Video - Envisionit Deep AI

Video and Interview Sessions with Dr. Jaishree Naidoo, CEO and Co-Founder

In setting the scene for this discussion, please give us an overview of your company – and the vision that has led to its success as an African Start-Up.
As a practising paediatric radiologist in South Africa I became absorbed with the question of how I and my peers would cope with the demand for quality healthcare in Africa. Africa is a continent with the youngest population in the world without sufficient access to paediatric diagnostic care. As one of just 20 specialist paediatric radiologists in Africa, this led to my vision to use Artificial Intelligence and machine learning to augment and multiply what I was doing in my daily work. After researching the concept part-time, I eventually left my academic and clinical career in 2019 to start Envisionit Deep AI together with my two co-founders. Since then, Envisionit has progressed exponentially to become a provider of world-class products and services that can support faster and more accurate diagnosis and triaging of patients in Africa, the UK, Middle East and most recently the USA.

Envisionit it now headquartered in the UK and recently received the approval of the United States’ Food and Drug Administration (FDA). How did this come about?
In 2020 Envisionit was appointed to the Global Entrepreneur Programme with the British Department of International Trade and hence we moved our head office to the UK. As a result we have been fortunate to travel extensively in order to participate in numerous international medical conferences and radiology symposiums, which eventually led to our successful receipt of a 510(k) FDA clearance for our AI assisted chest x-Ray solution – Radify.
This means that Envisionit Deep AI’s Radify Triage has been officially cleared to triage Pneumothorax and Pleural Effusion, two critical findings that represent crucial challenges in emergency rooms and intensive care units. In essence, Radify significantly reduces the time taken to alert ER doctors to critical pathologies that require urgent attention. In other words, our solution is a game changer that will enhance the efficiency of doctors, leading to better patient care.

More specifically, how does AI in radiology work – and what are the benefits?
AI in radiology refers to the use of artificial intelligence technologies, such as machine learning and deep learning algorithms, to assist radiologists in interpreting medical images. These algorithms analyse images such as X-rays, CT scans, and MRIs to identify abnormalities and assist in diagnosis.
AI offers several benefits to both radiologists and patients. For radiologists, AI can help in detecting abnormalities more accurately and efficiently, reducing the chances of oversight or misinterpretation. This can lead to faster diagnosis and treatment planning. For patients, AI can potentially lead to earlier detection of diseases, improved treatment outcomes, and reduced healthcare costs.

So, can AI completely replace human radiologists?
AI is not intended to replace human radiologists but rather to augment their capabilities. While AI algorithms can analyse images and identify patterns, human expertise is still crucial for interpreting complex cases, considering patient history, and making clinical decisions. AI serves as a valuable tool for radiologists, assisting them in their workflow and improving diagnostic accuracy.

How is AI in radiology regulated to ensure patient safety and data privacy?
Regulatory bodies such as the FDA (Food and Drug Administration) in the United States have established guidelines for the development and deployment of AI in medical imaging. These guidelines ensure that AI systems meet safety and efficacy standards before they can be used in clinical practice. Additionally, healthcare organizations must adhere to strict data privacy regulations, such as HIPAA (Health Insurance Portability and Accountability Act), to protect patient information when using AI systems.

What are the challenges and limitations of AI in radiology?
Despite its potential, AI in radiology faces challenges such as the need for large annotated datasets for training algorithms, variability in imaging techniques and equipment, and the potential for algorithmic biases. Additionally, integrating AI into existing healthcare workflows and gaining acceptance from radiologists and other healthcare professionals can be challenging. Continuous validation and improvement of AI algorithms are necessary to address these limitations and ensure their safe and effective use in clinical practice.

How can healthcare providers adopt AI in radiology effectively?
Healthcare providers can adopt AI in radiology effectively by collaborating with technology vendors and researchers, investing in infrastructure and resources for AI implementation, and providing training and education to radiologists and other staff members. It’s important to start with pilot projects to evaluate the performance and impact of AI algorithms in real-world clinical settings and to involve stakeholders in the decision-making process.

What does the future hold for AI in radiology?
The future of AI in radiology is promising, with ongoing research and development aimed at further improving the accuracy and efficiency of AI algorithms. As technology advances, AI is expected to play an increasingly important role in radiology, helping to enhance diagnostic capabilities, personalize treatment plans, and improve patient outcomes. Continued collaboration between healthcare providers, researchers, and technology developers will be essential to realizing the full potential of AI in radiology.