their messages accordingly. We propose to detect and categorize these rationales using social media data.
Electronic health records have begun to emerge as a new and valuable data source for mental health research, but clinical records can be a challenge to obtain, and provide only a narrow window into a patient's condition. Social media provides a complementary resource: now used regularly by over a billion people worldwide, social media offers insight into the thoughts, feelings, and behaviors of patients in their daily lives, and a finer-grained longitudinal view of their mental health. In this talk, we discuss our recent efforts working toward the combination of clinical records and social media posts, arguing that this combination provides a heretofore untapped potential for data-driven discovery, predictive modeling, and the development of personalized interventions in mental health.