“AI is now everywhere. It’s really exciting to be at the forefront of AI with predictive models that are constantly evolving and reasoning in real time.”
In the rapidly advancing field of artificial intelligence (AI), there are exceptional individuals whose contributions have reshaped the landscape of various industries. Anet Potgieter, a distinguished Computer Scientist specialising in Cognitive Distributed Artificial Intelligence, is one such trailblazer. As the co-founder of EPCON, she played a pivotal role in the development of the company’s groundbreaking AI health platform, which employs spatial data and predictive analytics to analyse disease burden.
In addition, Anet is the founder and technical director of Cognitive Systems, a South African software engineering company focusing on cognitive computing, distributed artificial intelligence in the internet of things, and big data environments.
AI, From research to innovation
Anet’s career encompasses a wide array of roles and achievements. After she completed her PhD in Computer Science, her endeavours were shaped by expertise in various domains, including software engineering, software architecture, and knowledge engineering. Notably, her doctoral research was significantly influenced by the work of Marvin Minsky, a distinguished American cognitive and computer scientist who made remarkable contributions to the field of artificial intelligence. Inspired by Minsky’s ‘Society of Mind’ theory, which views the human mind as well as other naturally evolved cognitive systems as a vast society of individually simple processes known as agents, Anet embarked on a journey to bring her research to the forefront of technological innovation.
EPCON’s work in infectious disease control owes its genesis to Anet's groundbreaking research. In 2018, the founders participated in the business acceleration program offered by imec, a Belgium-based tech accelerator. Their exemplary work earned them the top spot in the competition. The competition win marked the start of EPCON’s journey to becoming a viable company. One of the key strengths of EPCON’s AI technology, which formed the foundation of Anet's research, is the model's ability to learn from real-time feedback.
The power of real-time learning
Anet explains that swarm intelligence, a subfield of AI rooted in distributed and self-organized systems, forms the intellectual basis of EPCON’s model. Unlike other AI models that necessitate offline learning, EPCON's capacity for real-time updates ensures its adaptability to changing environments. Since disease prediction requires constant updating and reasoning, EPCON's model proves to be a game-changer in the field of epidemiology.
Not without its challenges
Despite the advancements achieved by EPCON's AI model, Anet explains that there are inherent challenges, particularly in acquiring high-quality and comprehensive data. Often, the available data is "noisy," plagued by biases and incompleteness. EPCON's model thrives on a substantial amount of accurate data to achieve optimal predictive performance.
Many of the regions where EPCON operates are remote, leading to poorly documented reported cases. Moreover, past disease outbreaks might not accurately reflect present conditions. In addition, infectious diseases with long incubation periods require the model to reason over time, further complicating data completeness. To address these challenges, EPCON's technology must grapple with uncertainty, paralleling, to some extent, the complexities faced by models predicting climate change.