Cancer surgery is a delicate dance, aiming to remove the diseased tissue while preserving as much healthy tissue as possible. It's a challenging task, often relying on pre- and post-operative imaging to locate and confirm the removal of cancerous cells. However, a groundbreaking technique developed by Lihong Wang and his team at Caltech and the City of Hope in California is set to revolutionize this process.
The Current Challenge: Balancing Precision and Preservation
Surgeons face a delicate balance when operating on cancer patients. While removing all cancerous tissue is crucial, preserving the affected region is equally important. Take lumpectomies for breast cancer, for instance. Studies show that patients survive just as well with this conservative approach as they do with more extensive procedures. However, the current method of ensuring complete cancer removal relies on post-operative pathology, which can lead to multiple surgeries for up to a third of breast cancer patients.
A Revolutionary Solution: AI-Assisted Intraoperative Histology
Lihong Wang, a renowned medical engineer at Caltech, has answered this challenge with a novel approach. By analyzing excised tissues during surgery using AI, surgeons can continue removing tissue until all cancerous cells are gone. This technique, known as ultraviolet photoacoustic microscopy (UV-PAM), eliminates the need for traditional, time-consuming tissue preparation methods.
Traditional Imaging vs. UV-PAM: A Comparison
Traditional imaging of tumor samples involves several steps: stabilizing the tissue, slicing it, and then staining it with specific chemicals to visualize the difference between cancerous and healthy cells. This process can be time-consuming and may cause distortions, especially with fatty breast tissue. Additionally, the accuracy of the analysis depends on the pathologist's skill, making standardization difficult.
UV-PAM, on the other hand, uses a low-energy laser to excite the tissue, causing the cell nuclei to appear brighter as they absorb the laser's light. This natural staining process, as Wang calls it, also leads to ultrasonic sound waves, allowing for precise imaging with a resolution of 200-300 nanometers. The image is then adjusted using AI to resemble traditional staining, making it easily readable for pathologists and surgeons.
The Benefits of UV-PAM: Speed, Accuracy, and Versatility
One of the key advantages of UV-PAM is its speed. Surgeons requested an analysis time limit of 10 minutes, and Wang's team is confident they can achieve this, providing results fast enough to guide decisions before closing the incision. Additionally, the technique provides more data than a single pathologist could analyze, and it appears to work equally well on various tissue types, including breast, bone, skin, and organ tissues.
The Future of Cancer Surgery: A Commercial Product in the Making
Currently, Wang's team is in the testing phase, but they aim to develop a commercial product that can be widely used. This technology has the potential to significantly improve cancer surgery outcomes, reducing the need for repeat surgeries and providing more accurate, timely results. The research, titled "Rapid cancer diagnosis using deep-learning–powered label-free subcellular-resolution photoacoustic histology," was published in Science Advances on November 21, 2025, and was funded by the National Institutes of Health and the National Research Foundation of the Korean government.
A Thought-Provoking Question:
As this technology advances, it raises an intriguing question: Could AI-assisted intraoperative histology become the new standard for cancer surgeries, revolutionizing the way we approach cancer treatment?
