2022 Cohort > eMoodie Limited

eMoodie Limited

Business Proposition

eMoodie provides digital mental health care solutions via a two-pronged strategy across our products.

eMoodie Labs is a mental health research platform which enables scientists to design complex psychological studies that establish valuable insights into mental health treatments. This platform, which consists of a web-based portal and a companion app available on both Android and iOS, is already fully functional. We license access to the platform to conduct clinical and academic research projects.

eMoodie Minds is a mental health app which facilitates early screening and data-driven tailored symptom-management using evidence-based digital interventions. Our USP is a validated screening algorithm, which can detect symptoms of depression, anxiety, social anxiety, and disordered sleep at a 95% accuracy level. We are currently expanding our screening technology using deep learning AI. It leverages voice and facial expression data for the assessment of a user’s current mood, stress level, and ultimately, burnout risk.

The combination of our products gives rise to a complementary strategy whereby the research platform helps translate and validate scientific evidence into digital solutions which then become part of the mental health app’s intervention/assessment suite. This will allow eMoodie to continuously innovate at a rapid pace – thus setting it apart from its competitors. There is tremendous value added as our research partnerships help to legitimize the eMoodie brand.

The mental health app provides digital therapeutic interventions bespoke to the user’s symptom profile. The first step is screening. We collect data from various sources, including surveys, smartphone sensors (e.g. accelerometer, GPS etc.), and voice and video recordings. From the raw input data, we extract low-level features (e.g. movement intensity can be extracted from GPS data, and paralinguistic features such as speech tone can be extracted from voice recordings etc.). The low-level features are then mapped to behavioral markers (e.g. stress can be determined from movement intensity and speech patterns), and, finally, those behavioral markers are mapped to clinical states (e.g. anxiety and depression). We build multimodal machine learning and deep learning models to achieve those mappings, but focus mainly on the latter as deep learning models have shown superior performance in many real-world applications, including mental health where it has shown some promising results.

The second step maps the screening results to relevant interventions. While users go through the recommended interventions, the app continues screening to monitor progress. The screening results are then used to continuously update the recommended interventions. This feedback loop ensures that the user always gets the most appropriate intervention given their symptom profile at any point in time. The mapping of symptoms to recommended interventions also leverages AI techniques, such as decision trees.

We hold all of the IP associated with our algorithms, software code, and graphics.

Investment Ask

£500k – £999k