Built on the foundation of our core technologies, we deliver interpretable and performant end-to-end solutions to maximize the values in your data and elevate your analytics capability.
Current prediction models in medicine are often built using logistic regression, with limited accuracy. Furthermore, the model often has many many variables and is hard to use in the clinical setting.
In collaboration with top Massachusetts General Hospital surgeons, we built an actionable surgical risk calculator. It has currently 500+ users on iOS and Android.
The tree outputs and easy-to-use physician questionnaire provide transparent rationale for physicians to understand and adopt, and also make it straightforward to discuss and iterate with medical collaborators.
In collaboration with oncologists from Dana Farber Cancer Institute and Massachusetts General Hospital, we built an actionable cancer mortality prediction tool that estimates the probability of mortality for a particular patient and a particular envisioned cancer treatment.
Because the structure of the prediction is based on decision trees, a physician or even a patient can easily understand the reasoning behind the algorithm. The model also identifies key predictors of mortality such as change in weight.
We helped a large investment firm to understand and predict fund flows from its financial advisors. With Optimal Trees, our solution delivered accurate predictions and understandable business insight.
The tree recommends different marketing strategies for clusters of the firm's clients, with an estimated lift of 8-15% in fund flow using the recommendations.
Illustrated example of the fund flow prediction tree.
We have successful built a number of other products using the Interpretable AI software as their analytics engine. Examples include