Project Objective
Viavi sought to develop an advanced online platform for health prediction and personalised wellness strategies. The project’s purpose was to analyse individual patient genotypic and phenotypic data, using proprietary pattern recognition methodologies to predict health-related factors and risks. By employing computational deterministic models and sophisticated AI techniques the company’s technical team hoped to be able to generate accurate predictions and outcomes tailored to each patient and client-specific reports within the clinic.
While the company observed existing similar solutions, none possessed the comprehensive capabilities that the company aimed to achieve. Consequently, the technical team embarked on a long journey of extensive research and development of to create their algorithms and data processing techniques in order to create an advanced system capable of predicting and improving client health outcomes in a unique and unprecedented manner.
Scientific Baseline
The baseline upon which this advance is being measured is the ability within the public domain to analyse individual patient genotypic and phenotypic data to predict health-related factors. Prior to this advance, the company faced significant uncertainties with existing technology in the healthcare industry. Essentially, the proprietary pattern recognition methodologies necessary for identifying correlations between patient data and health patterns were not available in any pre-existing solution.
The technical team conducted extensive research and development to create a novel solution. The ability to aggregate this data to train the predictive models proved to be challenging, given the lack of online resources and libraries. This necessitated independent research and problem-solving to structure and enhance system performance. Through the course of this project, the company aimed to pioneer a ground-breaking AI-driven platform capable of accurate health predictions tailored to their client needs.
Scientific Advancement Achieved
A key technological advancement the company achieved during the relevant period was the successful integration of AI and machine learning into their Java Spring-based platform. Particularly noteworthy was their advancement in data aggregation and data normalisation of large datasets. Leveraging Python, TensorFlow, and Keras, the team developed advanced neural networks and computational deterministic models, enabling efficient handling and analysis of the vast and diverse patient data. This breakthrough marked a significant milestone in the project, empowering the platform with the ability to process and make accurate predictions based on structured and normalized data, thereby enhancing its overall performance and predictive capabilities. These technologies enabled the platform to process vast amounts of genotypic and phenotypic data efficiently, facilitating accurate predictions of health-related factors and patient outcomes.
Furthermore, the company’s technical team employed proprietary pattern recognition methodologies to enable the system to identify correlations between patient data and health patterns, empowering medical professionals with valuable insights to provide tailored and data-driven treatments. Moreover, the data-gathering and storage ensured a scalable and robust infrastructure, facilitating the management of both structured and relational data. Additionally, they enabled the platform to meet the performance demands required for generating timely client reports within their clinic, improving patient outcomes.