SnooCODERED Healthcare Logistics System

The president of the Ghana Medical Association disclosed that according to a survey they conducted among 360 medical doctors in the country, 90% of them had considered leaving Ghana (GhanaWeb, 2022). On average, 500 nurses leave Ghana for the West every month, and the UK now employs more Ghanaian nurses than Ghana does. (Wilson Center, 2023)

Even more unfortunately, this problem is not limited to Ghana, but is rather quite pervasive across sub-Saharan Africa. According to the WHO, from 2015 data, Liberia had a mere 51 doctors for a population of 4.5 million, and Sierra Leone only had 136 doctors for 6 million people, making the doctor-to-population rate 0.1 and 0.2 per 1,000 people respectively. Egypt’s former health minister lamented that 65% of Egyptian doctors are working abroad, while in Nigeria, 9,000 doctors left to work in the UK, the US, and Canada from 2016 to 2018. (Wilson Center, 2023)

The problem affects the entire population, as we in Ghana contend with 1 doctor to 6,500 (WHO, 2022), against WHO’s recommended 1:1,000, but the people living in poverty are the worst affected. It is reported that 42% of doctors are in the capital Accra alone, and up to 81.3% of all doctors are concentrated in 5 regions (WHO, 2022), all of which are the significant centres of commerce and administration. This means the 11 poorer regions in the country, containing 42.5% of the population (i.e. over 13 million people), are left to be cared for by 18.7% of the country’s doctors. 

Beside the human resource deficit, financial challenges, long distances to health facilities, and attitude of health staff were found to influence access to health care among vulnerable populations (Acquah-Hagan et al, 2022). Overall, as of 2021, private and public hospitals were available to 33.9% of urban localities in Ghana, while 1.9% of the rural communities had access to hospitals. Clinics and healthcare centres were available to almost 63% of urban localities, compared to 11.3% for rural residential clusters (Statista, 2023).

The lack of access to healthcare in developing countries like Ghana is the specific problem we are solving.

SnooCODERED is a family of digital solutions we built for emergency and public healthcare, and epidemiology, after the identification of our core technology, SnooCODE, by the WHO’s African Network for Drugs and Diagnostics Innovation (ANDI), as “a key technology in providing timely emergency care across Africa and in the battle against infectious and vector-borne diseases,” following the 2014 Ebola outbreak. Our system for emergency care, SnooCODERED Control Centre (SRCC), enables responders to determine the nearest health assets (ambulance stations; hospitals; pharmacies; individual doctors) to a patient and navigate easily to and from each site, improving response time and chances of survival. SRCC is also capable of dispatching a drone from the user’s location to the emergency site (the user simply enters the 6-digit SnooCODE address of the site) for a quick survey or for first aid medical delivery. A live video feed can be transmitted back by the drone, displayed to the user through SRCC’s interface.  

SnooCODERED, a Royal Academy of Engineering Africa Prize alumna, is built on our core proprietary technology, SnooCODE, a digital addressing system using geocoding technology, that allows anybody anywhere on earth to generate a memorable 5- to 7-digit alphanumeric address code (a SnooCODE) for their location and share with anyone else for easy navigation to them or in the case of a drone, the designated location(s). As the first "postal code system and method" for smartphones and other mobile devices (See https://patents.google.com/patent/WO2014080186A3/en), SnooCODE has Prior Art, safeguarding its Intellectual Property. All our systems work offline (without the Internet or even cellular connectivity), including SRCC, making them ideal for use in remote settings, rural and other areas with little or no connectivity. We began integrating AI with SRCC this year through Project “Genie,” for comprehensive decision support and triage, providing a premium version that requires the Internet.  SnooCODERED is evolving from a tablet/smartphone emergency control centre into a modular solution that can be assembled and used in 4 different applications: stationary emergency call centre; mobile EMT (ambulance); stationary primary e-healthcare outpost; mobile clinic. Each application has 2 main components in common: satellite Internet; a tablet with the SnooCODERED Control Centre (SRCC) now integrated with AI for quick decision support. Please watch our SRCC product demo here: 

Our solution is designed to expedite health care delivery in resource-constrained settings such as found in Ghana. Our solution is especially suited for emergency, primary care and public health care delivery support. 

From estimates by the World Health Organization (WHO), for every ten cases referred by the national ambulance service in Ghana in 2022, one did not receive emergency support either due to lack of ambulances or bed space in receiving hospitals (WHO, 2023). According to the Ghana National Ambulance Service, out of 20,236 patients transported in 2014, only one was attended to within the recommended target response time of less than 8 minutes (Mahama et al, 2018). The median transportation time was 82 minutes and was as high as 552 minutes for some patients. Furthermore, at the hospitals, the reported doctor to patient ratio is 1:6500. 81.3% of doctors are in only 5 regions, and over 60% of doctors are in only 5 teaching hospitals in the country (WHO, 2022). This inequitable distribution of expertise, in a country with a poor emergency response and referral system, spells out disaster. The solution to this problem is multifaceted, and our SnooCODE RED system is designed to mitigate many of these glaring health disparities.

It does this in three ways. Firstly, it helps emergency responders to accurately locate people in need of assistance and to identify and coordinate with the appropriate health assets (with the right resources for the case) closest to the patient to reduce response times. Secondly, it provides first aid instructions to laypersons at the site of the emergency even as they await the arrival of trained emergency responders. Thirdly, it serves as a decision support tool for emergency responders, taking into context the setting and resources available at the location of the emergency. By providing the right context, we are developing solutions tailored to the resource constraints of developing countries.   

In our research and development process, we have been engaging with the people our solution would impact, from the people in resource-limited areas of the country to emergency responders and clinicians. We have continuously and iteratively incorporated their feedback into the tools we are building to meet their needs. For example, on a visit to a primary health facility in the Volta Region, the nurse-in-charge shared with us how pregnant women with complicated labours should have  had to arrange with a taxi driver prior to the onset of their labour so they could be rushed to a district hospital about 50km away and over unportable roads to receive specialist care as the ambulance service could not be relied on.  Furthermore, during interactions with clinicians at a conference by the Emergency Medicine Society of Ghana, they urged us to incorporate Artificial Intelligence (AI) capabilities within our solutions to better serve their needs. We have been building our SnooCODE RED Emergency app with these insights to address the pain points of our communities.

SnooCODE is a close-knit team of creators, engineers, mathematicians and social scientists working to provide class-leading location-based solutions to power core services around the world.

We have won awards & recognition from several notable institutions, such as the Royal Academy of Engineering, the WHO and the AFD.

We have learnt that this recognition is not solely for our innovation, but also for the deep knowledge we’ve gained living within the context of the problem.

In ensuring that our solution is clinically useful, we seek the input of clinicians who attend to medical emergencies in resource-deprived locations as we build, and we let them evaluate what we have built to ensure that it meets their needs. They have been quantitatively and qualitatively evaluating various iterations of our clinical decision support tool for managing medical emergencies, and we build better based on their feedback.

In trials conducted with West Africa’s 1st accredited training centre for emergency medicine residency, the Komfo Anokye Teaching Hospital (KATH) in Kumasi (in the Ashanti Region, where our project will be taking place), we helped reduce response times by as much as 56% with SRCC. After the system was validated, we worked with Ghana’s National Ambulance Service to train over 400 EMTs on its usage. Following that, in 2017, we executed a motorcycle ambulance project together with Oxford University, the National Ambulance Service and KATH, where we trained 60 EMTs in the use of a specially designed “portable toolbox for first-response emergency medicine.”

Our clinical decision support tool is being built by a team of Ghanaians with our Artificial Intelligence (AI) incorporation led by a physician and AI practitioner who has 25+ years of combined living and clinical work experience in the Ashanti Region, our pilot region. We also regularly consult Dr. Bonney, a specialist emergency physician at KATH, and the current president of the African Federation for Emergency Medicine.

Our series of evaluation studies so far have involved 40 clinicians based largely in Accra (capital of Ghana) and in the Ashanti Region, with levels of experience across all ends of the spectrum, from interns to emergency medicine specialists with 12+ years of experience. By iteratively improving our solution based on their feedback, we have seen the percentage satisfaction of clinicians with the differential diagnosis and first aid instructions outputted by our context-aware clinical decision support tool rise from 72% to 80.5%. Please see here for two of our completed studies, one of which has been accepted as a conference publication and the other undergoing peer review: (Study 1: https://doi.org/10.1101/2024.04.03.24305276 , Study 2: https://doi.org/10.1101/2024.04.17.24305971)

We are currently at TRL 6: Sub-system model demonstration in a relevant environment. We have been focused on developing our products to world-class standards and testing them through grant-funded projects and partnerships before going to market. 

Our SnooCODERED Control Centre tablet and smartphone app has been validated in trials by our emergency medical service (EMS) partners, including West Africa’s first accredited training centre for emergency medicine residency, Komfo Anokye Teaching Hospital (KATH), as well as with Oxford University and the National Ambulance Service. We have also collaborated on projects that test or demonstrate SRCC with the African Federation for Emergency Medicine, the Emergency Medicine Society of Ghana

We are now integrating AI with SRCC and developing it into a module with the addition of satellite Internet and solar energy. 

We are applying to Solve because we believe in the power of collaboration to address complex global challenges. Solve's platform provides a unique opportunity to connect with a diverse network of partners who can help us overcome specific barriers and accelerate the impact of our solution.

1.⁠ ⁠Financial: Access to funding is crucial for scaling our solution and reaching more underserved communities. Solve's network of partners can provide not only financial support but also access to resources and expertise to help us grow sustainably.

2.⁠ ⁠Technical: As we continue to innovate and evolve our solution, we may encounter technical challenges that require specialized knowledge or resources. Solve's network can connect us with technical experts who can provide guidance and support in overcoming these challenges.

3.⁠ ⁠Legal: Navigating legal frameworks, especially in different regions or countries, can be complex. Solve's partners can provide legal expertise to help us ensure compliance and navigate any legal barriers that may arise.

4.⁠ ⁠Cultural: Adapting our solution to different cultural contexts is essential for its success. Solve's network can provide insights and guidance on cultural considerations, helping us tailor our approach to better meet the needs of diverse communities.

5.⁠ ⁠Market: Understanding and navigating the healthcare market, particularly in underserved communities, is critical for scaling our solution. Solve's network can provide market insights and connections to help us expand our reach and impact.

Overall, we see Solve as a valuable partner in helping us overcome these barriers and advance our mission of improving access to emergency healthcare and epidemiological solutions for underserved communities. We are excited about the opportunity to collaborate with Solve and its partners to create meaningful and sustainable impact.

SnooCODERED is innovative in 4 main ways:

1. Core technology: SnooCODERED is developed on the backbone of our proprietary digital addressing system, SnooCODE, that works as a novel "postal code system and method," allowing anyone to generate a memorable 5-7 digit alphanumeric address code for their location. This enables accurate navigation and location mapping in areas without traditional addressing systems (majority of Africa).

2. Offline operation: The SnooCODERED Control Centre (SRCC) system can function entirely offline without Internet or cellular connectivity, making it suitable for remote or low-connectivity areas. The only time the Internet is needed is when using our new, premium AI-integrated version.

3. AI integration: We are finalising the integration of LLMs like ChatGPT and Gemini into SRCC to provide decision support and triage advice tailored to the specific emergency context, location type, patient demographics, etc., combining location intelligence with AI-powered decision support, e.g. the system can recommend that a patient with fever, lower abdominal pain and vomiting on a rural farm find a clean container, take in oral rehydration fluids, use a cool damp cloth on the forehead to reduce fever and to visit the nearest maternity home for further management (as a hospital is much farther away), compared to more generic advice given by mainstream models.

4. Modular solution: From a modest start with an Android tablet equipped with SRCC for EMTs and first responders, we are developing our prototype to fit 4 applications (stationary emergency call centre; mobile EMT; stationary e-health outpost; mobile clinic), with the e-health outpost being a fully-fledged solar- and Satellite Internet-powered health kiosk that houses our AI-integrated tablet, a human health assistant, CCTV cameras that can scan ID cards for patient data, and basic equipment like a thermal imaging gun, weighing scale, etc., for emergency public healthcare in both rural and urban areas. This also presents an innovative way to extend primary healthcare access to underserved rural and urban populations.

Our solution can catalyze positive impacts by improving healthcare accessibility, enabling early intervention, enhancing patient empowerment, and generating valuable health data to guide resource allocation and decisions. It has the potential to change the healthcare landscape, especially in low-resource settings, by complementing existing facilities with a decentralized mobile clinic model powered by innovative location-intelligence and decision support technologies.

The theory of change for how SnooCODERED expects to have an impact on improving access to healthcare:
Activities:

Develop and distribute SnooCODERED mobile app for emergency responders to locate nearest health facilities
Integrate AI decision support into app to provide triage advice based on situation
Deploy mobile clinics outfitted with SnooCODERED system to underserved areas
Train community health workers on using SnooCODERED for public health monitoring

Outputs:

Faster emergency response times getting patients to appropriate facilities
Earlier detection of health issues through community outreach via mobile clinics
More data collected on community health patterns to inform resource planning
Public education on when/how to summon emergency care via SnooCODERED

Outcomes:

Increased utilization of ambulance/emergency medical services
Increased primary care visits facilitated by mobile clinics
Lower mortality rates for time-sensitive emergencies like obstructed labor
Better containment of infectious disease outbreaks through surveillance

Impact:

Improved access to quality healthcare for underserved populations
Reduced burden on tertiary care hospitals as more cases managed at primary level
Healthier communities and populations across regions served

By bridging spatial and resource gaps, the SnooCODERED system links a sequence of activities to expand healthcare access, utilization and improved health outcomes in an evidence-based theory of change.

Impact Goals:

1. Increased Patronage of Ambulance Service: Increase ambulance service usage by at least 10% among underserved communities, reducing reliance on taxis and buses for emergency transportation.

2. Increased Patronage of Primary Healthcare Services: Increase the utilization of primary healthcare services with AI decision support by at least 10% among underserved communities, promoting early detection and management of health conditions.

3. Early Detection of Diseases: Facilitate early detection of diseases through increased healthcare patronage, leading to fewer complications and reduced burden (from referrals) on higher-level facilities.

4. Reduced Emergency Response Times: Reduce emergency response times by at least 40%, improving access to timely healthcare interventions.

5. Lower Mortality Rates: Lower mortality rates from time-sensitive emergencies by at least 10%, indicating improved outcomes for patients.

6. Reduced R0 in Epidemics: Decrease the R0 in epidemics and public health emergencies through granular contact tracing and epidemiological modeling, contributing to disease control and prevention.

How we will measure progress:

1. Ambulance Service Patronage: Track the number of ambulance service calls and the percentage increase in usage compared to previous years.

2. Primary Healthcare Service Patronage: Monitor the number of patients accessing primary healthcare services with AI decision support and the percentage increase in utilization.

3. Early Detection of Diseases: Analyze data from AI-based decision support tools to identify trends and patterns in early intervention.

4. Emergency Response Times: Measure the average emergency response time and the percentage reduction achieved over time.

5. Mortality Rates: Calculate the mortality rate for time-sensitive emergencies and track the percentage decrease in mortality rates.

6. R0 in Epidemics: Utilize epidemiological data to calculate the R0 for targeted diseases and monitor the reduction in R0 over time.

Data Collection and Analysis:

1. Implement electronic medical record systems on AI-integrated tablets for patient data collection.

2. Train staff in data management and analysis techniques.

3. Regularly analyze collected data to assess impact on healthcare access and outcomes.

These indicators align with several UN Sustainable Development Goals, including Goal 3 (Good Health and Well-being) and Goal 9 (Industry, Innovation, and Infrastructure), demonstrating the comprehensive impact of the solution on healthcare systems and communities.

Our core technology, SnooCODE, is powered by a combination of radio navigation systems, notably Global Positioning Systems (GPS), Global Navigation Satellite Systems (GNSS) and GLONASS (Soviet Union GNSS). 

The system is map-agnostic, acting as an addressing layer on top of the user’s selected mapping system, e.g. Google Maps, HERE maps, Apple Waze. This allows the user to leverage whichever mapping system they prefer or is suitable for them – HERE maps for instance, is ideal for ambulance services in the developing world, as it enables downloading of larger offline maps than Google’s.

The public can download the free SnooCODE digital addressing app from the iOS or Android app stores, stand outside their location entry point and press “Generate code” to immediately get their unique 6-digit alphanumeric code (e.g. 4XY-VTU) for Home or wherever they are – be it a large event or an accident site. They can save it, share and navigate without the Internet (the only time an Internet connection is needed is when downloading the app from the app store or when sharing the code via Internet-dependent platforms like WhatsApp), provided they have a smartphone. As situations change, parameters within the app can be changed using SMS.

In an emergency, a user simply needs to mention their 6-digit code to the emergency centre operator, who then inputs it in the SnooCODE RED Control Centre tablet app to view the nearest health centres in sequence, contact them to confirm availability, then dispatch their drivers to navigate to the emergency with the SnooCODE RED Dispatch mobile app, or input the SnooCODE in a drone with our Android drone control system for remote monitoring or medical delivery.

We are now finalising the integration of AI into our SRCC app and developing it into a modular solution to work in 4 applications, stationary on the go, with the addition of solar power & satellite Internet.

5 full-time; 4 part-time; 2 contractors; 2 advisors 

We have been working on SnooCODERED since 2015, as a sort of corporate social responsibility arm of our company, SnooCODE. This was inspired by the identification of SnooCODE, by the WHO’s African Network for Drugs and Diagnostics Innovation (ANDI), as a key technology in providing timely emergency care across Africa and in the battle against infectious and vector-borne diseases, following the 2014 Ebola outbreak.  We however begun directing more resources and focus towards SnooCODERED as a viable business unit in 2020 following the Covid-19 outbreak.

At SnooCODE, we strive to build a diverse, equitable and inclusive team environment. On our leadership team, we have made a concerted effort to have representation across gender, cultural and socioeconomic backgrounds. Half of our executive team identifies as women, and team members come from a range of ethnic groups across Ghana.

We believe diverse perspectives and experiences make us stronger as an organization. We actively work to minimize barriers and provide equal opportunities regardless of background. Our recruitment processes focus on skills and potential rather than pedigrees. We offer flexible work arrangements to accommodate different lifestyles and needs. All staff members receive training on unconscious bias.
Fostering an inclusive culture where everyone feels welcomed, valued and empowered to contribute is very important to us. We have employee resource groups that provide support networks. We celebrate cultural holidays and traditions from all the backgrounds represented on our team. An open-door policy allows staff to raise any concerns about inclusion.
Overall, while we still have progress to make, we are committed to actions that promote diversity of thought and experience at SnooCODE. We believe this allows us to better understand and serve the diverse communities we aim to support through our healthcare solutions.