DocMe

GIF
DocMe App Screen Shot

Inspiration

Our inspiration comes from our motivation to broaden access to healthcare. Our idea stemmed from the questions: “What if you could monitor your health without the long waiting times, expensive wearables, and uncertainty we see in everyday clinics? What if you could monitor your health with something much simpler — perhaps even a selfie?” This sparked a valuable discussion that led to the creation of our product: DocMe.

By providing remote diagnostics and guiding clinical decision making through a smartphone lens, we would not only allow you to be more informed about your everyday health but also be able to conveniently connect you with your doctor.

What it does

DocMe is a preventative health technology putting healthcare in the hands of individuals. By utilising edge-AI and computer vision, DocMe is able to:

easily capture the user’s vital signs directly from a smartphone camera and immediately receive insights powered by accredited research
connect users with their chosen medical professionals through a "privacy-first" telemedicine platform (Coming Soon!).

DocMe gets to work by first scanning the user’s face through their smartphone camera. The app then extracts minuscule details and analyses them. Today, the application can analyse 68 different regions on a user's face using AI, and, by magnifying subtle movements of the user’s skin, DocMe can even extract early warning symptoms of disease.

Importantly, our solution can be achieved without requiring additional wearable components. Thus, our technology enables remote, frictionless vital sign monitoring and the results can be accessed in real time, thereby enhancing the speed of clinical responses. Moreover, clinically, it is valuable to add physiological information to the patient’s history and symptoms in order to guide personalised care. Our edge-AI model processes and stores user data on their device. Hence, DocMe doesn't require a steady mobile connectivity to use its self-assessment tool. [At the moment, our demonstrator that uses a web server model due to building functionality from scratch in a narrow implementation time frame] Additionally, users can choose to contribute to anonymised data based on the user profile to feed the research component of DocMe.

Finally, due to our edge AI processing and our software architecture, DocMe will be GDPR and HIPPA compliant.

How we built it

We used the React Native framework to build a robust mobile app in JavaScript that runs on both iOS and Android devices. To create a secure offline database on the user’s device, we integrated the realm.io platform into the app. We implemented a REST API to interface between the user’s device and our server backend. OpenCV will be used to analyse and calculate heart rate, respiratory rate, and more from just a selfie video. In addition, we took advantage of a multitude of React Native component libraries to construct the live video feed, video upload mechanism, login/signup form validation, app light/dark theme, and more.

Challenges

Given that our team spanned six time-zones, initial communication between all team-members seemed very difficult. Throughout any time of the day, it was a challenge to contact team-members at times appropriate to their time-zone. However, we soon realised how to leverage this into an advantage. We proceeded to organise teams and gain more information about available hours for our team members in order to make the most of everybody’s daytime. This way, we were able to get a multitude of assignments and objectives completed in an extremely efficient manner.

Accomplishments that we're proud of

We are very proud of the amount of work we were able to accomplish in a relatively short period of time. DocMe was able to develop a working prototype of the App, formalise a USP for the service, continue biomedical research efforts in the format of White Papers, and foster strategic, mutually-beneficial partnerships Dell Medical School and Diya Health all in a span of 3 weeks. We are also proud of producing a beta version of our application in the time of this hackathon, as DocMe is ready for beta testing on iOS and Android mobile devices (iPhone X and above). Finally, we are especially proud of our diverse, interdisciplinary team ranging from Malaysia to Cambridge to San Francisco for leveraging their unique skills to exceed our initial expectations and we could not be happier.

Next Step

Our next steps would be to solidify DocMe as a viable business, conduct trials for efficacy, improve our AI diagnostics, and add features to create a more user-friendly experience.

Built With

Submitted to

UCSF COVID-19 Hackathon

Created by

Biomedical Engineering Student, University of Illinois at Urbana-Champaign | Worked as a part of the Clinical Research team to produce scientific White Papers through medical literature review. Also, contributed to Blog Post writing for the perfeXia website.

Amartya Dave
Bioengineering undergraduate at UIUC, Clinical Researcher, Data Analyst, and EMT-B. All things philanthropy, engineering, and medicine.
Chief Revenue Officer at Clinked.com | Advised and contributed on Business Development Strategy including product market positioning and value proposition of DocMe, persona & use case development, submitted video, business plan development and product marketing.

Sarah Broderick
Medical Graduate, University of Cambridge, Current Junior Doctor in the NHS at Lister Hospital, Stevenage | Performed feasibility analysis of the product from a translational (clinical and technical) point of view. In the process of integrating modified national risk stratifying protocols (e.g. NEWS) into the app. Lead the clinical feasibility study report (submitted as supplementary material).

Prabhvir Marway
NHS GP at Unity Health| Advised the team on real-life clinical practice and optimising doctor-patient communication.

Doctor Amy Kerstein
Computer Science Student, USC | Spearheaded app development on iOS platforms and improved UI and UX of DocMe App.

William Chang
PhD in Quantum Computing, University of Cambridge | As the team lead, I developed a truly multinational, multidisciplinary team spread across six timezones spanning from Malaysia to San Francisco and has the passion, skills, commitment, and more importantly, the ability to deliver regardless of the environment. My major focus was in managing the project, forming strategic partnerships with other participating teams, collaborating on the video content and design.

Pratyay Poddar
Chairman and CEO at perfeXia Health • Cambridge PhD in Quantum Computing • Industrial Fellow of the Royal Commission 1851
Experienced tax specialist and currently working as a business and technology specialist using an Agile methodology on projects, master's degree in Engineering business and management, Bauman Moscow State Technical University | Assisted the team with market analysis and checked commercial feasibility of the project to turn it into a viable business.

Veronika Koshkina
Computer Science Graduate, Monash University | Analysed Face Emotion Recognition (FER) using Deep Learning [Tech Stack: Tensorflow, Keras, OpenCV, Tensorflow Lite (Deployment)]

Vicky Feliren
Philomath.
Clinical Medical Student, University of Cambridge | Assisted the clinical advisors by conducting medical literature research and produced a clinical feasibility study report (submitted as supplementary material).

Jye Quan Teh
Medic with skills in Data Science, ML and Web/App Development
Software Architect and entrepreneur |Drafted the architecture and contributed as the Lead Developer for the prototype (beta test ready). Also, contributed to the implementation of OpenCV based detection of Respiratory Rate and Heart Rate.

Miro Svetlik
Chemical Engineer, University of Cambridge | Assisted in the back-end development of DocMe. Additionally, utilised data analysis methods to determine Respiratory Rate using Heart Rate Variability (HRV).

MikkaAlon
Computer Science Student, Bangalore Institute of Technology | Worked on the frontend development of DocMe. Also, built the landing page of the website (https://perfexia.health/docXme)

Rishabh Gupta
Clinical Medical Student, University of Cambridge, Investment Banking Spring Intern | Advised on medical aspect of product offering, business landscape analysis and business strategy development. Assisted with deliverable video creation and marketing.

Usama Safeer
Medical aspect of product offering, business landscape analysis and business strategy development.