Artificial intelligence is emerging as a transformative force in cancer diagnostics, offering pathologists and oncologists powerful new tools to improve accuracy, efficiency, and patient outcomes. While experts emphasize that AI won't replace human expertise, those who embrace the technology may gain significant advantages in cancer detection and treatment planning.
"AI will never replace pathologists [or oncologists], but it's the next new technology that pathologists will employ to help them make better, more accurate diagnoses," said Stuart J. Schnitt, MD, chief of breast oncologic pathology at Dana-Farber Brigham Cancer Center and professor of pathology at Harvard Medical School.
Addressing Key Challenges in Pathology
The field of pathology faces several pressing challenges that AI could help address. A global shortage of pathologists, increasing workloads, and growing case complexity have created significant pressure on diagnostic services. According to Schnitt, AI applications in diagnostic pathology can improve workflow, reduce turnaround times, and increase diagnostic accuracy and reproducibility.
"It can in many ways level the playing field and increase the accuracy and interobserver reproducibility of diagnosis," Schnitt explained. "Theoretically, it could bring people who have less experience up to a higher level by providing this adjunctive way of looking at cases."
AI can also handle tasks that pathologists find tedious or time-consuming. For example, screening lymph nodes for tumors—a task few pathologists enjoy—could be automated with AI algorithms that identify nodes likely to contain tumors, saving time and potentially improving accuracy.
Perhaps most intriguing is AI's ability to analyze and integrate features in ways humans cannot. "There are AI algorithms out there, and I think they'll get more sophisticated going forward, that can do things that humans can't possibly do," Schnitt noted.
Enhancing Precision Oncology
In the era of personalized medicine, AI could play a crucial role in ensuring patients receive the most appropriate treatments. By first confirming diagnostic accuracy—essential for proper treatment selection—AI can then provide prognostic and predictive information beyond what pathologists can determine visually.
Schnitt described a recent study where an AI algorithm analyzing the tissue microenvironment in breast cancer patients with residual disease after neoadjuvant chemotherapy outperformed the standard residual cancer burden (RCB) score in predicting outcomes. "The AI algorithm was better than the RCB, and the RCB didn't add anything to this algorithm," he said.
Such capabilities could significantly impact treatment decisions and patient management. AI algorithms can also predict receptor status from routine hematoxylin and eosin stained sections, potentially bringing advanced diagnostic capabilities to low-resource settings where immunohistochemical staining might be unavailable.
Implementation Challenges
Despite its promise, AI implementation in pathology faces significant hurdles. The primary challenge is the requirement for digitized slides, which remains limited in the United States compared to Europe.
"Unfortunately, in the United States, there are not many institutions that are completely digitized and have gone from glass slides to digitized slides," Schnitt explained. "To me, the major roadblock in all of this is having digitized slides. Once you have digitized slides, then these algorithms can be applied."
Digitization is expensive, requiring numerous scanners, personnel, and space—especially for large departments handling tens of thousands of cases annually. Additional concerns include ethical questions about diagnostic responsibility, patient privacy with digital slides, and regulatory issues regarding FDA approval for AI algorithms.
AI in Early Cancer Detection
AI is already making inroads in cancer screening, particularly in mammography interpretation. Studies have shown that AI can increase breast cancer detection rates, though the combination of radiologist expertise and AI produces better results than either alone.
"There have been a number of studies that have documented the fact that the use of AI can increase the detection of breast cancer," Schnitt noted. "But these studies have also shown that the combination of the radiologist and AI is better than either alone."
This collaborative approach addresses a common patient concern—that diagnoses might be made solely by machines without human oversight. Schnitt emphasized that AI will serve as an adjunct to human expertise, not a replacement.
Reducing Disparities in Cancer Care
AI could potentially reduce global disparities in cancer care by standardizing diagnoses across different experience levels and resource settings. Innovative approaches, such as using smartphone cameras with microscopes to digitize images for AI analysis, might bring advanced diagnostic capabilities to low-resource regions unable to afford expensive scanning equipment.
"From a global medical care point of view, AI can do a lot to standardize things across countries and across levels of experience," Schnitt said.
The Future of AI in Cancer Diagnostics
Looking ahead, Schnitt envisions AI fundamentally changing pathology practice. Future workflows might involve AI algorithms pre-screening cases, prioritizing those with detected cancer, and allowing pathologists to focus their expertise where it's most needed.
David Craig, PhD, professor and chair in the Department of Integrative Translational Science at City of Hope, highlights additional applications of AI in histopathology, including automated detection and quantification of tissue features, multi-omics data integration, and biomarker discovery.
Despite challenges in data requirements, generalizability, regulatory approval, and ethical considerations, the potential benefits are substantial. AI could augment pathologists' expertise, leading to more accurate diagnoses and improved patient care.
"I think it's going to fundamentally change the way we practice," Schnitt concluded. "Ultimately, everybody's going to benefit, particularly patients."
