Using machine learning to enable a future where every building makes us healthier.
Pitch us on your solution
Despite being at the center of healthcare, hospital environments are making us sicker. Studies show that inefficient ward layouts can slow patient recovery, and arduous commutes to/from nursing stations can trigger staff burnout. $200B of new hospital construction is slated for the next decade, and these new environments may continue to unwittingly jeopardize our health without thoughtful design.
Hospital administrators don’t intentionally create these dire environments; but rather, they lack data to prove that facility architecture affects health outcomes. Design is therefore deprioritized because it isn't translated into metrics that decision-makers care about.
We help healthcare organizations quantify the value of design by translating qualitative floor plans into quantitative spatial data, revealing how this data connects to hospital health outcomes, and recommending targeted interventions to optimize day-to-day performance. By starting in healthcare, we can create a building standard that optimizes wellness across urban hubs like community workspaces, universities, and homes.
Film your elevator pitch
What is the problem you are solving?
We are passionate about contributing to the Healthy Cities challenge by creating a wellness standard for buildings at a global scale. According to the Environmental Protection Agency (EPA), the average American spends 87% of his/her time indoors. Thus, if we want to promote healthier cities, we have to plan and promote healthier indoor environments. With hospital facilities being at the center of healthcare and serving as natural community hubs, we would like to start with optimizing their building layouts to prevent the spread of disease.
Our core solution relates to a growing body of research that links the architecture of a hospital to its rate of nosocomial infections (Zimring 2013) and medical errors (Buchanan 1991). As a result of this research, our data set includes quantifying aspects like locations and visibilities of sinks and sanitizers, as well as classifying the types of seating arrangements through the facility.
Our initial focus is serving that 35M patients that visit the 6,210 inpatient facilities in the US every year (source: American Hospital Association), but we are passionate about rewriting the standard for the ~16,500 hospitals worldwide and expanding our services to other types of buildings around the world.
Who are you serving?
While our primary buyer is described below, we are motivated by the opportunity to serve the 35M inpatients and their 40M visitors that frequent US hospitals every year. On the staff side, we hope to do our part to curb both the physician and nurse shortages by championing a better workplace environment that prevents burnout (sources: American Association of Colleges of Nursing, Association of American Medical Colleges, CNN). Over the course of >100 user interviews, we empathize with the physical and cognitive burden that patients, visitors, and staff endure in a hospital setting. These constituents have a voice through staff satisfaction surveys, HCAHPS scores, and patient-family advisory councils, but their feedback is largely left qualitative. By elevating these stakeholders’ voices in the form of quantified outcomes, we can ensure that their experience is fairly represented when evaluating opportunities to improve hospital spaces.
What is your solution?
The three aspects of our first product include translating, revealing, and recommending. By collecting qualitative inputs as simple as floor plans, we are able to analyze the static image and extract quantitative spatial features based on architectural principles. This allows us to provide our customers with new data features that they haven’t historically had. We then employ statistical methods such as natural experiments, difference-in-difference scenarios, and causal machine learning techniques to make sense of the new data, revealing a facility’s existing design gaps and how they are affecting outcomes. We tie this new data to relevant performance metrics that a hospital is already tracking; examples include staff turnover, patient HCAHPS and satisfaction ratings, and patient throughput. Based on the aggregate data across sites, we can recommend a tailored ‘kit of parts’ of evidence-based interventions based on a site’s individualized needs. By providing the recommendations through actionable summaries, we're empowering our partners to drive relevant conversations with decision makers and incorporate design decisions into their renovation plans. By quantifying the value of design, we can help payors and providers to save money, improve operational efficiency, and improve perceptions of care.
We are first piloting our solution with hospital administrators, namely a VP of Facilities or VP of Support Services. As derived from our primary market research, we’ve realized that these professionals not only have control over allocating resources for capital projects but are also incentivized to effectively champion design by improving metrics tied to their personal performance. 70% of US hospitals intend to renew or renovate their facilities in the next three years, and our ideal engagement in their process is during the pre-design process of an upcoming renovation. In the long term, we envision creating a health index in the long term that could span to healthcare payors (i.e. insurance companies), allowing them to track building health within their provider network over time. This index product would also allow for a solution that yields recurring revenue and can span to measure the wellness of other urban floor plans (office spaces, retail, universities, etc.). By focusing on streamlined analytics behind design effectiveness, we can ultimately drive systematic change by incentivizing professionals to design for health at scale.
Select only the most relevant.
Where is your solution team headquartered?Cambridge, Massachusetts, USA
Our solution's stage of development:
Select one of the below:
Describe what makes your solution innovative.
Jim has personally felt the pain of the lack of a data-driven and unbiased approach in how architecture firms value their work during post-occupancy evaluations. Our approach brings a quantitative rigor that hasn't been applied at scale to date.
There are a number of healthcare analytics companies that have validated the business model of creating standards around patient safety or clinical trials, but none are looking at the environmental determinants of health. As an example, Pascal Metrics has established itself as a registered Patient Safety Organization (PSO) with the US government to gain credibility in its risk score of patient falls. Our approach focuses on an important social determinant of health that has often been neglected despite strong links to hospital performance.
Finally, existing certification programs are largely audit-based. As a parallel, Delos has created a wellness standard that has contributed to the company's 800M valuation with investors like the Bill Gates Foundation. However, current certifications aren't quantifying the exact impact of a building's decisions beyond the required threshold.
While we are leveraging business models that are validated by existing industries, we're doubling down on our unique premise that design affects our health. We can differentiate ourselves from existing solutions to help healthcare administrators bring statistical rigor to a usually touchy-feely process, take data creation a step further to connect to outcomes, and create personalized insights based on THEIR floor plans, but with models that are operating at scale.
Describe the core technology that your solution utilizes.
We see the first phase as being a new technology. We are developing the unique process of automatically extracting features from a partner's floor plans and turning layouts into distinct, measurable spatial features. This technology is unique to the market in that it would replace a considerable amount of manual labor that is required to prepare floor plans to extract pre-identified features within a floor plan. Our primary market research validates that many industries beyond healthcare are interested in making use of floor plans and other qualitative design aspects to map to unique insights about their spaces.
We see the second phase of our work as using existing analysis methods in a new application. We use various methods of statistical analysis in order to connect this newly generated spatial data to a hospital's health outcomes and performance. While the methods themselves are not new, using them to connect architecture and health outcomes is what makes us unique. In our earlier pilots, we plan to draw connections with the help of natural experiments and difference-in-difference scenarios. In the long term, we plan to use this newly created spatial data to employ causal machine learning techniques and offer predictive scenarios that could help facilities choose between multiple design decisions based on their potential impacts.
Please select the technologies currently used in your solution:
Why do you expect your solution to address the problem?
We have outlined our theory of change to help drive our efforts to our company's end goal: to inspire systemic change of improving healthcare design. The theory below also includes our long term vision to include multiple aspects of the built environment.
As a single example, we have outlined the specific ways we can help hospital administrators enhance patients' perceptions of care:
Select the key characteristics of the population your solution serves.
In which countries do you currently operate?
In which countries will you be operating within the next year?
How many people are you currently serving with your solution? How many will you be serving in one year? How about in five years?
With the existing proof of concept analyzing six different outpatient sites, we can estimate that we are impacting 500 physicians and their supporting staff members out of 651 of the physicians within the specific health network. Based on the average patient impact of 1,000 patient discharges per year, we anticipate affecting half, or 500, that will visit one of the outpatient centers this year (source: partner website).
Our goal is to secure up to 10 pilots with hospital networks within the next year. With 6,210 inpatient hospitals and an average of 35M inpatients being admitted to US hospitals, each pilot could impact approximately 5,000 patients, or up to 50,000 patients in the next year.
Assuming a milestone of 100 customers over the next five years, we hope to make buildings healthier for up to 500,000 urban citizens that frequent hospitals, office spaces, and other community hubs.
What are your goals within the next year and within the next five years?
Within the next year, we have set milestones to validate our product-market fit in the healthcare sector. In the next five years, we hope to not only help more hospital networks, but also to expand into other aspects of real estate that would impact population health. If we succeed, we can significantly contribute to reducing disease and improving wellness across cities. We plan to prioritize markets for subsequent entry after we understand the impact we can make in healthcare.
What are the barriers that currently exist for you to accomplish your goals for the next year and for the next five years?
1. Next Year: Our current offering still involves manual processes to test the initial interest of our product.
2. Next Year: We are familiarizing ourselves with the requirements of analyzing non-identified health information for healthcare companies.
3. Next Year: We are working to prioritize health outcomes that matter most for healthcare administrators.
4. Next Five Years: We are relying on our first pilots to help determine the indicators to develop our future health index.
How are you planning to overcome these barriers?
The following map to the numbered aspects in the prior question:
1. We are prioritizing certain feature development in order to be able to automate the most labor-intensive steps within the next year.
2. We are setting up tools and technical resources that allow us to comply with the industry's regulations, but would sincerely appreciate guidance from Solve community experts.
3. We would love the support of Solve sponsors and community members to continue our primary market research. To date, our proof of concept has already shed great insights with our first partner.
4. Our interactions with a mock board as participation in the MIT Delta V program will help to strategically work towards a recurring revenue product.
Select an option below:
How many people work on your solution team?
2 Co-founders (full-time staff)
1 Undergraduate Intern (full-time for Summer 2019)
1 Consultant (part-time for Summer 2019)
For how many years have you been working on your solution?
Why are you and your team best-placed to deliver this solution?
Jim Peraino - MIT SMArchS Design & Computation ‘20
While designing and evaluating hospital facilities as a licensed architect, Jim witnessed first-hand how thoughtful design could improve health outcomes. He developed computational tools to analyze and automate design for health outcomes as part of his master’s thesis at Harvard. His expertise in machine learning and data analytics can be applied to deliver high-performance healthcare architecture insights.
Sonal Singh - MIT Sloan MBA ‘19
Sonal’s experience in product management for the consumer health industry brings the UI/UX perspective of creating human-centered experiences. She is passionate about the notion that nuances in environment can drive culture, engagement, and community wellness.
Both Jim and Sonal share personal experiences visiting loved ones in a hospital only to spend hours in an ill-equipped cafeteria. We ate cold cheeseburgers because they were all the hospital sold, scrambled to find places to keep up with outside work, and had no place to catch our breath and relax. Our experiences as visitors provided a glimpse into the daily reality of healing and operating in a healthcare environment. Our healthcare system is witness to our country’s urban burnout culture, and we are passionate about the use of design as a retaliation to these cultural norms while still maintaining the regulatory requirements of health centers. Based on the team’s momentum from the MIT DesignX and Delta V programs, we decided to pursue this venture together to create havens that replace the dire memories of their own visits.
With what organizations are you currently partnering, if any? How are you working with them?
We would be grateful for the opportunity to join the Solve community in order to:
- Expanding our network within the healthcare payor and provider communities
- Growing our team with people who are passionate about making cities healthier, one building at a time
- Receiving mentorship to drive our go-to-market technical and product strategy
What is your business model?
We plan to invest in our relationships with healthcare organizations by first offering a one-time snapshot of their facilities across the network. This 'foot-in-door' one-time data alignment fee would cost our partners $20,000 and include:
- The generation of new spatial data (which they own),
- Statistical analysis and insights in relation to the network's site- or room-specific health outcomes, and
- A set of targeted, prioritized recommendations based on cost-effective impact
This will allow us to learn about a hospital’s full customer life cycle and refine our analysis before offering a recurring revenue product.
What is your path to financial sustainability?
While our first product provides a one-time snapshot to drive an upcoming capital-intensive project, we don't want to do this for customers just once. Our plan is to use the revenue from these one-time pilots to fund our future product development. We are excited to work with our pilot partners to create a follow-on offering that tracks building health across a healthcare network over time. By focusing on the analytics behind design effectiveness, we can ultimately drive systematic change by incentivizing professionals to design for health at scale. We plan to learn from our pilots and develop a health index that healthcare networks and other industries can use to evaluate their spaces. Finally, we also envision partnering with enabling vendors in the space (green wall builders, smart glass providers, etc.) to ensure that we bridge the gap between our recommendations and our customers’ ability to execute upon them.
Why are you applying to Solve?
We were excited by the crowd-sourcing culture that Solve embodies and are eager to get feedback on our concept.
If we receive the opportunity to participate in additional Solve events, we would be able to connect with the broader expert community to fuel our product development and understand how we can truly build sustainable solutions that make our cities healthier.
Finally, we are passionate about bringing our concept to reality and would love the opportunity to contribute to the cohort, meet like-minded entrepreneurial individuals, and potentially connect with future team members.
What types of connections and partnerships would be most catalytic for your solution?
With what organizations would you like to partner, and how would you like to partner with them?
We have identified more than 50 healthcare organizations that we would love to test our product with. Examples include broad healthcare networks like Mount Sinai, Partners, Kaiser, United Health Group, Sutter Health, and others that may be affiliated within the Solve network.