For decades, mammography has been the go-to method for breast cancer screening. It’s saved countless lives—but it’s also come with trade-offs: discomfort from compression, exposure to ionizing radiation, and a frustratingly high rate of false positives, especially in patients with dense breast tissue. For many women, the experience is painful enough to delay or avoid screening altogether. But a new technique developed by Caltech researchers may finally offer a gentler, smarter alternative—one that replaces compression plates and X-rays with laser pulses, ultrasound waves, and artificial intelligence.
The technology is called photoacoustic computed tomography, or PACT. It’s been in development for over two decades, and in recent clinical trials involving 39 patients at City of Hope Comprehensive Cancer Center, it performed as well as or better than mammography and MRI in identifying suspicious and benign tissue. Unlike traditional imaging methods, PACT doesn’t require contrast agents, doesn’t expose patients to radiation, and doesn’t involve any painful squeezing of breast tissue. Instead, it uses near-infrared laser pulses to illuminate the breast. These pulses are absorbed by blood vessels and other tissues, which then emit ultrasonic waves. The system captures these waves to construct detailed images of the internal structure—essentially turning light into sound, and sound into insight.
What sets PACT apart isn’t just its comfort or safety—it’s the speed and precision of its AI-assisted analysis. In some cases, tumors were identified in as little as 15 seconds. The machine learning algorithms help differentiate between healthy and abnormal tissue, reducing the need for repeat scans and minimizing diagnostic uncertainty. That’s especially important for patients with dense breast tissue, where traditional mammograms often struggle to provide clear results. Ultrasound, while safe, depends heavily on the skill of the operator. MRI, though accurate, is expensive and not suitable for everyone—particularly those with implants, claustrophobia, or allergies to contrast agents. PACT sidesteps all of these limitations.
The system’s design also lends itself to accessibility. Because it doesn’t rely on radiation or contrast dyes, it could be used more frequently and in a wider range of settings—from hospitals to mobile clinics. That’s a big deal for underserved communities, where access to advanced imaging is often limited. And for patients who avoid screenings due to pain or fear, PACT could be the difference between early detection and missed opportunity.
Article from Caltech: AI-Assisted Technique Offers Safe, Effective, Painless Breast Imaging Alternative
Abstract from Nature Biomedical Engineering: Panoramic photoacoustic computed tomography with learning-based classification enhances breast lesion characterization