Panoramic photoacoustic computed tomography (PACT) with machine learning assistance could be a safe, noninvasive, and sensitive alternative to mammography, ultrasound, and magnetic resonance imaging (MRI) for breast cancer screening, according to study results that were published in Nature Biomedical Engineering.
When compared with standard imaging methods, PACT performed comparably in its classification between normal and suspicious breast tissue.
"We'd like to make PACT a clinical tool that benefits patients—to help detect breast cancer without patients taking the risk of getting cancer or worrying about an allergic reaction," said corresponding study author Lihong V. Wang, PhD, the Bren Professor of Medical Engineering and Electrical Engineering at Caltech.
Background
Standard imaging approaches for breast cancer screenings can be flawed: mammography involves ionizing radiation, may produce false-positives, and is less effective for radiographically dense breasts; ultrasonography can lack specificity and has operator-dependent image quality; and MRIs are costly, take longer, and are not an option for many patients due to allergies, implants, and claustrophobia.
"We were strongly motivated to work on this problem because none of the current techniques are perfect," said Dr. Wang, who is also the Andrew and Peggy Cherng Medical Engineering Leadership Chair and Executive Officer for Medical Engineering at Caltech. "The future of medicine has to be better than that."
Study Methods and Findings
The researchers from Caltech have developed and refined the PACT technique over the past 20 years. PACT consists of a laser-sonic scanner that can identify tumors in as little as 15 seconds. "We use light to see the molecules, but we use sound to define the spatial location," Dr. Wang explained. Panoramic PACT also offers detailed visualization of vasculature regardless of breast density, and can detect hemoglobin and tumor hypoxia as well.
"We basically use molecules to figure out the body's physiology," Dr. Wang said. "That's the beauty of photoacoustic tomography: By detecting molecules, we can figure out exactly how the body is functioning. When there's a functional difference, that means we can potentially detect disease better."
The PACT system was also trained on images of malignant and benign growths or lumps as well as suspicious and healthy tissue, improving its ability to notice subtle variations in the type of tissue imaged.
The researchers tested PACT on the features of 78 breasts in 39 patients to distinguish between healthy and suspicious breast tissue. The system achieved a maximum area under the receiver operating characteristic curve of 0.89, which was comparable with mammography and MRI results.
Going forward, the researchers intend to enhance the technique with additional laser wavelengths and new features, validate the system with a larger dataset, and improve the classification model.
Disclosure: The study was funded in part by grants from the National Institutes of Health. For full disclosures of the study authors, visit nature.com.
