As artificial intelligence (AI) continues to transform a broad spectrum of industries, its impact in the healthcare field has been particularly important for many people. Renowned AI thought leader Hassan Taher counts himself among those people, calling the effect of AI on healthcare “groundbreaking” and its use to detect diseases at the nanoscale “a particularly exciting development.”
The author of The Rise of Intelligent Machines, AI and Ethics: Navigating the Moral Maze, and The Future of Work in an AI-Powered World, Taher is an established tech expert who has offered countless valuable insights about AI and its various applications. As the head of the consulting firm Taher AI Solutions, he has served clients across sectors that range from manufacturing to finance.
However, Hassan Taher is probably most devoted to capitalizing on the vast potential of AI in healthcare. During a 2023 interview, he was asked to single out one trend in AI that excites him the most. “The growing use of AI in the healthcare industry,” he answered. “With the potential to revolutionize patient care and outcomes, I believe that AI has the power to make a significant impact on people’s lives.”
Considering his well-established interest in healthcare-related AI, it should come as no surprise that Taher was among the first to voice his praise of a recent joint research project by the Center for Genomic Regulation, the University of the Basque Country (UPV/EHU), the Donostia International Physics Center, and the Instituto Biofisika. This project developed a highly specialized AI tool that can differentiate cancerous cells from healthy ones. It has also proven particularly adept at identifying the earliest stages of viral infection inside cells.
Called AINU (which is short for AI of the nucleus), this tool employs state-of-the-art STORM (stochastic optical reconstruction microscopy) imaging to create a picture that captures many finer details than no regular microscope can detect. In fact, AINU can register and accurately present nanoscale details, meaning it can zoom in to capture high-definition images at one billionth of a meter. These images can serve as an incredibly precise tool in the hands of medical professionals, and this level of precision is particularly important in cancer diagnosis. Cells exhibit extremely small anomalies during the onset of cancer that can be difficult to detect. But early detection is quite important in the world of cancer healthcare because most forms of cancer are far more treatable in their earlier stages.
The promising results of the AINU project were detailed in the peer-reviewed journal Nature Machine Intelligence on August 27, 2024. “The resolution of these images is powerful enough for our AI to recognize specific patterns and differences with remarkable accuracy, including changes in how DNA is arranged inside cells, helping spot alterations very soon after they occur,” says ICREA research professor Pia Cosma, co-corresponding author of the study and researcher at the Center for Genomic Regulation in Barcelona. “We think that, one day, this type of information can buy doctors valuable time to monitor disease, personalize treatments and improve patient outcomes.”
To accurately identify cellular anomalies at nanoscale, the AINU project relies on a subset of AI called machine learning (ML). Briefly defined ML allows intelligent machines and AI software to complete autonomous but directed learning processes that resemble traditional human learning in many respects. Specifically, ML involves the development of AI that can recognize patterns, make predictions, and (most importantly for the AINU project) detect anomalies.
The AINU project utilized a form of ML called deep learning. “Deep learning, a subset of machine learning, involves the use of neural networks with multiple layers to analyze complex data,” explains Hassan Taher. “In healthcare, deep learning has proven to be an invaluable tool, particularly in fields that rely heavily on image analysis, such as radiology and pathology. Unlike traditional machine learning, which often requires manual feature extraction, deep learning models automatically learn to identify patterns and features from vast amounts of data, making them exceptionally effective at tasks like image recognition and classification.”
Taher goes on to identify the analysis of medical images as one of the most significant applications of deep learning in healthcare. “Radiologists and pathologists have traditionally relied on manual examination of images, which can be time-consuming and subject to human error,” he writes. “Deep learning models, however, can process thousands of images in a fraction of the time, identifying patterns that may not be visible to the human eye.”
Beyond its oncology applications, AINU has proven effective in the diagnosis of viral infections. It was able to detect changes in the nucleus of the cell in as little as one hour after infection by the type-1 herpes simplex virus. Specifically, it was able to identify the presence of the virus by identifying slight differences in how tightly cellular DNA is packed. A loosening of DNA strands occurs as a virus begins to alter the structure of the cell’s nucleus.
“Our method can detect cells that have been infected by a virus very soon after the infection starts” says UPV/EHU research associate and co-corresponding AINU study author Ignacio Arganda-Carreras. “Normally, it takes time for doctors to spot an infection because they rely on visible symptoms or larger changes in the body. But with AINU, we can see tiny changes in the cell’s nucleus right away.”
From cancer to viral infections, deep learning models have incredible potential when it comes to analyzing diseases at the nanoscale. Hassan Taher is particularly excited about the level of detailed information deep learning can supply about a patient’s condition to drive the development of highly personalized treatment plans. “This approach could lead to more effective treatments with fewer side effects, improving patient outcomes and reducing healthcare costs in the long run,” he writes. “As researchers continue to refine these techniques and address the challenges associated with their implementation, the impact of AI on healthcare is likely to grow, transforming the way we detect and treat diseases.”
Next: Exploring the Future of AI, Startups, and Remote Work Trends