Projects

CollaboRhythm

Redefining the doctor-patient relationship - John Moore MD

 

(Special thanks to Luis Blackaller, Nicole Prowell, and Max Wagenblass for their hard work in producing this video and to Jhonatan Rotberg and the Next Billion Network for their support.)

 

The doctor-patient relationship is deteriorating. And today’s information technology solutions are exacerbating the problem. They perpetuate paternalistic decision-making and episodic care, and they fail to assist doctors in making persuasive arguments to their patients.

 

CollaboRhythm is a technological framework that encourages new paradigms in doctor-patient interaction to improve health outcomes and the patient experience. It uses ubiquitous connectivity, collaborative decision-making, and compelling interfaces and visualizations to educate patients, improve treatment adherence, and deliver care at any point in time or space with seamless transitions.

 

The foundation of CollaboRhythm is a speech- and touch-controlled collaborative interface for the office where doctor and patient make shared decisions. Patients can actively engage with their data, so they can take action in their lives with doctors serving as coaches rather than commanders.

 

Patients own their data in CollaboRhythm: everything they see in the doctor’s office is available at home, or when they visit another doctor, or change jobs, or move across the world. Just as importantly, patients can contribute data of their own, things that doctors fail to see in the face of too many lab tests: data and perceptions about social support, diet, alternative therapies, and their effect on the patient’s quality of life. (Patients and physicians disagree on the reason for an office visit nearly 50% of the time.)

 

Although the hub of CollaboRhythm is located in the doctor’s office, the system’s goal is to connect with the patient from any place at any time. However, it isn’t a tele-presence system – which beams the doctor into the patient’s life – but a system for tele-collaboration. Patients need support from their health-care providers, and doctors need to encourage patients to actively track their performance in collaboration with providers. No more letting patients fall through the cracks between visits every six months.

 

Working together, CollaboRhythm’s components bring the promise of ubiquitous connectivity to primary care. For example, in the near future, the doctor may be able to push out medication reminders to the patient’s bathroom mirror or television. The patient can interact with an intelligent conversational agent before doctor visits to prepare. The doctor can send the patients visualizations of their progress in fighting disease, providing detail on a microscopic level in a form that’s understandable and actionable by the patient.

 

Patients of the future will know more about their health than their doctors. They have to. By making patients active, informed participants in their own care, we believe we can reduce health care costs, increase quality, and improve health outcomes.

 

I'm Listening

Giving patients the chance to speak - John Moore MD

Increasing understanding of how to categorize patient symptoms for efficient diagnosis has led to structured patient interviews and diagnostic flowcharts that can provide diagnostic accuracy and save valuable physician time. But the rigidity of predefined questions and controlled vocabulary for answers can leave patients feeling over-constrained, as if the doctor (or computer system) is not really attending to them. I’m Listening is a system for automatically conducting patient pre-visit interviews. It does not replace a human doctor, but can be used before an office visit to prepare the patient, deliver educational materials, triage care, and preorder appropriate tests, making better use of both doctor and patient time. It uses an on-screen avatar and natural language processing to (partially) understand the patient's response. Key is a commonsense reasoning system that lets patients express themselves in unconstrained natural language, even using metaphor, and that maps the language to medically relevant categories.

 

The avatar animation framework used in I'm listening is powered by Oddcast.com. The commonsense reasoning is powered by Open Mind Commons and ConceptNet from the Software Agents group at the MIT Media Lab.

Collective Discovery

Discovering cures in 'everyday experiments' - Ian Eslick - https://www.lamsight.org

A fundamental challenge faced by the medical establishment, from general practice to biomedical research, is the lack of detailed data professionals have about the true context of a patient's life. The choices we make about diet, environment, over the counter medications, alternative therapies, and potential off-label prescriptions constitute a massive, ongoing 'everyday experiment'. This unrecorded dataset is a powerful yet untapped resource.

 

In the case of rare diseases, this failure is particularly poignant as it delays both the diagnosis of a rare condition and limits the information available to form good hypotheses in clinical research.

 

Collective Discovery aims to leverage the intuition and insights of patient communities en-mass to capture and mine information about everyday experiences; this enables communities to "think with data" and collaborate in the formation and support of novel hypotheses. This new mode of community discourse will lead to better decision making, stronger self-advocacy, identification of novel therapies, and inspiration of better hypotheses in traditional research, accelerating the search for treatments.

 

The unique characteristic of Collective Discovery is the use of knowledge representation and natural language processing to compensate for both methodological errors and self-reporting bias through mediation of communal hypotheses generation. This model is being validated in a real world context as part of a partnership with the LAM Treatment Alliance and the greater LAM community.

 

HealthMap

Automated disease outbreak tracking and visualization - Clark Freifeld - http://www.healthmap.org

HealthMap is a multi-lingual, real-time disease outbreak tracking and visualization system. Launched in 2006, the Web site collects over 300 reports per day in English, Spanish, French, Russian, and Chinese, from both general news media and public health sources around the world. Updated hourly, the system filters these reports to determine relevance, disease, location and duplication clustering by means of a series of custom-designed automated text-processing algorithms. Relevant reports are then aggregated and displayed on a freely available dashboard where users can tailor the view according to date, disease, location, and source. HealthMap provides an overview of real-time information on emerging infectious diseases, and has particular interest for public health officials and international travelers.